62 files changed, 3968 insertions, 574 deletions

diff --git a/src/video_core/CMakeLists.txt b/src/video_core/CMakeLists.txt
index 6aac7f305..91a30fef7 100644
--- a/src/video_core/CMakeLists.txt
+++ b/src/video_core/CMakeLists.txt
@@ -132,6 +132,8 @@ add_library(video_core STATIC
     renderer_vulkan/vk_descriptor_pool.h
     renderer_vulkan/vk_fence_manager.cpp
     renderer_vulkan/vk_fence_manager.h
+    renderer_vulkan/vk_fsr.cpp
+    renderer_vulkan/vk_fsr.h
     renderer_vulkan/vk_graphics_pipeline.cpp
     renderer_vulkan/vk_graphics_pipeline.h
     renderer_vulkan/vk_master_semaphore.cpp
diff --git a/src/video_core/buffer_cache/buffer_cache.h b/src/video_core/buffer_cache/buffer_cache.h
index d350c9b36..43bed63ac 100644
--- a/src/video_core/buffer_cache/buffer_cache.h
+++ b/src/video_core/buffer_cache/buffer_cache.h
@@ -853,12 +853,14 @@ void BufferCache<P>::CommitAsyncFlushesHigh() {
     }
     if constexpr (USE_MEMORY_MAPS) {
         auto download_staging = runtime.DownloadStagingBuffer(total_size_bytes);
+        runtime.PreCopyBarrier();
         for (auto& [copy, buffer_id] : downloads) {
             // Have in mind the staging buffer offset for the copy
             copy.dst_offset += download_staging.offset;
             const std::array copies{copy};
-            runtime.CopyBuffer(download_staging.buffer, slot_buffers[buffer_id], copies);
+            runtime.CopyBuffer(download_staging.buffer, slot_buffers[buffer_id], copies, false);
         }
+        runtime.PostCopyBarrier();
         runtime.Finish();
         for (const auto& [copy, buffer_id] : downloads) {
             const Buffer& buffer = slot_buffers[buffer_id];
diff --git a/src/video_core/dirty_flags.h b/src/video_core/dirty_flags.h
index f0d545f90..d63ad5a35 100644
--- a/src/video_core/dirty_flags.h
+++ b/src/video_core/dirty_flags.h
@@ -29,6 +29,8 @@ enum : u8 {
     ColorBuffer6,
     ColorBuffer7,
     ZetaBuffer,
+    RescaleViewports,
+    RescaleScissors,
 
     VertexBuffers,
     VertexBuffer0,
diff --git a/src/video_core/host_shaders/CMakeLists.txt b/src/video_core/host_shaders/CMakeLists.txt
index 20d748c12..d779a967a 100644
--- a/src/video_core/host_shaders/CMakeLists.txt
+++ b/src/video_core/host_shaders/CMakeLists.txt
@@ -1,3 +1,11 @@
+set(FIDELITYFX_INCLUDE_DIR ${CMAKE_SOURCE_DIR}/externals/FidelityFX-FSR/ffx-fsr)
+
+set(GLSL_INCLUDES
+    fidelityfx_fsr.comp
+    ${FIDELITYFX_INCLUDE_DIR}/ffx_a.h
+    ${FIDELITYFX_INCLUDE_DIR}/ffx_fsr1.h
+)
+
 set(SHADER_FILES
     astc_decoder.comp
     block_linear_unswizzle_2d.comp
@@ -5,14 +13,25 @@ set(SHADER_FILES
     convert_depth_to_float.frag
     convert_float_to_depth.frag
     full_screen_triangle.vert
+    fxaa.frag
+    fxaa.vert
     opengl_copy_bc4.comp
     opengl_present.frag
     opengl_present.vert
+    opengl_present_scaleforce.frag
     pitch_unswizzle.comp
+    present_bicubic.frag
+    present_gaussian.frag
     vulkan_blit_color_float.frag
     vulkan_blit_depth_stencil.frag
+    vulkan_fidelityfx_fsr_easu_fp16.comp
+    vulkan_fidelityfx_fsr_easu_fp32.comp
+    vulkan_fidelityfx_fsr_rcas_fp16.comp
+    vulkan_fidelityfx_fsr_rcas_fp32.comp
     vulkan_present.frag
     vulkan_present.vert
+    vulkan_present_scaleforce_fp16.frag
+    vulkan_present_scaleforce_fp32.frag
     vulkan_quad_indexed.comp
     vulkan_uint8.comp
 )
@@ -76,7 +95,7 @@ foreach(FILENAME IN ITEMS ${SHADER_FILES})
             OUTPUT
                 ${SPIRV_HEADER_FILE}
             COMMAND
-                ${GLSLANGVALIDATOR} -V ${QUIET_FLAG} ${GLSL_FLAGS} --variable-name ${SPIRV_VARIABLE_NAME} -o ${SPIRV_HEADER_FILE} ${SOURCE_FILE}
+                ${GLSLANGVALIDATOR} -V ${QUIET_FLAG} -I"${FIDELITYFX_INCLUDE_DIR}" ${GLSL_FLAGS} --variable-name ${SPIRV_VARIABLE_NAME} -o ${SPIRV_HEADER_FILE} ${SOURCE_FILE}
             MAIN_DEPENDENCY
                 ${SOURCE_FILE}
         )
@@ -84,9 +103,12 @@ foreach(FILENAME IN ITEMS ${SHADER_FILES})
     endif()
 endforeach()
 
+set(SHADER_SOURCES ${SHADER_FILES})
+list(APPEND SHADER_SOURCES ${GLSL_INCLUDES})
+
 add_custom_target(host_shaders
     DEPENDS
         ${SHADER_HEADERS}
     SOURCES
-        ${SHADER_FILES}
+        ${SHADER_SOURCES}
 )
diff --git a/src/video_core/host_shaders/fidelityfx_fsr.comp b/src/video_core/host_shaders/fidelityfx_fsr.comp
new file mode 100644
index 000000000..6b97f789d
--- /dev/null
+++ b/src/video_core/host_shaders/fidelityfx_fsr.comp
@@ -0,0 +1,116 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+//!#version 460 core
+#extension GL_ARB_separate_shader_objects : enable
+#extension GL_ARB_shading_language_420pack : enable
+#extension GL_GOOGLE_include_directive : enable
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+// FidelityFX Super Resolution Sample
+//
+// Copyright (c) 2021 Advanced Micro Devices, Inc. All rights reserved.
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files(the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and / or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions :
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+layout( push_constant ) uniform constants {
+    uvec4 Const0;
+    uvec4 Const1;
+    uvec4 Const2;
+    uvec4 Const3;
+};
+
+layout(set=0,binding=0) uniform sampler2D InputTexture;
+layout(set=0,binding=1,rgba16f) uniform image2D OutputTexture;
+
+#define A_GPU 1
+#define A_GLSL 1
+
+#ifndef YUZU_USE_FP16
+    #include "ffx_a.h"
+
+    #if USE_EASU
+        #define FSR_EASU_F 1
+        AF4 FsrEasuRF(AF2 p) { AF4 res = textureGather(InputTexture, p, 0); return res; }
+        AF4 FsrEasuGF(AF2 p) { AF4 res = textureGather(InputTexture, p, 1); return res; }
+        AF4 FsrEasuBF(AF2 p) { AF4 res = textureGather(InputTexture, p, 2); return res; }
+    #endif
+    #if USE_RCAS
+        #define FSR_RCAS_F 1
+        AF4 FsrRcasLoadF(ASU2 p) { return texelFetch(InputTexture, ASU2(p), 0); }
+        void FsrRcasInputF(inout AF1 r, inout AF1 g, inout AF1 b) {}
+    #endif
+#else
+    #define A_HALF
+    #include "ffx_a.h"
+
+    #if USE_EASU
+        #define FSR_EASU_H 1
+        AH4 FsrEasuRH(AF2 p) { AH4 res = AH4(textureGather(InputTexture, p, 0)); return res; }
+        AH4 FsrEasuGH(AF2 p) { AH4 res = AH4(textureGather(InputTexture, p, 1)); return res; }
+        AH4 FsrEasuBH(AF2 p) { AH4 res = AH4(textureGather(InputTexture, p, 2)); return res; }
+    #endif
+    #if USE_RCAS
+        #define FSR_RCAS_H 1
+        AH4 FsrRcasLoadH(ASW2 p) { return AH4(texelFetch(InputTexture, ASU2(p), 0)); }
+        void FsrRcasInputH(inout AH1 r,inout AH1 g,inout AH1 b){}
+    #endif
+#endif
+
+#include "ffx_fsr1.h"
+
+void CurrFilter(AU2 pos) {
+#if USE_BILINEAR
+    AF2 pp = (AF2(pos) * AF2_AU2(Const0.xy) + AF2_AU2(Const0.zw)) * AF2_AU2(Const1.xy) + AF2(0.5, -0.5) * AF2_AU2(Const1.zw);
+    imageStore(OutputTexture, ASU2(pos), textureLod(InputTexture, pp, 0.0));
+#endif
+#if USE_EASU
+    #ifndef YUZU_USE_FP16
+        AF3 c;
+        FsrEasuF(c, pos, Const0, Const1, Const2, Const3);
+        imageStore(OutputTexture, ASU2(pos), AF4(c, 1));
+    #else
+        AH3 c;
+        FsrEasuH(c, pos, Const0, Const1, Const2, Const3);
+        imageStore(OutputTexture, ASU2(pos), AH4(c, 1));
+    #endif
+#endif
+#if USE_RCAS
+    #ifndef YUZU_USE_FP16
+        AF3 c;
+        FsrRcasF(c.r, c.g, c.b, pos, Const0);
+        imageStore(OutputTexture, ASU2(pos), AF4(c, 1));
+    #else
+        AH3 c;
+        FsrRcasH(c.r, c.g, c.b, pos, Const0);
+        imageStore(OutputTexture, ASU2(pos), AH4(c, 1));
+    #endif
+#endif
+}
+
+layout(local_size_x=64) in;
+void main() {
+    // Do remapping of local xy in workgroup for a more PS-like swizzle pattern.
+    AU2 gxy = ARmp8x8(gl_LocalInvocationID.x) + AU2(gl_WorkGroupID.x << 4u, gl_WorkGroupID.y << 4u);
+    CurrFilter(gxy);
+    gxy.x += 8u;
+    CurrFilter(gxy);
+    gxy.y += 8u;
+    CurrFilter(gxy);
+    gxy.x -= 8u;
+    CurrFilter(gxy);
+}
diff --git a/src/video_core/host_shaders/fxaa.frag b/src/video_core/host_shaders/fxaa.frag
new file mode 100644
index 000000000..02f4068d1
--- /dev/null
+++ b/src/video_core/host_shaders/fxaa.frag
@@ -0,0 +1,76 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+// Source code is adapted from
+// https://www.geeks3d.com/20110405/fxaa-fast-approximate-anti-aliasing-demo-glsl-opengl-test-radeon-geforce/3/
+
+#version 460
+
+#ifdef VULKAN
+
+#define BINDING_COLOR_TEXTURE 1
+
+#else // ^^^ Vulkan ^^^ // vvv OpenGL vvv
+
+#define BINDING_COLOR_TEXTURE 0
+
+#endif
+
+layout (location = 0) in vec4 posPos;
+
+layout (location = 0) out vec4 frag_color;
+
+layout (binding = BINDING_COLOR_TEXTURE) uniform sampler2D input_texture;
+
+const float FXAA_SPAN_MAX = 8.0;
+const float FXAA_REDUCE_MUL = 1.0 / 8.0;
+const float FXAA_REDUCE_MIN = 1.0 / 128.0;
+
+#define FxaaTexLod0(t, p) textureLod(t, p, 0.0)
+#define FxaaTexOff(t, p, o) textureLodOffset(t, p, 0.0, o)
+
+vec3 FxaaPixelShader(vec4 posPos, sampler2D tex) {
+
+    vec3 rgbNW = FxaaTexLod0(tex, posPos.zw).xyz;
+    vec3 rgbNE = FxaaTexOff(tex, posPos.zw, ivec2(1,0)).xyz;
+    vec3 rgbSW = FxaaTexOff(tex, posPos.zw, ivec2(0,1)).xyz;
+    vec3 rgbSE = FxaaTexOff(tex, posPos.zw, ivec2(1,1)).xyz;
+    vec3 rgbM  = FxaaTexLod0(tex, posPos.xy).xyz;
+/*---------------------------------------------------------*/
+    vec3 luma = vec3(0.299, 0.587, 0.114);
+    float lumaNW = dot(rgbNW, luma);
+    float lumaNE = dot(rgbNE, luma);
+    float lumaSW = dot(rgbSW, luma);
+    float lumaSE = dot(rgbSE, luma);
+    float lumaM  = dot(rgbM,  luma);
+/*---------------------------------------------------------*/
+    float lumaMin = min(lumaM, min(min(lumaNW, lumaNE), min(lumaSW, lumaSE)));
+    float lumaMax = max(lumaM, max(max(lumaNW, lumaNE), max(lumaSW, lumaSE)));
+/*---------------------------------------------------------*/
+    vec2 dir;
+    dir.x = -((lumaNW + lumaNE) - (lumaSW + lumaSE));
+    dir.y =  ((lumaNW + lumaSW) - (lumaNE + lumaSE));
+/*---------------------------------------------------------*/
+    float dirReduce = max(
+        (lumaNW + lumaNE + lumaSW + lumaSE) * (0.25 * FXAA_REDUCE_MUL),
+        FXAA_REDUCE_MIN);
+    float rcpDirMin = 1.0/(min(abs(dir.x), abs(dir.y)) + dirReduce);
+    dir = min(vec2( FXAA_SPAN_MAX,  FXAA_SPAN_MAX),
+          max(vec2(-FXAA_SPAN_MAX, -FXAA_SPAN_MAX),
+          dir * rcpDirMin)) / textureSize(tex, 0);
+/*--------------------------------------------------------*/
+    vec3 rgbA = (1.0 / 2.0) * (
+        FxaaTexLod0(tex, posPos.xy + dir * (1.0 / 3.0 - 0.5)).xyz +
+        FxaaTexLod0(tex, posPos.xy + dir * (2.0 / 3.0 - 0.5)).xyz);
+    vec3 rgbB = rgbA * (1.0 / 2.0) + (1.0 / 4.0) * (
+        FxaaTexLod0(tex, posPos.xy + dir * (0.0 / 3.0 - 0.5)).xyz +
+        FxaaTexLod0(tex, posPos.xy + dir * (3.0 / 3.0 - 0.5)).xyz);
+    float lumaB = dot(rgbB, luma);
+    if((lumaB < lumaMin) || (lumaB > lumaMax)) return rgbA;
+    return rgbB;
+}
+
+void main() {
+  frag_color = vec4(FxaaPixelShader(posPos, input_texture), 1.0);
+}
diff --git a/src/video_core/host_shaders/fxaa.vert b/src/video_core/host_shaders/fxaa.vert
new file mode 100644
index 000000000..ac20c04e9
--- /dev/null
+++ b/src/video_core/host_shaders/fxaa.vert
@@ -0,0 +1,38 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#version 460
+
+out gl_PerVertex {
+    vec4 gl_Position;
+};
+
+const vec2 vertices[4] =
+    vec2[4](vec2(-1.0, 1.0), vec2(1.0, 1.0), vec2(-1.0, -1.0), vec2(1.0, -1.0));
+
+layout (location = 0) out vec4 posPos;
+
+#ifdef VULKAN
+
+#define BINDING_COLOR_TEXTURE 0
+#define VERTEX_ID gl_VertexIndex
+
+#else // ^^^ Vulkan ^^^ // vvv OpenGL vvv
+
+#define BINDING_COLOR_TEXTURE 0
+#define VERTEX_ID gl_VertexID
+
+#endif
+
+layout (binding = BINDING_COLOR_TEXTURE) uniform sampler2D input_texture;
+
+const float FXAA_SUBPIX_SHIFT = 0;
+
+void main() {
+  vec2 vertex = vertices[VERTEX_ID];
+  gl_Position = vec4(vertex, 0.0, 1.0);
+  vec2 vert_tex_coord = (vertex + 1.0) / 2.0;
+  posPos.xy = vert_tex_coord;
+  posPos.zw = vert_tex_coord - (0.5 + FXAA_SUBPIX_SHIFT) / textureSize(input_texture, 0);
+}
diff --git a/src/video_core/host_shaders/opengl_present_scaleforce.frag b/src/video_core/host_shaders/opengl_present_scaleforce.frag
new file mode 100644
index 000000000..71ff9e1e3
--- /dev/null
+++ b/src/video_core/host_shaders/opengl_present_scaleforce.frag
@@ -0,0 +1,130 @@
+// MIT License
+//
+// Copyright (c) 2020 BreadFish64
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in all
+// copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+// SOFTWARE.
+
+// Adapted from https://github.com/BreadFish64/ScaleFish/tree/master/scaleforce
+
+//! #version 460
+
+#extension GL_ARB_separate_shader_objects : enable
+
+#ifdef YUZU_USE_FP16
+
+#extension GL_AMD_gpu_shader_half_float : enable
+#extension GL_NV_gpu_shader5 : enable
+
+#define lfloat float16_t
+#define lvec2 f16vec2
+#define lvec3 f16vec3
+#define lvec4 f16vec4
+
+#else
+
+#define lfloat float
+#define lvec2 vec2
+#define lvec3 vec3
+#define lvec4 vec4
+
+#endif
+
+#ifdef VULKAN
+
+#define BINDING_COLOR_TEXTURE 1
+
+#else // ^^^ Vulkan ^^^ // vvv OpenGL vvv
+
+#define BINDING_COLOR_TEXTURE 0
+
+#endif
+
+layout (location = 0) in vec2 tex_coord;
+
+layout (location = 0) out vec4 frag_color;
+
+layout (binding = BINDING_COLOR_TEXTURE) uniform sampler2D input_texture;
+
+const bool ignore_alpha = true;
+
+lfloat ColorDist1(lvec4 a, lvec4 b) {
+    // https://en.wikipedia.org/wiki/YCbCr#ITU-R_BT.2020_conversion
+    const lvec3 K = lvec3(0.2627, 0.6780, 0.0593);
+    const lfloat scaleB = lfloat(0.5) / (lfloat(1.0) - K.b);
+    const lfloat scaleR = lfloat(0.5) / (lfloat(1.0) - K.r);
+    lvec4 diff = a - b;
+    lfloat Y = dot(diff.rgb, K);
+    lfloat Cb = scaleB * (diff.b - Y);
+    lfloat Cr = scaleR * (diff.r - Y);
+    lvec3 YCbCr = lvec3(Y, Cb, Cr);
+    lfloat d = length(YCbCr);
+    if (ignore_alpha) {
+        return d;
+    }
+    return sqrt(a.a * b.a * d * d + diff.a * diff.a);
+}
+
+lvec4 ColorDist(lvec4 ref, lvec4 A, lvec4 B, lvec4 C, lvec4 D) {
+    return lvec4(
+            ColorDist1(ref, A),
+            ColorDist1(ref, B),
+            ColorDist1(ref, C),
+            ColorDist1(ref, D)
+        );
+}
+
+vec4 Scaleforce(sampler2D tex, vec2 tex_coord) {
+    lvec4 bl = lvec4(textureOffset(tex, tex_coord, ivec2(-1, -1)));
+    lvec4 bc = lvec4(textureOffset(tex, tex_coord, ivec2(0, -1)));
+    lvec4 br = lvec4(textureOffset(tex, tex_coord, ivec2(1, -1)));
+    lvec4 cl = lvec4(textureOffset(tex, tex_coord, ivec2(-1, 0)));
+    lvec4 cc = lvec4(texture(tex, tex_coord));
+    lvec4 cr = lvec4(textureOffset(tex, tex_coord, ivec2(1, 0)));
+    lvec4 tl = lvec4(textureOffset(tex, tex_coord, ivec2(-1, 1)));
+    lvec4 tc = lvec4(textureOffset(tex, tex_coord, ivec2(0, 1)));
+    lvec4 tr = lvec4(textureOffset(tex, tex_coord, ivec2(1, 1)));
+
+    lvec4 offset_tl = ColorDist(cc, tl, tc, tr, cr);
+    lvec4 offset_br = ColorDist(cc, br, bc, bl, cl);
+
+    // Calculate how different cc is from the texels around it
+    const lfloat plus_weight = lfloat(1.5);
+    const lfloat cross_weight = lfloat(1.5);
+    lfloat total_dist = dot(offset_tl + offset_br, lvec4(cross_weight, plus_weight, cross_weight, plus_weight));
+
+    if (total_dist == lfloat(0.0)) {
+        return cc;
+    } else {
+        // Add together all the distances with direction taken into account
+        lvec4 tmp = offset_tl - offset_br;
+        lvec2 total_offset = tmp.wy * plus_weight + (tmp.zz + lvec2(-tmp.x, tmp.x)) * cross_weight;
+
+        // When the image has thin points, they tend to split apart.
+        // This is because the texels all around are different and total_offset reaches into clear areas.
+        // This works pretty well to keep the offset in bounds for these cases.
+        lfloat clamp_val = length(total_offset) / total_dist;
+        vec2 final_offset = vec2(clamp(total_offset, -clamp_val, clamp_val)) / textureSize(tex, 0);
+
+        return texture(tex, tex_coord - final_offset);
+    }
+}
+
+void main() {
+    frag_color = Scaleforce(input_texture, tex_coord);
+}
diff --git a/src/video_core/host_shaders/present_bicubic.frag b/src/video_core/host_shaders/present_bicubic.frag
new file mode 100644
index 000000000..902b70c2b
--- /dev/null
+++ b/src/video_core/host_shaders/present_bicubic.frag
@@ -0,0 +1,67 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#version 460 core
+
+#ifdef VULKAN
+
+#define BINDING_COLOR_TEXTURE 1
+
+#else // ^^^ Vulkan ^^^ // vvv OpenGL vvv
+
+#define BINDING_COLOR_TEXTURE 0
+
+#endif
+
+
+layout (location = 0) in vec2 frag_tex_coord;
+
+layout (location = 0) out vec4 color;
+
+layout (binding = BINDING_COLOR_TEXTURE) uniform sampler2D color_texture;
+
+vec4 cubic(float v) {
+    vec4 n = vec4(1.0, 2.0, 3.0, 4.0) - v;
+    vec4 s = n * n * n;
+    float x = s.x;
+    float y = s.y - 4.0 * s.x;
+    float z = s.z - 4.0 * s.y + 6.0 * s.x;
+    float w = 6.0 - x - y - z;
+    return vec4(x, y, z, w) * (1.0 / 6.0);
+}
+
+vec4 textureBicubic( sampler2D textureSampler, vec2 texCoords ) {
+
+    vec2 texSize = textureSize(textureSampler, 0);
+    vec2 invTexSize = 1.0 / texSize;
+
+    texCoords = texCoords * texSize - 0.5;
+
+    vec2 fxy = fract(texCoords);
+    texCoords -= fxy;
+
+    vec4 xcubic = cubic(fxy.x);
+    vec4 ycubic = cubic(fxy.y);
+
+    vec4 c = texCoords.xxyy + vec2(-0.5, +1.5).xyxy;
+
+    vec4 s = vec4(xcubic.xz + xcubic.yw, ycubic.xz + ycubic.yw);
+    vec4 offset = c + vec4(xcubic.yw, ycubic.yw) / s;
+
+    offset *= invTexSize.xxyy;
+
+    vec4 sample0 = texture(textureSampler, offset.xz);
+    vec4 sample1 = texture(textureSampler, offset.yz);
+    vec4 sample2 = texture(textureSampler, offset.xw);
+    vec4 sample3 = texture(textureSampler, offset.yw);
+
+    float sx = s.x / (s.x + s.y);
+    float sy = s.z / (s.z + s.w);
+
+    return mix(mix(sample3, sample2, sx), mix(sample1, sample0, sx), sy);
+}
+
+void main() {
+    color = vec4(textureBicubic(color_texture, frag_tex_coord).rgb, 1.0f);
+}
diff --git a/src/video_core/host_shaders/present_gaussian.frag b/src/video_core/host_shaders/present_gaussian.frag
new file mode 100644
index 000000000..66fed3238
--- /dev/null
+++ b/src/video_core/host_shaders/present_gaussian.frag
@@ -0,0 +1,70 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+// Code adapted from the following sources:
+// - https://learnopengl.com/Advanced-Lighting/Bloom
+// - https://www.rastergrid.com/blog/2010/09/efficient-gaussian-blur-with-linear-sampling/
+
+#version 460 core
+
+#ifdef VULKAN
+
+#define BINDING_COLOR_TEXTURE 1
+
+#else // ^^^ Vulkan ^^^ // vvv OpenGL vvv
+
+#define BINDING_COLOR_TEXTURE 0
+
+#endif
+
+layout(location = 0) in vec2 frag_tex_coord;
+
+layout(location = 0) out vec4 color;
+
+layout(binding = BINDING_COLOR_TEXTURE) uniform sampler2D color_texture;
+
+const float offset[3] = float[](0.0, 1.3846153846, 3.2307692308);
+const float weight[3] = float[](0.2270270270, 0.3162162162, 0.0702702703);
+
+vec4 blurVertical(sampler2D textureSampler, vec2 coord, vec2 norm) {
+    vec4 result = vec4(0.0f);
+    for (int i = 1; i < 3; i++) {
+        result += texture(textureSampler, vec2(coord) + (vec2(0.0, offset[i]) * norm)) * weight[i];
+        result += texture(textureSampler, vec2(coord) - (vec2(0.0, offset[i]) * norm)) * weight[i];
+    }
+    return result;
+}
+
+vec4 blurHorizontal(sampler2D textureSampler, vec2 coord, vec2 norm) {
+    vec4 result = vec4(0.0f);
+    for (int i = 1; i < 3; i++) {
+        result += texture(textureSampler, vec2(coord) + (vec2(offset[i], 0.0) * norm)) * weight[i];
+        result += texture(textureSampler, vec2(coord) - (vec2(offset[i], 0.0) * norm)) * weight[i];
+    }
+    return result;
+}
+
+vec4 blurDiagonal(sampler2D textureSampler, vec2 coord, vec2 norm) {
+    vec4 result = vec4(0.0f);
+    for (int i = 1; i < 3; i++) {
+        result +=
+            texture(textureSampler, vec2(coord) + (vec2(offset[i], offset[i]) * norm)) * weight[i];
+        result +=
+            texture(textureSampler, vec2(coord) - (vec2(offset[i], offset[i]) * norm)) * weight[i];
+    }
+    return result;
+}
+
+void main() {
+    vec3 base = texture(color_texture, vec2(frag_tex_coord)).rgb * weight[0];
+    vec2 tex_offset = 1.0f / textureSize(color_texture, 0);
+
+    // TODO(Blinkhawk): This code can be optimized through shader group instructions.
+    vec3 horizontal = blurHorizontal(color_texture, frag_tex_coord, tex_offset).rgb;
+    vec3 vertical = blurVertical(color_texture, frag_tex_coord, tex_offset).rgb;
+    vec3 diagonalA = blurDiagonal(color_texture, frag_tex_coord, tex_offset).rgb;
+    vec3 diagonalB = blurDiagonal(color_texture, frag_tex_coord, tex_offset * vec2(1.0, -1.0)).rgb;
+    vec3 combination = mix(mix(horizontal, vertical, 0.5f), mix(diagonalA, diagonalB, 0.5f), 0.5f);
+    color = vec4(combination + base, 1.0f);
+}
diff --git a/src/video_core/host_shaders/vulkan_fidelityfx_fsr_easu_fp16.comp b/src/video_core/host_shaders/vulkan_fidelityfx_fsr_easu_fp16.comp
new file mode 100644
index 000000000..1c96a7905
--- /dev/null
+++ b/src/video_core/host_shaders/vulkan_fidelityfx_fsr_easu_fp16.comp
@@ -0,0 +1,11 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#version 460 core
+#extension GL_GOOGLE_include_directive : enable
+
+#define YUZU_USE_FP16
+#define USE_EASU 1
+
+#include "fidelityfx_fsr.comp"
diff --git a/src/video_core/host_shaders/vulkan_fidelityfx_fsr_easu_fp32.comp b/src/video_core/host_shaders/vulkan_fidelityfx_fsr_easu_fp32.comp
new file mode 100644
index 000000000..f4daff739
--- /dev/null
+++ b/src/video_core/host_shaders/vulkan_fidelityfx_fsr_easu_fp32.comp
@@ -0,0 +1,10 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#version 460 core
+#extension GL_GOOGLE_include_directive : enable
+
+#define USE_EASU 1
+
+#include "fidelityfx_fsr.comp"
diff --git a/src/video_core/host_shaders/vulkan_fidelityfx_fsr_rcas_fp16.comp b/src/video_core/host_shaders/vulkan_fidelityfx_fsr_rcas_fp16.comp
new file mode 100644
index 000000000..6b6796dd1
--- /dev/null
+++ b/src/video_core/host_shaders/vulkan_fidelityfx_fsr_rcas_fp16.comp
@@ -0,0 +1,11 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#version 460 core
+#extension GL_GOOGLE_include_directive : enable
+
+#define YUZU_USE_FP16
+#define USE_RCAS 1
+
+#include "fidelityfx_fsr.comp"
diff --git a/src/video_core/host_shaders/vulkan_fidelityfx_fsr_rcas_fp32.comp b/src/video_core/host_shaders/vulkan_fidelityfx_fsr_rcas_fp32.comp
new file mode 100644
index 000000000..f785eebf3
--- /dev/null
+++ b/src/video_core/host_shaders/vulkan_fidelityfx_fsr_rcas_fp32.comp
@@ -0,0 +1,10 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#version 460 core
+#extension GL_GOOGLE_include_directive : enable
+
+#define USE_RCAS 1
+
+#include "fidelityfx_fsr.comp"
diff --git a/src/video_core/host_shaders/vulkan_present_scaleforce_fp16.frag b/src/video_core/host_shaders/vulkan_present_scaleforce_fp16.frag
new file mode 100644
index 000000000..924c03060
--- /dev/null
+++ b/src/video_core/host_shaders/vulkan_present_scaleforce_fp16.frag
@@ -0,0 +1,7 @@
+#version 460
+
+#extension GL_GOOGLE_include_directive : enable
+
+#define YUZU_USE_FP16
+
+#include "opengl_present_scaleforce.frag"
diff --git a/src/video_core/host_shaders/vulkan_present_scaleforce_fp32.frag b/src/video_core/host_shaders/vulkan_present_scaleforce_fp32.frag
new file mode 100644
index 000000000..a594b83ca
--- /dev/null
+++ b/src/video_core/host_shaders/vulkan_present_scaleforce_fp32.frag
@@ -0,0 +1,5 @@
+#version 460
+
+#extension GL_GOOGLE_include_directive : enable
+
+#include "opengl_present_scaleforce.frag"
diff --git a/src/video_core/renderer_opengl/gl_buffer_cache.cpp b/src/video_core/renderer_opengl/gl_buffer_cache.cpp
index 187a28e4d..d4dd10bb6 100644
--- a/src/video_core/renderer_opengl/gl_buffer_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_buffer_cache.cpp
@@ -5,6 +5,7 @@
 #include <algorithm>
 #include <span>
 
+#include "shader_recompiler/backend/glasm/emit_glasm.h"
 #include "video_core/buffer_cache/buffer_cache.h"
 #include "video_core/renderer_opengl/gl_buffer_cache.h"
 #include "video_core/renderer_opengl/gl_device.h"
@@ -229,8 +230,10 @@ void BufferCacheRuntime::BindStorageBuffer(size_t stage, u32 binding_index, Buff
             .padding = 0,
         };
         buffer.MakeResident(is_written ? GL_READ_WRITE : GL_READ_ONLY);
-        glProgramLocalParametersI4uivNV(PROGRAM_LUT[stage], binding_index, 1,
-                                        reinterpret_cast<const GLuint*>(&ssbo));
+        glProgramLocalParametersI4uivNV(
+            PROGRAM_LUT[stage],
+            Shader::Backend::GLASM::PROGRAM_LOCAL_PARAMETER_STORAGE_BUFFER_BASE + binding_index, 1,
+            reinterpret_cast<const GLuint*>(&ssbo));
     }
 }
 
@@ -250,8 +253,10 @@ void BufferCacheRuntime::BindComputeStorageBuffer(u32 binding_index, Buffer& buf
             .padding = 0,
         };
         buffer.MakeResident(is_written ? GL_READ_WRITE : GL_READ_ONLY);
-        glProgramLocalParametersI4uivNV(GL_COMPUTE_PROGRAM_NV, binding_index, 1,
-                                        reinterpret_cast<const GLuint*>(&ssbo));
+        glProgramLocalParametersI4uivNV(
+            GL_COMPUTE_PROGRAM_NV,
+            Shader::Backend::GLASM::PROGRAM_LOCAL_PARAMETER_STORAGE_BUFFER_BASE + binding_index, 1,
+            reinterpret_cast<const GLuint*>(&ssbo));
     }
 }
 
diff --git a/src/video_core/renderer_opengl/gl_compute_pipeline.cpp b/src/video_core/renderer_opengl/gl_compute_pipeline.cpp
index aa1cc592f..5c1f21c65 100644
--- a/src/video_core/renderer_opengl/gl_compute_pipeline.cpp
+++ b/src/video_core/renderer_opengl/gl_compute_pipeline.cpp
@@ -19,15 +19,6 @@ using VideoCommon::ImageId;
 constexpr u32 MAX_TEXTURES = 64;
 constexpr u32 MAX_IMAGES = 16;
 
-template <typename Range>
-u32 AccumulateCount(const Range& range) {
-    u32 num{};
-    for (const auto& desc : range) {
-        num += desc.count;
-    }
-    return num;
-}
-
 size_t ComputePipelineKey::Hash() const noexcept {
     return static_cast<size_t>(
         Common::CityHash64(reinterpret_cast<const char*>(this), sizeof *this));
@@ -58,17 +49,17 @@ ComputePipeline::ComputePipeline(const Device& device, TextureCache& texture_cac
     std::copy_n(info.constant_buffer_used_sizes.begin(), uniform_buffer_sizes.size(),
                 uniform_buffer_sizes.begin());
 
-    num_texture_buffers = AccumulateCount(info.texture_buffer_descriptors);
-    num_image_buffers = AccumulateCount(info.image_buffer_descriptors);
+    num_texture_buffers = Shader::NumDescriptors(info.texture_buffer_descriptors);
+    num_image_buffers = Shader::NumDescriptors(info.image_buffer_descriptors);
 
-    const u32 num_textures{num_texture_buffers + AccumulateCount(info.texture_descriptors)};
+    const u32 num_textures{num_texture_buffers + Shader::NumDescriptors(info.texture_descriptors)};
     ASSERT(num_textures <= MAX_TEXTURES);
 
-    const u32 num_images{num_image_buffers + AccumulateCount(info.image_descriptors)};
+    const u32 num_images{num_image_buffers + Shader::NumDescriptors(info.image_descriptors)};
     ASSERT(num_images <= MAX_IMAGES);
 
     const bool is_glasm{assembly_program.handle != 0};
-    const u32 num_storage_buffers{AccumulateCount(info.storage_buffers_descriptors)};
+    const u32 num_storage_buffers{Shader::NumDescriptors(info.storage_buffers_descriptors)};
     use_storage_buffers =
         !is_glasm || num_storage_buffers < device.GetMaxGLASMStorageBufferBlocks();
     writes_global_memory = !use_storage_buffers &&
@@ -88,8 +79,7 @@ void ComputePipeline::Configure() {
     }
     texture_cache.SynchronizeComputeDescriptors();
 
-    std::array<ImageViewId, MAX_TEXTURES + MAX_IMAGES> image_view_ids;
-    boost::container::static_vector<u32, MAX_TEXTURES + MAX_IMAGES> image_view_indices;
+    boost::container::static_vector<VideoCommon::ImageViewInOut, MAX_TEXTURES + MAX_IMAGES> views;
     std::array<GLuint, MAX_TEXTURES> samplers;
     std::array<GLuint, MAX_TEXTURES> textures;
     std::array<GLuint, MAX_IMAGES> images;
@@ -119,33 +109,39 @@ void ComputePipeline::Configure() {
         }
         return TexturePair(gpu_memory.Read<u32>(addr), via_header_index);
     }};
-    const auto add_image{[&](const auto& desc) {
+    const auto add_image{[&](const auto& desc, bool blacklist) {
         for (u32 index = 0; index < desc.count; ++index) {
             const auto handle{read_handle(desc, index)};
-            image_view_indices.push_back(handle.first);
+            views.push_back({
+                .index = handle.first,
+                .blacklist = blacklist,
+                .id = {},
+            });
         }
     }};
     for (const auto& desc : info.texture_buffer_descriptors) {
         for (u32 index = 0; index < desc.count; ++index) {
             const auto handle{read_handle(desc, index)};
-            image_view_indices.push_back(handle.first);
+            views.push_back({handle.first});
             samplers[sampler_binding++] = 0;
         }
     }
-    std::ranges::for_each(info.image_buffer_descriptors, add_image);
+    for (const auto& desc : info.image_buffer_descriptors) {
+        add_image(desc, false);
+    }
     for (const auto& desc : info.texture_descriptors) {
         for (u32 index = 0; index < desc.count; ++index) {
             const auto handle{read_handle(desc, index)};
-            image_view_indices.push_back(handle.first);
+            views.push_back({handle.first});
 
             Sampler* const sampler = texture_cache.GetComputeSampler(handle.second);
             samplers[sampler_binding++] = sampler->Handle();
         }
     }
-    std::ranges::for_each(info.image_descriptors, add_image);
-
-    const std::span indices_span(image_view_indices.data(), image_view_indices.size());
-    texture_cache.FillComputeImageViews(indices_span, image_view_ids);
+    for (const auto& desc : info.image_descriptors) {
+        add_image(desc, desc.is_written);
+    }
+    texture_cache.FillComputeImageViews(std::span(views.data(), views.size()));
 
     if (assembly_program.handle != 0) {
         program_manager.BindComputeAssemblyProgram(assembly_program.handle);
@@ -161,7 +157,7 @@ void ComputePipeline::Configure() {
             if constexpr (is_image) {
                 is_written = desc.is_written;
             }
-            ImageView& image_view{texture_cache.GetImageView(image_view_ids[texbuf_index])};
+            ImageView& image_view{texture_cache.GetImageView(views[texbuf_index].id)};
             buffer_cache.BindComputeTextureBuffer(texbuf_index, image_view.GpuAddr(),
                                                   image_view.BufferSize(), image_view.format,
                                                   is_written, is_image);
@@ -177,23 +173,45 @@ void ComputePipeline::Configure() {
     buffer_cache.runtime.SetImagePointers(textures.data(), images.data());
     buffer_cache.BindHostComputeBuffers();
 
-    const ImageId* views_it{image_view_ids.data() + num_texture_buffers + num_image_buffers};
+    const VideoCommon::ImageViewInOut* views_it{views.data() + num_texture_buffers +
+                                                num_image_buffers};
     texture_binding += num_texture_buffers;
     image_binding += num_image_buffers;
 
+    u32 texture_scaling_mask{};
     for (const auto& desc : info.texture_descriptors) {
         for (u32 index = 0; index < desc.count; ++index) {
-            ImageView& image_view{texture_cache.GetImageView(*(views_it++))};
-            textures[texture_binding++] = image_view.Handle(desc.type);
+            ImageView& image_view{texture_cache.GetImageView((views_it++)->id)};
+            textures[texture_binding] = image_view.Handle(desc.type);
+            if (texture_cache.IsRescaling(image_view)) {
+                texture_scaling_mask |= 1u << texture_binding;
+            }
+            ++texture_binding;
         }
     }
+    u32 image_scaling_mask{};
     for (const auto& desc : info.image_descriptors) {
         for (u32 index = 0; index < desc.count; ++index) {
-            ImageView& image_view{texture_cache.GetImageView(*(views_it++))};
+            ImageView& image_view{texture_cache.GetImageView((views_it++)->id)};
             if (desc.is_written) {
                 texture_cache.MarkModification(image_view.image_id);
             }
-            images[image_binding++] = image_view.StorageView(desc.type, desc.format);
+            images[image_binding] = image_view.StorageView(desc.type, desc.format);
+            if (texture_cache.IsRescaling(image_view)) {
+                image_scaling_mask |= 1u << image_binding;
+            }
+            ++image_binding;
+        }
+    }
+    if (info.uses_rescaling_uniform) {
+        const f32 float_texture_scaling_mask{Common::BitCast<f32>(texture_scaling_mask)};
+        const f32 float_image_scaling_mask{Common::BitCast<f32>(image_scaling_mask)};
+        if (assembly_program.handle != 0) {
+            glProgramLocalParameter4fARB(GL_COMPUTE_PROGRAM_NV, 0, float_texture_scaling_mask,
+                                         float_image_scaling_mask, 0.0f, 0.0f);
+        } else {
+            glProgramUniform4f(source_program.handle, 0, float_texture_scaling_mask,
+                               float_image_scaling_mask, 0.0f, 0.0f);
         }
     }
     if (texture_binding != 0) {
diff --git a/src/video_core/renderer_opengl/gl_graphics_pipeline.cpp b/src/video_core/renderer_opengl/gl_graphics_pipeline.cpp
index bccb37a58..f8495896c 100644
--- a/src/video_core/renderer_opengl/gl_graphics_pipeline.cpp
+++ b/src/video_core/renderer_opengl/gl_graphics_pipeline.cpp
@@ -15,7 +15,7 @@
 #include "video_core/renderer_opengl/gl_shader_util.h"
 #include "video_core/renderer_opengl/gl_state_tracker.h"
 #include "video_core/shader_notify.h"
-#include "video_core/texture_cache/texture_cache_base.h"
+#include "video_core/texture_cache/texture_cache.h"
 
 #if defined(_MSC_VER) && defined(NDEBUG)
 #define LAMBDA_FORCEINLINE [[msvc::forceinline]]
@@ -27,6 +27,7 @@ namespace OpenGL {
 namespace {
 using Shader::ImageBufferDescriptor;
 using Shader::ImageDescriptor;
+using Shader::NumDescriptors;
 using Shader::TextureBufferDescriptor;
 using Shader::TextureDescriptor;
 using Tegra::Texture::TexturePair;
@@ -35,15 +36,6 @@ using VideoCommon::ImageId;
 constexpr u32 MAX_TEXTURES = 64;
 constexpr u32 MAX_IMAGES = 8;
 
-template <typename Range>
-u32 AccumulateCount(const Range& range) {
-    u32 num{};
-    for (const auto& desc : range) {
-        num += desc.count;
-    }
-    return num;
-}
-
 GLenum Stage(size_t stage_index) {
     switch (stage_index) {
     case 0:
@@ -204,23 +196,23 @@ GraphicsPipeline::GraphicsPipeline(
             base_uniform_bindings[stage + 1] = base_uniform_bindings[stage];
             base_storage_bindings[stage + 1] = base_storage_bindings[stage];
 
-            base_uniform_bindings[stage + 1] += AccumulateCount(info.constant_buffer_descriptors);
-            base_storage_bindings[stage + 1] += AccumulateCount(info.storage_buffers_descriptors);
+            base_uniform_bindings[stage + 1] += NumDescriptors(info.constant_buffer_descriptors);
+            base_storage_bindings[stage + 1] += NumDescriptors(info.storage_buffers_descriptors);
         }
         enabled_uniform_buffer_masks[stage] = info.constant_buffer_mask;
         std::ranges::copy(info.constant_buffer_used_sizes, uniform_buffer_sizes[stage].begin());
 
-        const u32 num_tex_buffer_bindings{AccumulateCount(info.texture_buffer_descriptors)};
+        const u32 num_tex_buffer_bindings{NumDescriptors(info.texture_buffer_descriptors)};
         num_texture_buffers[stage] += num_tex_buffer_bindings;
         num_textures += num_tex_buffer_bindings;
 
-        const u32 num_img_buffers_bindings{AccumulateCount(info.image_buffer_descriptors)};
+        const u32 num_img_buffers_bindings{NumDescriptors(info.image_buffer_descriptors)};
         num_image_buffers[stage] += num_img_buffers_bindings;
         num_images += num_img_buffers_bindings;
 
-        num_textures += AccumulateCount(info.texture_descriptors);
-        num_images += AccumulateCount(info.image_descriptors);
-        num_storage_buffers += AccumulateCount(info.storage_buffers_descriptors);
+        num_textures += NumDescriptors(info.texture_descriptors);
+        num_images += NumDescriptors(info.image_descriptors);
+        num_storage_buffers += NumDescriptors(info.storage_buffers_descriptors);
 
         writes_global_memory |= std::ranges::any_of(
             info.storage_buffers_descriptors, [](const auto& desc) { return desc.is_written; });
@@ -288,10 +280,9 @@ GraphicsPipeline::GraphicsPipeline(
 
 template <typename Spec>
 void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
-    std::array<ImageId, MAX_TEXTURES + MAX_IMAGES> image_view_ids;
-    std::array<u32, MAX_TEXTURES + MAX_IMAGES> image_view_indices;
+    std::array<VideoCommon::ImageViewInOut, MAX_TEXTURES + MAX_IMAGES> views;
     std::array<GLuint, MAX_TEXTURES> samplers;
-    size_t image_view_index{};
+    size_t views_index{};
     GLsizei sampler_binding{};
 
     texture_cache.SynchronizeGraphicsDescriptors();
@@ -336,30 +327,34 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
             }
             return TexturePair(gpu_memory.Read<u32>(addr), via_header_index);
         }};
-        const auto add_image{[&](const auto& desc) {
+        const auto add_image{[&](const auto& desc, bool blacklist) LAMBDA_FORCEINLINE {
             for (u32 index = 0; index < desc.count; ++index) {
                 const auto handle{read_handle(desc, index)};
-                image_view_indices[image_view_index++] = handle.first;
+                views[views_index++] = {
+                    .index = handle.first,
+                    .blacklist = blacklist,
+                    .id = {},
+                };
             }
         }};
         if constexpr (Spec::has_texture_buffers) {
             for (const auto& desc : info.texture_buffer_descriptors) {
                 for (u32 index = 0; index < desc.count; ++index) {
                     const auto handle{read_handle(desc, index)};
-                    image_view_indices[image_view_index++] = handle.first;
+                    views[views_index++] = {handle.first};
                     samplers[sampler_binding++] = 0;
                 }
             }
         }
         if constexpr (Spec::has_image_buffers) {
             for (const auto& desc : info.image_buffer_descriptors) {
-                add_image(desc);
+                add_image(desc, false);
             }
         }
         for (const auto& desc : info.texture_descriptors) {
             for (u32 index = 0; index < desc.count; ++index) {
                 const auto handle{read_handle(desc, index)};
-                image_view_indices[image_view_index++] = handle.first;
+                views[views_index++] = {handle.first};
 
                 Sampler* const sampler{texture_cache.GetGraphicsSampler(handle.second)};
                 samplers[sampler_binding++] = sampler->Handle();
@@ -367,7 +362,7 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
         }
         if constexpr (Spec::has_images) {
             for (const auto& desc : info.image_descriptors) {
-                add_image(desc);
+                add_image(desc, desc.is_written);
             }
         }
     }};
@@ -386,13 +381,12 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
     if constexpr (Spec::enabled_stages[4]) {
         config_stage(4);
     }
-    const std::span indices_span(image_view_indices.data(), image_view_index);
-    texture_cache.FillGraphicsImageViews(indices_span, image_view_ids);
+    texture_cache.FillGraphicsImageViews<Spec::has_images>(std::span(views.data(), views_index));
 
     texture_cache.UpdateRenderTargets(false);
     state_tracker.BindFramebuffer(texture_cache.GetFramebuffer()->Handle());
 
-    ImageId* texture_buffer_index{image_view_ids.data()};
+    VideoCommon::ImageViewInOut* texture_buffer_it{views.data()};
     const auto bind_stage_info{[&](size_t stage) LAMBDA_FORCEINLINE {
         size_t index{};
         const auto add_buffer{[&](const auto& desc) {
@@ -402,12 +396,12 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
                 if constexpr (is_image) {
                     is_written = desc.is_written;
                 }
-                ImageView& image_view{texture_cache.GetImageView(*texture_buffer_index)};
+                ImageView& image_view{texture_cache.GetImageView(texture_buffer_it->id)};
                 buffer_cache.BindGraphicsTextureBuffer(stage, index, image_view.GpuAddr(),
                                                        image_view.BufferSize(), image_view.format,
                                                        is_written, is_image);
                 ++index;
-                ++texture_buffer_index;
+                ++texture_buffer_it;
             }
         }};
         const Shader::Info& info{stage_infos[stage]};
@@ -423,13 +417,9 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
                 add_buffer(desc);
             }
         }
-        for (const auto& desc : info.texture_descriptors) {
-            texture_buffer_index += desc.count;
-        }
+        texture_buffer_it += Shader::NumDescriptors(info.texture_descriptors);
         if constexpr (Spec::has_images) {
-            for (const auto& desc : info.image_descriptors) {
-                texture_buffer_index += desc.count;
-            }
+            texture_buffer_it += Shader::NumDescriptors(info.image_descriptors);
         }
     }};
     if constexpr (Spec::enabled_stages[0]) {
@@ -453,12 +443,13 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
     if (!is_built.load(std::memory_order::relaxed)) {
         WaitForBuild();
     }
-    if (assembly_programs[0].handle != 0) {
+    const bool use_assembly{assembly_programs[0].handle != 0};
+    if (use_assembly) {
         program_manager.BindAssemblyPrograms(assembly_programs, enabled_stages_mask);
     } else {
         program_manager.BindSourcePrograms(source_programs);
     }
-    const ImageId* views_it{image_view_ids.data()};
+    const VideoCommon::ImageViewInOut* views_it{views.data()};
     GLsizei texture_binding = 0;
     GLsizei image_binding = 0;
     std::array<GLuint, MAX_TEXTURES> textures;
@@ -473,20 +464,49 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
         views_it += num_texture_buffers[stage];
         views_it += num_image_buffers[stage];
 
+        u32 texture_scaling_mask{};
+        u32 image_scaling_mask{};
+        u32 stage_texture_binding{};
+        u32 stage_image_binding{};
+
         const auto& info{stage_infos[stage]};
         for (const auto& desc : info.texture_descriptors) {
             for (u32 index = 0; index < desc.count; ++index) {
-                ImageView& image_view{texture_cache.GetImageView(*(views_it++))};
-                textures[texture_binding++] = image_view.Handle(desc.type);
+                ImageView& image_view{texture_cache.GetImageView((views_it++)->id)};
+                textures[texture_binding] = image_view.Handle(desc.type);
+                if (texture_cache.IsRescaling(image_view)) {
+                    texture_scaling_mask |= 1u << stage_texture_binding;
+                }
+                ++texture_binding;
+                ++stage_texture_binding;
             }
         }
         for (const auto& desc : info.image_descriptors) {
             for (u32 index = 0; index < desc.count; ++index) {
-                ImageView& image_view{texture_cache.GetImageView(*(views_it++))};
+                ImageView& image_view{texture_cache.GetImageView((views_it++)->id)};
                 if (desc.is_written) {
                     texture_cache.MarkModification(image_view.image_id);
                 }
-                images[image_binding++] = image_view.StorageView(desc.type, desc.format);
+                images[image_binding] = image_view.StorageView(desc.type, desc.format);
+                if (texture_cache.IsRescaling(image_view)) {
+                    image_scaling_mask |= 1u << stage_image_binding;
+                }
+                ++image_binding;
+                ++stage_image_binding;
+            }
+        }
+        if (info.uses_rescaling_uniform) {
+            const f32 float_texture_scaling_mask{Common::BitCast<f32>(texture_scaling_mask)};
+            const f32 float_image_scaling_mask{Common::BitCast<f32>(image_scaling_mask)};
+            const bool is_rescaling{texture_cache.IsRescaling()};
+            const f32 config_down_factor{Settings::values.resolution_info.down_factor};
+            const f32 down_factor{is_rescaling ? config_down_factor : 1.0f};
+            if (use_assembly) {
+                glProgramLocalParameter4fARB(AssemblyStage(stage), 0, float_texture_scaling_mask,
+                                             float_image_scaling_mask, down_factor, 0.0f);
+            } else {
+                glProgramUniform4f(source_programs[stage].handle, 0, float_texture_scaling_mask,
+                                   float_image_scaling_mask, down_factor, 0.0f);
             }
         }
     }};
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.cpp b/src/video_core/renderer_opengl/gl_rasterizer.cpp
index a6d9f7c43..9b516c64f 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer.cpp
+++ b/src/video_core/renderer_opengl/gl_rasterizer.cpp
@@ -184,6 +184,10 @@ void RasterizerOpenGL::Clear() {
     SyncRasterizeEnable();
     SyncStencilTestState();
 
+    std::scoped_lock lock{texture_cache.mutex};
+    texture_cache.UpdateRenderTargets(true);
+    state_tracker.BindFramebuffer(texture_cache.GetFramebuffer()->Handle());
+    SyncViewport();
     if (regs.clear_flags.scissor) {
         SyncScissorTest();
     } else {
@@ -192,10 +196,6 @@ void RasterizerOpenGL::Clear() {
     }
     UNIMPLEMENTED_IF(regs.clear_flags.viewport);
 
-    std::scoped_lock lock{texture_cache.mutex};
-    texture_cache.UpdateRenderTargets(true);
-    state_tracker.BindFramebuffer(texture_cache.GetFramebuffer()->Handle());
-
     if (use_color) {
         glClearBufferfv(GL_COLOR, regs.clear_buffers.RT, regs.clear_color);
     }
@@ -214,8 +214,6 @@ void RasterizerOpenGL::Draw(bool is_indexed, bool is_instanced) {
 
     query_cache.UpdateCounters();
 
-    SyncState();
-
     GraphicsPipeline* const pipeline{shader_cache.CurrentGraphicsPipeline()};
     if (!pipeline) {
         return;
@@ -223,6 +221,8 @@ void RasterizerOpenGL::Draw(bool is_indexed, bool is_instanced) {
     std::scoped_lock lock{buffer_cache.mutex, texture_cache.mutex};
     pipeline->Configure(is_indexed);
 
+    SyncState();
+
     const GLenum primitive_mode = MaxwellToGL::PrimitiveTopology(maxwell3d.regs.draw.topology);
     BeginTransformFeedback(pipeline, primitive_mode);
 
@@ -533,7 +533,8 @@ void RasterizerOpenGL::SyncViewport() {
     auto& flags = maxwell3d.dirty.flags;
     const auto& regs = maxwell3d.regs;
 
-    const bool dirty_viewport = flags[Dirty::Viewports];
+    const bool rescale_viewports = flags[VideoCommon::Dirty::RescaleViewports];
+    const bool dirty_viewport = flags[Dirty::Viewports] || rescale_viewports;
     const bool dirty_clip_control = flags[Dirty::ClipControl];
 
     if (dirty_clip_control || flags[Dirty::FrontFace]) {
@@ -553,8 +554,7 @@ void RasterizerOpenGL::SyncViewport() {
         }
         glFrontFace(mode);
     }
-
-    if (dirty_viewport || flags[Dirty::ClipControl]) {
+    if (dirty_viewport || dirty_clip_control) {
         flags[Dirty::ClipControl] = false;
 
         bool flip_y = false;
@@ -570,37 +570,58 @@ void RasterizerOpenGL::SyncViewport() {
         state_tracker.ClipControl(origin, depth);
         state_tracker.SetYNegate(regs.screen_y_control.y_negate != 0);
     }
+    const bool is_rescaling{texture_cache.IsRescaling()};
+    const float scale = is_rescaling ? Settings::values.resolution_info.up_factor : 1.0f;
+    const auto conv = [scale](float value) -> GLfloat {
+        float new_value = value * scale;
+        if (scale < 1.0f) {
+            const bool sign = std::signbit(value);
+            new_value = std::round(std::abs(new_value));
+            new_value = sign ? -new_value : new_value;
+        }
+        return static_cast<GLfloat>(new_value);
+    };
 
     if (dirty_viewport) {
         flags[Dirty::Viewports] = false;
 
-        const bool force = flags[Dirty::ViewportTransform];
+        const bool force = flags[Dirty::ViewportTransform] || rescale_viewports;
         flags[Dirty::ViewportTransform] = false;
+        flags[VideoCommon::Dirty::RescaleViewports] = false;
 
-        for (std::size_t i = 0; i < Maxwell::NumViewports; ++i) {
-            if (!force && !flags[Dirty::Viewport0 + i]) {
+        for (size_t index = 0; index < Maxwell::NumViewports; ++index) {
+            if (!force && !flags[Dirty::Viewport0 + index]) {
                 continue;
             }
-            flags[Dirty::Viewport0 + i] = false;
+            flags[Dirty::Viewport0 + index] = false;
 
-            const auto& src = regs.viewport_transform[i];
-            const Common::Rectangle<f32> rect{src.GetRect()};
-            glViewportIndexedf(static_cast<GLuint>(i), rect.left, rect.bottom, rect.GetWidth(),
-                               rect.GetHeight());
+            const auto& src = regs.viewport_transform[index];
+            GLfloat x = conv(src.translate_x - src.scale_x);
+            GLfloat y = conv(src.translate_y - src.scale_y);
+            GLfloat width = conv(src.scale_x * 2.0f);
+            GLfloat height = conv(src.scale_y * 2.0f);
+
+            if (height < 0) {
+                y += height;
+                height = -height;
+            }
+            glViewportIndexedf(static_cast<GLuint>(index), x, y, width != 0.0f ? width : 1.0f,
+                               height != 0.0f ? height : 1.0f);
 
             const GLdouble reduce_z = regs.depth_mode == Maxwell::DepthMode::MinusOneToOne;
             const GLdouble near_depth = src.translate_z - src.scale_z * reduce_z;
             const GLdouble far_depth = src.translate_z + src.scale_z;
             if (device.HasDepthBufferFloat()) {
-                glDepthRangeIndexeddNV(static_cast<GLuint>(i), near_depth, far_depth);
+                glDepthRangeIndexeddNV(static_cast<GLuint>(index), near_depth, far_depth);
             } else {
-                glDepthRangeIndexed(static_cast<GLuint>(i), near_depth, far_depth);
+                glDepthRangeIndexed(static_cast<GLuint>(index), near_depth, far_depth);
             }
 
             if (!GLAD_GL_NV_viewport_swizzle) {
                 continue;
             }
-            glViewportSwizzleNV(static_cast<GLuint>(i), MaxwellToGL::ViewportSwizzle(src.swizzle.x),
+            glViewportSwizzleNV(static_cast<GLuint>(index),
+                                MaxwellToGL::ViewportSwizzle(src.swizzle.x),
                                 MaxwellToGL::ViewportSwizzle(src.swizzle.y),
                                 MaxwellToGL::ViewportSwizzle(src.swizzle.z),
                                 MaxwellToGL::ViewportSwizzle(src.swizzle.w));
@@ -903,14 +924,34 @@ void RasterizerOpenGL::SyncLogicOpState() {
 
 void RasterizerOpenGL::SyncScissorTest() {
     auto& flags = maxwell3d.dirty.flags;
-    if (!flags[Dirty::Scissors]) {
+    if (!flags[Dirty::Scissors] && !flags[VideoCommon::Dirty::RescaleScissors]) {
         return;
     }
     flags[Dirty::Scissors] = false;
 
+    const bool force = flags[VideoCommon::Dirty::RescaleScissors];
+    flags[VideoCommon::Dirty::RescaleScissors] = false;
+
     const auto& regs = maxwell3d.regs;
+
+    const auto& resolution = Settings::values.resolution_info;
+    const bool is_rescaling{texture_cache.IsRescaling()};
+    const u32 up_scale = is_rescaling ? resolution.up_scale : 1U;
+    const u32 down_shift = is_rescaling ? resolution.down_shift : 0U;
+    const auto scale_up = [up_scale, down_shift](u32 value) -> u32 {
+        if (value == 0) {
+            return 0U;
+        }
+        const u32 upset = value * up_scale;
+        u32 acumm{};
+        if ((up_scale >> down_shift) == 0) {
+            acumm = upset % 2;
+        }
+        const u32 converted_value = upset >> down_shift;
+        return std::max<u32>(converted_value + acumm, 1U);
+    };
     for (std::size_t index = 0; index < Maxwell::NumViewports; ++index) {
-        if (!flags[Dirty::Scissor0 + index]) {
+        if (!force && !flags[Dirty::Scissor0 + index]) {
             continue;
         }
         flags[Dirty::Scissor0 + index] = false;
@@ -918,8 +959,8 @@ void RasterizerOpenGL::SyncScissorTest() {
         const auto& src = regs.scissor_test[index];
         if (src.enable) {
             glEnablei(GL_SCISSOR_TEST, static_cast<GLuint>(index));
-            glScissorIndexed(static_cast<GLuint>(index), src.min_x, src.min_y,
-                             src.max_x - src.min_x, src.max_y - src.min_y);
+            glScissorIndexed(static_cast<GLuint>(index), scale_up(src.min_x), scale_up(src.min_y),
+                             scale_up(src.max_x - src.min_x), scale_up(src.max_y - src.min_y));
         } else {
             glDisablei(GL_SCISSOR_TEST, static_cast<GLuint>(index));
         }
@@ -935,8 +976,9 @@ void RasterizerOpenGL::SyncPointState() {
 
     oglEnable(GL_POINT_SPRITE, maxwell3d.regs.point_sprite_enable);
     oglEnable(GL_PROGRAM_POINT_SIZE, maxwell3d.regs.vp_point_size.enable);
-
-    glPointSize(std::max(1.0f, maxwell3d.regs.point_size));
+    const bool is_rescaling{texture_cache.IsRescaling()};
+    const float scale = is_rescaling ? Settings::values.resolution_info.up_factor : 1.0f;
+    glPointSize(std::max(1.0f, maxwell3d.regs.point_size * scale));
 }
 
 void RasterizerOpenGL::SyncLineState() {
diff --git a/src/video_core/renderer_opengl/gl_resource_manager.cpp b/src/video_core/renderer_opengl/gl_resource_manager.cpp
index 8695c29e3..5e7101d28 100644
--- a/src/video_core/renderer_opengl/gl_resource_manager.cpp
+++ b/src/video_core/renderer_opengl/gl_resource_manager.cpp
@@ -166,7 +166,12 @@ void OGLFramebuffer::Create() {
         return;
 
     MICROPROFILE_SCOPE(OpenGL_ResourceCreation);
+    // Bind to READ_FRAMEBUFFER to stop Nvidia's driver from creating an EXT_framebuffer instead of
+    // a core framebuffer. EXT framebuffer attachments have to match in size and can be shared
+    // across contexts. yuzu doesn't share framebuffers across contexts and we need attachments with
+    // mismatching size, this is why core framebuffers are preferred.
     glGenFramebuffers(1, &handle);
+    glBindFramebuffer(GL_READ_FRAMEBUFFER, handle);
 }
 
 void OGLFramebuffer::Release() {
diff --git a/src/video_core/renderer_opengl/gl_shader_cache.cpp b/src/video_core/renderer_opengl/gl_shader_cache.cpp
index 02682bd76..42ef67628 100644
--- a/src/video_core/renderer_opengl/gl_shader_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_shader_cache.cpp
@@ -426,16 +426,14 @@ std::unique_ptr<GraphicsPipeline> ShaderCache::CreateGraphicsPipeline(
             // Normal path
             programs[index] = TranslateProgram(pools.inst, pools.block, env, cfg, host_info);
 
-            for (const auto& desc : programs[index].info.storage_buffers_descriptors) {
-                total_storage_buffers += desc.count;
-            }
+            total_storage_buffers +=
+                Shader::NumDescriptors(programs[index].info.storage_buffers_descriptors);
         } else {
             // VertexB path when VertexA is present.
             auto& program_va{programs[0]};
             auto program_vb{TranslateProgram(pools.inst, pools.block, env, cfg, host_info)};
-            for (const auto& desc : program_vb.info.storage_buffers_descriptors) {
-                total_storage_buffers += desc.count;
-            }
+            total_storage_buffers +=
+                Shader::NumDescriptors(program_vb.info.storage_buffers_descriptors);
             programs[index] = MergeDualVertexPrograms(program_va, program_vb, env);
         }
     }
@@ -510,10 +508,7 @@ std::unique_ptr<ComputePipeline> ShaderCache::CreateComputePipeline(
     Shader::Maxwell::Flow::CFG cfg{env, pools.flow_block, env.StartAddress()};
     auto program{TranslateProgram(pools.inst, pools.block, env, cfg, host_info)};
 
-    u32 num_storage_buffers{};
-    for (const auto& desc : program.info.storage_buffers_descriptors) {
-        num_storage_buffers += desc.count;
-    }
+    const u32 num_storage_buffers{Shader::NumDescriptors(program.info.storage_buffers_descriptors)};
     Shader::RuntimeInfo info;
     info.glasm_use_storage_buffers = num_storage_buffers <= device.GetMaxGLASMStorageBufferBlocks();
 
diff --git a/src/video_core/renderer_opengl/gl_texture_cache.cpp b/src/video_core/renderer_opengl/gl_texture_cache.cpp
index 8c3ca3d82..2f7d98d8b 100644
--- a/src/video_core/renderer_opengl/gl_texture_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_texture_cache.cpp
@@ -9,8 +9,8 @@
 
 #include <glad/glad.h>
 
+#include "common/literals.h"
 #include "common/settings.h"
-
 #include "video_core/renderer_opengl/gl_device.h"
 #include "video_core/renderer_opengl/gl_shader_manager.h"
 #include "video_core/renderer_opengl/gl_state_tracker.h"
@@ -42,6 +42,7 @@ using VideoCore::Surface::IsPixelFormatSRGB;
 using VideoCore::Surface::MaxPixelFormat;
 using VideoCore::Surface::PixelFormat;
 using VideoCore::Surface::SurfaceType;
+using namespace Common::Literals;
 
 struct CopyOrigin {
     GLint level;
@@ -316,6 +317,52 @@ void AttachTexture(GLuint fbo, GLenum attachment, const ImageView* image_view) {
     }
 }
 
+OGLTexture MakeImage(const VideoCommon::ImageInfo& info, GLenum gl_internal_format) {
+    const GLenum target = ImageTarget(info);
+    const GLsizei width = info.size.width;
+    const GLsizei height = info.size.height;
+    const GLsizei depth = info.size.depth;
+    const int max_host_mip_levels = std::bit_width(info.size.width);
+    const GLsizei num_levels = std::min(info.resources.levels, max_host_mip_levels);
+    const GLsizei num_layers = info.resources.layers;
+    const GLsizei num_samples = info.num_samples;
+
+    GLuint handle = 0;
+    OGLTexture texture;
+    if (target != GL_TEXTURE_BUFFER) {
+        texture.Create(target);
+        handle = texture.handle;
+    }
+    switch (target) {
+    case GL_TEXTURE_1D_ARRAY:
+        glTextureStorage2D(handle, num_levels, gl_internal_format, width, num_layers);
+        break;
+    case GL_TEXTURE_2D_ARRAY:
+        glTextureStorage3D(handle, num_levels, gl_internal_format, width, height, num_layers);
+        break;
+    case GL_TEXTURE_2D_MULTISAMPLE_ARRAY: {
+        // TODO: Where should 'fixedsamplelocations' come from?
+        const auto [samples_x, samples_y] = SamplesLog2(info.num_samples);
+        glTextureStorage3DMultisample(handle, num_samples, gl_internal_format, width >> samples_x,
+                                      height >> samples_y, num_layers, GL_FALSE);
+        break;
+    }
+    case GL_TEXTURE_RECTANGLE:
+        glTextureStorage2D(handle, num_levels, gl_internal_format, width, height);
+        break;
+    case GL_TEXTURE_3D:
+        glTextureStorage3D(handle, num_levels, gl_internal_format, width, height, depth);
+        break;
+    case GL_TEXTURE_BUFFER:
+        UNREACHABLE();
+        break;
+    default:
+        UNREACHABLE_MSG("Invalid target=0x{:x}", target);
+        break;
+    }
+    return texture;
+}
+
 [[nodiscard]] bool IsPixelFormatBGR(PixelFormat format) {
     switch (format) {
     case PixelFormat::B5G6R5_UNORM:
@@ -359,7 +406,8 @@ ImageBufferMap::~ImageBufferMap() {
 
 TextureCacheRuntime::TextureCacheRuntime(const Device& device_, ProgramManager& program_manager,
                                          StateTracker& state_tracker_)
-    : device{device_}, state_tracker{state_tracker_}, util_shaders(program_manager) {
+    : device{device_}, state_tracker{state_tracker_},
+      util_shaders(program_manager), resolution{Settings::values.resolution_info} {
     static constexpr std::array TARGETS{GL_TEXTURE_1D_ARRAY, GL_TEXTURE_2D_ARRAY, GL_TEXTURE_3D};
     for (size_t i = 0; i < TARGETS.size(); ++i) {
         const GLenum target = TARGETS[i];
@@ -426,6 +474,13 @@ TextureCacheRuntime::TextureCacheRuntime(const Device& device_, ProgramManager&
     set_view(Shader::TextureType::ColorArray1D, null_image_1d_array.handle);
     set_view(Shader::TextureType::ColorArray2D, null_image_view_2d_array.handle);
     set_view(Shader::TextureType::ColorArrayCube, null_image_cube_array.handle);
+
+    if (resolution.active) {
+        for (size_t i = 0; i < rescale_draw_fbos.size(); ++i) {
+            rescale_draw_fbos[i].Create();
+            rescale_read_fbos[i].Create();
+        }
+    }
 }
 
 TextureCacheRuntime::~TextureCacheRuntime() = default;
@@ -442,6 +497,15 @@ ImageBufferMap TextureCacheRuntime::DownloadStagingBuffer(size_t size) {
     return download_buffers.RequestMap(size, false);
 }
 
+u64 TextureCacheRuntime::GetDeviceLocalMemory() const {
+    if (GLAD_GL_NVX_gpu_memory_info) {
+        GLint cur_avail_mem_kb = 0;
+        glGetIntegerv(GL_GPU_MEMORY_INFO_CURRENT_AVAILABLE_VIDMEM_NVX, &cur_avail_mem_kb);
+        return static_cast<u64>(cur_avail_mem_kb) * 1_KiB;
+    }
+    return 2_GiB; // Return minimum requirements
+}
+
 void TextureCacheRuntime::CopyImage(Image& dst_image, Image& src_image,
                                     std::span<const ImageCopy> copies) {
     const GLuint dst_name = dst_image.Handle();
@@ -605,13 +669,13 @@ std::optional<size_t> TextureCacheRuntime::StagingBuffers::FindBuffer(size_t req
     return found;
 }
 
-Image::Image(TextureCacheRuntime& runtime, const VideoCommon::ImageInfo& info_, GPUVAddr gpu_addr_,
+Image::Image(TextureCacheRuntime& runtime_, const VideoCommon::ImageInfo& info_, GPUVAddr gpu_addr_,
              VAddr cpu_addr_)
-    : VideoCommon::ImageBase(info_, gpu_addr_, cpu_addr_) {
-    if (CanBeAccelerated(runtime, info)) {
+    : VideoCommon::ImageBase(info_, gpu_addr_, cpu_addr_), runtime{&runtime_} {
+    if (CanBeAccelerated(*runtime, info)) {
         flags |= ImageFlagBits::AcceleratedUpload;
     }
-    if (IsConverted(runtime.device, info.format, info.type)) {
+    if (IsConverted(runtime->device, info.format, info.type)) {
         flags |= ImageFlagBits::Converted;
         gl_internal_format = IsPixelFormatSRGB(info.format) ? GL_SRGB8_ALPHA8 : GL_RGBA8;
         gl_format = GL_RGBA;
@@ -622,58 +686,25 @@ Image::Image(TextureCacheRuntime& runtime, const VideoCommon::ImageInfo& info_,
         gl_format = tuple.format;
         gl_type = tuple.type;
     }
-    const GLenum target = ImageTarget(info);
-    const GLsizei width = info.size.width;
-    const GLsizei height = info.size.height;
-    const GLsizei depth = info.size.depth;
-    const int max_host_mip_levels = std::bit_width(info.size.width);
-    const GLsizei num_levels = std::min(info.resources.levels, max_host_mip_levels);
-    const GLsizei num_layers = info.resources.layers;
-    const GLsizei num_samples = info.num_samples;
-
-    GLuint handle = 0;
-    if (target != GL_TEXTURE_BUFFER) {
-        texture.Create(target);
-        handle = texture.handle;
-    }
-    switch (target) {
-    case GL_TEXTURE_1D_ARRAY:
-        glTextureStorage2D(handle, num_levels, gl_internal_format, width, num_layers);
-        break;
-    case GL_TEXTURE_2D_ARRAY:
-        glTextureStorage3D(handle, num_levels, gl_internal_format, width, height, num_layers);
-        break;
-    case GL_TEXTURE_2D_MULTISAMPLE_ARRAY: {
-        // TODO: Where should 'fixedsamplelocations' come from?
-        const auto [samples_x, samples_y] = SamplesLog2(info.num_samples);
-        glTextureStorage3DMultisample(handle, num_samples, gl_internal_format, width >> samples_x,
-                                      height >> samples_y, num_layers, GL_FALSE);
-        break;
-    }
-    case GL_TEXTURE_RECTANGLE:
-        glTextureStorage2D(handle, num_levels, gl_internal_format, width, height);
-        break;
-    case GL_TEXTURE_3D:
-        glTextureStorage3D(handle, num_levels, gl_internal_format, width, height, depth);
-        break;
-    case GL_TEXTURE_BUFFER:
-        UNREACHABLE();
-        break;
-    default:
-        UNREACHABLE_MSG("Invalid target=0x{:x}", target);
-        break;
-    }
-    if (runtime.device.HasDebuggingToolAttached()) {
+    texture = MakeImage(info, gl_internal_format);
+    current_texture = texture.handle;
+    if (runtime->device.HasDebuggingToolAttached()) {
         const std::string name = VideoCommon::Name(*this);
-        glObjectLabel(target == GL_TEXTURE_BUFFER ? GL_BUFFER : GL_TEXTURE, handle,
-                      static_cast<GLsizei>(name.size()), name.data());
+        glObjectLabel(ImageTarget(info) == GL_TEXTURE_BUFFER ? GL_BUFFER : GL_TEXTURE,
+                      texture.handle, static_cast<GLsizei>(name.size()), name.data());
     }
 }
 
+Image::Image(const VideoCommon::NullImageParams& params) : VideoCommon::ImageBase{params} {}
+
 Image::~Image() = default;
 
 void Image::UploadMemory(const ImageBufferMap& map,
                          std::span<const VideoCommon::BufferImageCopy> copies) {
+    const bool is_rescaled = True(flags & ImageFlagBits::Rescaled);
+    if (is_rescaled) {
+        ScaleDown(true);
+    }
     glBindBuffer(GL_PIXEL_UNPACK_BUFFER, map.buffer);
     glFlushMappedBufferRange(GL_PIXEL_UNPACK_BUFFER, map.offset, unswizzled_size_bytes);
 
@@ -693,12 +724,18 @@ void Image::UploadMemory(const ImageBufferMap& map,
         }
         CopyBufferToImage(copy, map.offset);
     }
+    if (is_rescaled) {
+        ScaleUp();
+    }
 }
 
 void Image::DownloadMemory(ImageBufferMap& map,
                            std::span<const VideoCommon::BufferImageCopy> copies) {
+    const bool is_rescaled = True(flags & ImageFlagBits::Rescaled);
+    if (is_rescaled) {
+        ScaleDown();
+    }
     glMemoryBarrier(GL_PIXEL_BUFFER_BARRIER_BIT); // TODO: Move this to its own API
-
     glBindBuffer(GL_PIXEL_PACK_BUFFER, map.buffer);
     glPixelStorei(GL_PACK_ALIGNMENT, 1);
 
@@ -716,6 +753,9 @@ void Image::DownloadMemory(ImageBufferMap& map,
         }
         CopyImageToBuffer(copy, map.offset);
     }
+    if (is_rescaled) {
+        ScaleUp(true);
+    }
 }
 
 GLuint Image::StorageHandle() noexcept {
@@ -741,11 +781,11 @@ GLuint Image::StorageHandle() noexcept {
             return store_view.handle;
         }
         store_view.Create();
-        glTextureView(store_view.handle, ImageTarget(info), texture.handle, GL_RGBA8, 0,
+        glTextureView(store_view.handle, ImageTarget(info), current_texture, GL_RGBA8, 0,
                       info.resources.levels, 0, info.resources.layers);
         return store_view.handle;
     default:
-        return texture.handle;
+        return current_texture;
     }
 }
 
@@ -849,6 +889,140 @@ void Image::CopyImageToBuffer(const VideoCommon::BufferImageCopy& copy, size_t b
     }
 }
 
+void Image::Scale(bool up_scale) {
+    const auto format_type = GetFormatType(info.format);
+    const GLenum attachment = [format_type] {
+        switch (format_type) {
+        case SurfaceType::ColorTexture:
+            return GL_COLOR_ATTACHMENT0;
+        case SurfaceType::Depth:
+            return GL_DEPTH_ATTACHMENT;
+        case SurfaceType::DepthStencil:
+            return GL_DEPTH_STENCIL_ATTACHMENT;
+        default:
+            UNREACHABLE();
+            return GL_COLOR_ATTACHMENT0;
+        }
+    }();
+    const GLenum mask = [format_type] {
+        switch (format_type) {
+        case SurfaceType::ColorTexture:
+            return GL_COLOR_BUFFER_BIT;
+        case SurfaceType::Depth:
+            return GL_DEPTH_BUFFER_BIT;
+        case SurfaceType::DepthStencil:
+            return GL_STENCIL_BUFFER_BIT | GL_DEPTH_BUFFER_BIT;
+        default:
+            UNREACHABLE();
+            return GL_COLOR_BUFFER_BIT;
+        }
+    }();
+    const size_t fbo_index = [format_type] {
+        switch (format_type) {
+        case SurfaceType::ColorTexture:
+            return 0;
+        case SurfaceType::Depth:
+            return 1;
+        case SurfaceType::DepthStencil:
+            return 2;
+        default:
+            UNREACHABLE();
+            return 0;
+        }
+    }();
+    const bool is_2d = info.type == ImageType::e2D;
+    const bool is_color{(mask & GL_COLOR_BUFFER_BIT) != 0};
+    // Integer formats must use NEAREST filter
+    const bool linear_color_format{is_color && !IsPixelFormatInteger(info.format)};
+    const GLenum filter = linear_color_format ? GL_LINEAR : GL_NEAREST;
+
+    const auto& resolution = runtime->resolution;
+    const u32 scaled_width = resolution.ScaleUp(info.size.width);
+    const u32 scaled_height = is_2d ? resolution.ScaleUp(info.size.height) : info.size.height;
+    const u32 original_width = info.size.width;
+    const u32 original_height = info.size.height;
+
+    if (!upscaled_backup.handle) {
+        auto dst_info = info;
+        dst_info.size.width = scaled_width;
+        dst_info.size.height = scaled_height;
+        upscaled_backup = MakeImage(dst_info, gl_internal_format);
+    }
+    const u32 src_width = up_scale ? original_width : scaled_width;
+    const u32 src_height = up_scale ? original_height : scaled_height;
+    const u32 dst_width = up_scale ? scaled_width : original_width;
+    const u32 dst_height = up_scale ? scaled_height : original_height;
+    const auto src_handle = up_scale ? texture.handle : upscaled_backup.handle;
+    const auto dst_handle = up_scale ? upscaled_backup.handle : texture.handle;
+
+    // TODO (ameerj): Investigate other GL states that affect blitting.
+    glDisablei(GL_SCISSOR_TEST, 0);
+    glViewportIndexedf(0, 0.0f, 0.0f, static_cast<GLfloat>(dst_width),
+                       static_cast<GLfloat>(dst_height));
+
+    const GLuint read_fbo = runtime->rescale_read_fbos[fbo_index].handle;
+    const GLuint draw_fbo = runtime->rescale_draw_fbos[fbo_index].handle;
+    for (s32 layer = 0; layer < info.resources.layers; ++layer) {
+        for (s32 level = 0; level < info.resources.levels; ++level) {
+            const u32 src_level_width = std::max(1u, src_width >> level);
+            const u32 src_level_height = std::max(1u, src_height >> level);
+            const u32 dst_level_width = std::max(1u, dst_width >> level);
+            const u32 dst_level_height = std::max(1u, dst_height >> level);
+
+            glNamedFramebufferTextureLayer(read_fbo, attachment, src_handle, level, layer);
+            glNamedFramebufferTextureLayer(draw_fbo, attachment, dst_handle, level, layer);
+
+            glBlitNamedFramebuffer(read_fbo, draw_fbo, 0, 0, src_level_width, src_level_height, 0,
+                                   0, dst_level_width, dst_level_height, mask, filter);
+        }
+    }
+    current_texture = dst_handle;
+    auto& state_tracker = runtime->GetStateTracker();
+    state_tracker.NotifyViewport0();
+    state_tracker.NotifyScissor0();
+}
+
+bool Image::ScaleUp(bool ignore) {
+    if (True(flags & ImageFlagBits::Rescaled)) {
+        return false;
+    }
+    if (gl_format == 0 && gl_type == 0) {
+        // compressed textures
+        return false;
+    }
+    if (info.type == ImageType::Linear) {
+        UNREACHABLE();
+        return false;
+    }
+    flags |= ImageFlagBits::Rescaled;
+    if (!runtime->resolution.active) {
+        return false;
+    }
+    has_scaled = true;
+    if (ignore) {
+        current_texture = upscaled_backup.handle;
+        return true;
+    }
+    Scale(true);
+    return true;
+}
+
+bool Image::ScaleDown(bool ignore) {
+    if (False(flags & ImageFlagBits::Rescaled)) {
+        return false;
+    }
+    flags &= ~ImageFlagBits::Rescaled;
+    if (!runtime->resolution.active) {
+        return false;
+    }
+    if (ignore) {
+        current_texture = texture.handle;
+        return true;
+    }
+    Scale(false);
+    return true;
+}
+
 ImageView::ImageView(TextureCacheRuntime& runtime, const VideoCommon::ImageViewInfo& info,
                      ImageId image_id_, Image& image)
     : VideoCommon::ImageViewBase{info, image.info, image_id_}, views{runtime.null_image_views} {
@@ -862,7 +1036,7 @@ ImageView::ImageView(TextureCacheRuntime& runtime, const VideoCommon::ImageViewI
     flat_range = info.range;
     set_object_label = device.HasDebuggingToolAttached();
     is_render_target = info.IsRenderTarget();
-    original_texture = image.texture.handle;
+    original_texture = image.Handle();
     num_samples = image.info.num_samples;
     if (!is_render_target) {
         swizzle[0] = info.x_source;
@@ -950,9 +1124,11 @@ ImageView::ImageView(TextureCacheRuntime&, const VideoCommon::ImageInfo& info,
                      const VideoCommon::ImageViewInfo& view_info)
     : VideoCommon::ImageViewBase{info, view_info} {}
 
-ImageView::ImageView(TextureCacheRuntime& runtime, const VideoCommon::NullImageParams& params)
+ImageView::ImageView(TextureCacheRuntime& runtime, const VideoCommon::NullImageViewParams& params)
     : VideoCommon::ImageViewBase{params}, views{runtime.null_image_views} {}
 
+ImageView::~ImageView() = default;
+
 GLuint ImageView::StorageView(Shader::TextureType texture_type, Shader::ImageFormat image_format) {
     if (image_format == Shader::ImageFormat::Typeless) {
         return Handle(texture_type);
@@ -1037,7 +1213,8 @@ Sampler::Sampler(TextureCacheRuntime& runtime, const TSCEntry& config) {
     glSamplerParameterfv(handle, GL_TEXTURE_BORDER_COLOR, config.BorderColor().data());
 
     if (GLAD_GL_ARB_texture_filter_anisotropic || GLAD_GL_EXT_texture_filter_anisotropic) {
-        glSamplerParameterf(handle, GL_TEXTURE_MAX_ANISOTROPY, config.MaxAnisotropy());
+        const f32 max_anisotropy = std::clamp(config.MaxAnisotropy(), 1.0f, 16.0f);
+        glSamplerParameterf(handle, GL_TEXTURE_MAX_ANISOTROPY, max_anisotropy);
     } else {
         LOG_WARNING(Render_OpenGL, "GL_ARB_texture_filter_anisotropic is required");
     }
@@ -1056,13 +1233,8 @@ Sampler::Sampler(TextureCacheRuntime& runtime, const TSCEntry& config) {
 
 Framebuffer::Framebuffer(TextureCacheRuntime& runtime, std::span<ImageView*, NUM_RT> color_buffers,
                          ImageView* depth_buffer, const VideoCommon::RenderTargets& key) {
-    // Bind to READ_FRAMEBUFFER to stop Nvidia's driver from creating an EXT_framebuffer instead of
-    // a core framebuffer. EXT framebuffer attachments have to match in size and can be shared
-    // across contexts. yuzu doesn't share framebuffers across contexts and we need attachments with
-    // mismatching size, this is why core framebuffers are preferred.
-    GLuint handle;
-    glGenFramebuffers(1, &handle);
-    glBindFramebuffer(GL_READ_FRAMEBUFFER, handle);
+    framebuffer.Create();
+    GLuint handle = framebuffer.handle;
 
     GLsizei num_buffers = 0;
     std::array<GLenum, NUM_RT> gl_draw_buffers;
@@ -1110,31 +1282,31 @@ Framebuffer::Framebuffer(TextureCacheRuntime& runtime, std::span<ImageView*, NUM
         const std::string name = VideoCommon::Name(key);
         glObjectLabel(GL_FRAMEBUFFER, handle, static_cast<GLsizei>(name.size()), name.data());
     }
-    framebuffer.handle = handle;
 }
 
+Framebuffer::~Framebuffer() = default;
+
 void BGRCopyPass::CopyBGR(Image& dst_image, Image& src_image,
                           std::span<const VideoCommon::ImageCopy> copies) {
     static constexpr VideoCommon::Offset3D zero_offset{0, 0, 0};
-    const u32 requested_pbo_size =
-        std::max(src_image.unswizzled_size_bytes, dst_image.unswizzled_size_bytes);
-
-    if (bgr_pbo_size < requested_pbo_size) {
-        bgr_pbo.Create();
-        bgr_pbo_size = requested_pbo_size;
-        glNamedBufferData(bgr_pbo.handle, bgr_pbo_size, nullptr, GL_STREAM_COPY);
-    }
+    const u32 img_bpp = BytesPerBlock(src_image.info.format);
     for (const ImageCopy& copy : copies) {
         ASSERT(copy.src_offset == zero_offset);
         ASSERT(copy.dst_offset == zero_offset);
-
+        const u32 num_src_layers = static_cast<u32>(copy.src_subresource.num_layers);
+        const u32 copy_size = copy.extent.width * copy.extent.height * num_src_layers * img_bpp;
+        if (bgr_pbo_size < copy_size) {
+            bgr_pbo.Create();
+            bgr_pbo_size = copy_size;
+            glNamedBufferData(bgr_pbo.handle, bgr_pbo_size, nullptr, GL_STREAM_COPY);
+        }
         // Copy from source to PBO
         glPixelStorei(GL_PACK_ALIGNMENT, 1);
         glPixelStorei(GL_PACK_ROW_LENGTH, copy.extent.width);
         glBindBuffer(GL_PIXEL_PACK_BUFFER, bgr_pbo.handle);
         glGetTextureSubImage(src_image.Handle(), 0, 0, 0, 0, copy.extent.width, copy.extent.height,
-                             copy.src_subresource.num_layers, src_image.GlFormat(),
-                             src_image.GlType(), static_cast<GLsizei>(bgr_pbo_size), nullptr);
+                             num_src_layers, src_image.GlFormat(), src_image.GlType(),
+                             static_cast<GLsizei>(bgr_pbo_size), nullptr);
 
         // Copy from PBO to destination in desired GL format
         glPixelStorei(GL_UNPACK_ALIGNMENT, 1);
diff --git a/src/video_core/renderer_opengl/gl_texture_cache.h b/src/video_core/renderer_opengl/gl_texture_cache.h
index 1ca2c90be..1bb762568 100644
--- a/src/video_core/renderer_opengl/gl_texture_cache.h
+++ b/src/video_core/renderer_opengl/gl_texture_cache.h
@@ -15,6 +15,10 @@
 #include "video_core/texture_cache/image_view_base.h"
 #include "video_core/texture_cache/texture_cache_base.h"
 
+namespace Settings {
+struct ResolutionScalingInfo;
+}
+
 namespace OpenGL {
 
 class Device;
@@ -78,9 +82,11 @@ public:
 
     ImageBufferMap DownloadStagingBuffer(size_t size);
 
+    u64 GetDeviceLocalMemory() const;
+
     void CopyImage(Image& dst, Image& src, std::span<const VideoCommon::ImageCopy> copies);
 
-    void ConvertImage(Framebuffer* dst, ImageView& dst_view, ImageView& src_view) {
+    void ConvertImage(Framebuffer* dst, ImageView& dst_view, ImageView& src_view, bool rescaled) {
         UNIMPLEMENTED();
     }
 
@@ -110,6 +116,12 @@ public:
 
     bool HasNativeASTC() const noexcept;
 
+    void TickFrame() {}
+
+    StateTracker& GetStateTracker() {
+        return state_tracker;
+    }
+
 private:
     struct StagingBuffers {
         explicit StagingBuffers(GLenum storage_flags_, GLenum map_flags_);
@@ -149,6 +161,10 @@ private:
     OGLTextureView null_image_view_cube;
 
     std::array<GLuint, Shader::NUM_TEXTURE_TYPES> null_image_views{};
+
+    std::array<OGLFramebuffer, 3> rescale_draw_fbos;
+    std::array<OGLFramebuffer, 3> rescale_read_fbos;
+    const Settings::ResolutionScalingInfo& resolution;
 };
 
 class Image : public VideoCommon::ImageBase {
@@ -157,6 +173,7 @@ class Image : public VideoCommon::ImageBase {
 public:
     explicit Image(TextureCacheRuntime&, const VideoCommon::ImageInfo& info, GPUVAddr gpu_addr,
                    VAddr cpu_addr);
+    explicit Image(const VideoCommon::NullImageParams&);
 
     ~Image();
 
@@ -174,7 +191,7 @@ public:
     GLuint StorageHandle() noexcept;
 
     GLuint Handle() const noexcept {
-        return texture.handle;
+        return current_texture;
     }
 
     GLuint GlFormat() const noexcept {
@@ -185,16 +202,25 @@ public:
         return gl_type;
     }
 
+    bool ScaleUp(bool ignore = false);
+
+    bool ScaleDown(bool ignore = false);
+
 private:
     void CopyBufferToImage(const VideoCommon::BufferImageCopy& copy, size_t buffer_offset);
 
     void CopyImageToBuffer(const VideoCommon::BufferImageCopy& copy, size_t buffer_offset);
 
+    void Scale(bool up_scale);
+
     OGLTexture texture;
+    OGLTexture upscaled_backup;
     OGLTextureView store_view;
     GLenum gl_internal_format = GL_NONE;
     GLenum gl_format = GL_NONE;
     GLenum gl_type = GL_NONE;
+    TextureCacheRuntime* runtime{};
+    GLuint current_texture{};
 };
 
 class ImageView : public VideoCommon::ImageViewBase {
@@ -206,7 +232,15 @@ public:
                        const VideoCommon::ImageViewInfo&, GPUVAddr);
     explicit ImageView(TextureCacheRuntime&, const VideoCommon::ImageInfo& info,
                        const VideoCommon::ImageViewInfo& view_info);
-    explicit ImageView(TextureCacheRuntime&, const VideoCommon::NullImageParams&);
+    explicit ImageView(TextureCacheRuntime&, const VideoCommon::NullImageViewParams&);
+
+    ~ImageView();
+
+    ImageView(const ImageView&) = delete;
+    ImageView& operator=(const ImageView&) = delete;
+
+    ImageView(ImageView&&) = default;
+    ImageView& operator=(ImageView&&) = default;
 
     [[nodiscard]] GLuint StorageView(Shader::TextureType texture_type,
                                      Shader::ImageFormat image_format);
@@ -276,6 +310,14 @@ public:
     explicit Framebuffer(TextureCacheRuntime&, std::span<ImageView*, NUM_RT> color_buffers,
                          ImageView* depth_buffer, const VideoCommon::RenderTargets& key);
 
+    ~Framebuffer();
+
+    Framebuffer(const Framebuffer&) = delete;
+    Framebuffer& operator=(const Framebuffer&) = delete;
+
+    Framebuffer(Framebuffer&&) = default;
+    Framebuffer& operator=(Framebuffer&&) = default;
+
     [[nodiscard]] GLuint Handle() const noexcept {
         return framebuffer.handle;
     }
@@ -293,7 +335,7 @@ struct TextureCacheParams {
     static constexpr bool ENABLE_VALIDATION = true;
     static constexpr bool FRAMEBUFFER_BLITS = true;
     static constexpr bool HAS_EMULATED_COPIES = true;
-    static constexpr bool HAS_DEVICE_MEMORY_INFO = false;
+    static constexpr bool HAS_DEVICE_MEMORY_INFO = true;
 
     using Runtime = OpenGL::TextureCacheRuntime;
     using Image = OpenGL::Image;
diff --git a/src/video_core/renderer_opengl/renderer_opengl.cpp b/src/video_core/renderer_opengl/renderer_opengl.cpp
index 7d7cba69c..28daacd82 100644
--- a/src/video_core/renderer_opengl/renderer_opengl.cpp
+++ b/src/video_core/renderer_opengl/renderer_opengl.cpp
@@ -21,8 +21,13 @@
 #include "core/memory.h"
 #include "core/perf_stats.h"
 #include "core/telemetry_session.h"
+#include "video_core/host_shaders/fxaa_frag.h"
+#include "video_core/host_shaders/fxaa_vert.h"
 #include "video_core/host_shaders/opengl_present_frag.h"
+#include "video_core/host_shaders/opengl_present_scaleforce_frag.h"
 #include "video_core/host_shaders/opengl_present_vert.h"
+#include "video_core/host_shaders/present_bicubic_frag.h"
+#include "video_core/host_shaders/present_gaussian_frag.h"
 #include "video_core/renderer_opengl/gl_rasterizer.h"
 #include "video_core/renderer_opengl/gl_shader_manager.h"
 #include "video_core/renderer_opengl/gl_shader_util.h"
@@ -208,7 +213,9 @@ void RendererOpenGL::LoadFBToScreenInfo(const Tegra::FramebufferConfig& framebuf
     framebuffer_crop_rect = framebuffer.crop_rect;
 
     const VAddr framebuffer_addr{framebuffer.address + framebuffer.offset};
-    if (rasterizer.AccelerateDisplay(framebuffer, framebuffer_addr, framebuffer.stride)) {
+    screen_info.was_accelerated =
+        rasterizer.AccelerateDisplay(framebuffer, framebuffer_addr, framebuffer.stride);
+    if (screen_info.was_accelerated) {
         return;
     }
 
@@ -251,12 +258,25 @@ void RendererOpenGL::LoadColorToActiveGLTexture(u8 color_r, u8 color_g, u8 color
 
 void RendererOpenGL::InitOpenGLObjects() {
     // Create shader programs
+    fxaa_vertex = CreateProgram(HostShaders::FXAA_VERT, GL_VERTEX_SHADER);
+    fxaa_fragment = CreateProgram(HostShaders::FXAA_FRAG, GL_FRAGMENT_SHADER);
     present_vertex = CreateProgram(HostShaders::OPENGL_PRESENT_VERT, GL_VERTEX_SHADER);
-    present_fragment = CreateProgram(HostShaders::OPENGL_PRESENT_FRAG, GL_FRAGMENT_SHADER);
+    present_bilinear_fragment = CreateProgram(HostShaders::OPENGL_PRESENT_FRAG, GL_FRAGMENT_SHADER);
+    present_bicubic_fragment = CreateProgram(HostShaders::PRESENT_BICUBIC_FRAG, GL_FRAGMENT_SHADER);
+    present_gaussian_fragment =
+        CreateProgram(HostShaders::PRESENT_GAUSSIAN_FRAG, GL_FRAGMENT_SHADER);
+    present_scaleforce_fragment =
+        CreateProgram(fmt::format("#version 460\n{}", HostShaders::OPENGL_PRESENT_SCALEFORCE_FRAG),
+                      GL_FRAGMENT_SHADER);
 
     // Generate presentation sampler
     present_sampler.Create();
     glSamplerParameteri(present_sampler.handle, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
+    glSamplerParameteri(present_sampler.handle, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
+
+    present_sampler_nn.Create();
+    glSamplerParameteri(present_sampler_nn.handle, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
+    glSamplerParameteri(present_sampler_nn.handle, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
 
     // Generate VBO handle for drawing
     vertex_buffer.Create();
@@ -274,6 +294,8 @@ void RendererOpenGL::InitOpenGLObjects() {
 
     // Clear screen to black
     LoadColorToActiveGLTexture(0, 0, 0, 0, screen_info.texture);
+
+    fxaa_framebuffer.Create();
 }
 
 void RendererOpenGL::AddTelemetryFields() {
@@ -325,18 +347,130 @@ void RendererOpenGL::ConfigureFramebufferTexture(TextureInfo& texture,
     texture.resource.Release();
     texture.resource.Create(GL_TEXTURE_2D);
     glTextureStorage2D(texture.resource.handle, 1, internal_format, texture.width, texture.height);
+    fxaa_texture.Release();
+    fxaa_texture.Create(GL_TEXTURE_2D);
+    glTextureStorage2D(fxaa_texture.handle, 1, GL_RGBA16F,
+                       Settings::values.resolution_info.ScaleUp(screen_info.texture.width),
+                       Settings::values.resolution_info.ScaleUp(screen_info.texture.height));
+    glNamedFramebufferTexture(fxaa_framebuffer.handle, GL_COLOR_ATTACHMENT0, fxaa_texture.handle,
+                              0);
 }
 
 void RendererOpenGL::DrawScreen(const Layout::FramebufferLayout& layout) {
+    // TODO: Signal state tracker about these changes
+    state_tracker.NotifyScreenDrawVertexArray();
+    state_tracker.NotifyPolygonModes();
+    state_tracker.NotifyViewport0();
+    state_tracker.NotifyScissor0();
+    state_tracker.NotifyColorMask(0);
+    state_tracker.NotifyBlend0();
+    state_tracker.NotifyFramebuffer();
+    state_tracker.NotifyFrontFace();
+    state_tracker.NotifyCullTest();
+    state_tracker.NotifyDepthTest();
+    state_tracker.NotifyStencilTest();
+    state_tracker.NotifyPolygonOffset();
+    state_tracker.NotifyRasterizeEnable();
+    state_tracker.NotifyFramebufferSRGB();
+    state_tracker.NotifyLogicOp();
+    state_tracker.NotifyClipControl();
+    state_tracker.NotifyAlphaTest();
+
+    state_tracker.ClipControl(GL_LOWER_LEFT, GL_ZERO_TO_ONE);
+
     // Update background color before drawing
     glClearColor(Settings::values.bg_red.GetValue() / 255.0f,
                  Settings::values.bg_green.GetValue() / 255.0f,
                  Settings::values.bg_blue.GetValue() / 255.0f, 1.0f);
 
+    glEnable(GL_CULL_FACE);
+    glDisable(GL_COLOR_LOGIC_OP);
+    glDisable(GL_DEPTH_TEST);
+    glDisable(GL_STENCIL_TEST);
+    glDisable(GL_POLYGON_OFFSET_FILL);
+    glDisable(GL_RASTERIZER_DISCARD);
+    glDisable(GL_ALPHA_TEST);
+    glDisablei(GL_BLEND, 0);
+    glPolygonMode(GL_FRONT_AND_BACK, GL_FILL);
+    glCullFace(GL_BACK);
+    glFrontFace(GL_CW);
+    glColorMaski(0, GL_TRUE, GL_TRUE, GL_TRUE, GL_TRUE);
+
+    glBindTextureUnit(0, screen_info.display_texture);
+
+    if (Settings::values.anti_aliasing.GetValue() == Settings::AntiAliasing::Fxaa) {
+        program_manager.BindPresentPrograms(fxaa_vertex.handle, fxaa_fragment.handle);
+
+        glEnablei(GL_SCISSOR_TEST, 0);
+        auto viewport_width = screen_info.texture.width;
+        auto scissor_width = framebuffer_crop_rect.GetWidth();
+        if (scissor_width <= 0) {
+            scissor_width = viewport_width;
+        }
+        auto viewport_height = screen_info.texture.height;
+        auto scissor_height = framebuffer_crop_rect.GetHeight();
+        if (scissor_height <= 0) {
+            scissor_height = viewport_height;
+        }
+        if (screen_info.was_accelerated) {
+            viewport_width = Settings::values.resolution_info.ScaleUp(viewport_width);
+            scissor_width = Settings::values.resolution_info.ScaleUp(scissor_width);
+            viewport_height = Settings::values.resolution_info.ScaleUp(viewport_height);
+            scissor_height = Settings::values.resolution_info.ScaleUp(scissor_height);
+        }
+        glScissorIndexed(0, 0, 0, scissor_width, scissor_height);
+        glViewportIndexedf(0, 0.0f, 0.0f, static_cast<GLfloat>(viewport_width),
+                           static_cast<GLfloat>(viewport_height));
+        glDepthRangeIndexed(0, 0.0, 0.0);
+
+        glBindSampler(0, present_sampler.handle);
+        GLint old_read_fb;
+        GLint old_draw_fb;
+        glGetIntegerv(GL_READ_FRAMEBUFFER_BINDING, &old_read_fb);
+        glGetIntegerv(GL_DRAW_FRAMEBUFFER_BINDING, &old_draw_fb);
+        glBindFramebuffer(GL_DRAW_FRAMEBUFFER, fxaa_framebuffer.handle);
+
+        glDrawArrays(GL_TRIANGLE_STRIP, 0, 4);
+
+        glBindFramebuffer(GL_READ_FRAMEBUFFER, old_read_fb);
+        glBindFramebuffer(GL_DRAW_FRAMEBUFFER, old_draw_fb);
+
+        glBindTextureUnit(0, fxaa_texture.handle);
+    }
+
     // Set projection matrix
     const std::array ortho_matrix =
         MakeOrthographicMatrix(static_cast<float>(layout.width), static_cast<float>(layout.height));
-    program_manager.BindPresentPrograms(present_vertex.handle, present_fragment.handle);
+
+    GLuint fragment_handle;
+    const auto filter = Settings::values.scaling_filter.GetValue();
+    switch (filter) {
+    case Settings::ScalingFilter::NearestNeighbor:
+        fragment_handle = present_bilinear_fragment.handle;
+        break;
+    case Settings::ScalingFilter::Bilinear:
+        fragment_handle = present_bilinear_fragment.handle;
+        break;
+    case Settings::ScalingFilter::Bicubic:
+        fragment_handle = present_bicubic_fragment.handle;
+        break;
+    case Settings::ScalingFilter::Gaussian:
+        fragment_handle = present_gaussian_fragment.handle;
+        break;
+    case Settings::ScalingFilter::ScaleForce:
+        fragment_handle = present_scaleforce_fragment.handle;
+        break;
+    case Settings::ScalingFilter::Fsr:
+        LOG_WARNING(
+            Render_OpenGL,
+            "FidelityFX FSR Super Sampling is not supported in OpenGL, changing to ScaleForce");
+        fragment_handle = present_scaleforce_fragment.handle;
+        break;
+    default:
+        fragment_handle = present_bilinear_fragment.handle;
+        break;
+    }
+    program_manager.BindPresentPrograms(present_vertex.handle, fragment_handle);
     glProgramUniformMatrix3x2fv(present_vertex.handle, ModelViewMatrixLocation, 1, GL_FALSE,
                                 ortho_matrix.data());
 
@@ -370,6 +504,11 @@ void RendererOpenGL::DrawScreen(const Layout::FramebufferLayout& layout) {
         scale_v = static_cast<f32>(framebuffer_crop_rect.GetHeight()) /
                   static_cast<f32>(screen_info.texture.height);
     }
+    if (Settings::values.anti_aliasing.GetValue() == Settings::AntiAliasing::Fxaa &&
+        !screen_info.was_accelerated) {
+        scale_u /= Settings::values.resolution_info.up_factor;
+        scale_v /= Settings::values.resolution_info.up_factor;
+    }
 
     const auto& screen = layout.screen;
     const std::array vertices = {
@@ -380,47 +519,14 @@ void RendererOpenGL::DrawScreen(const Layout::FramebufferLayout& layout) {
     };
     glNamedBufferSubData(vertex_buffer.handle, 0, sizeof(vertices), std::data(vertices));
 
-    // TODO: Signal state tracker about these changes
-    state_tracker.NotifyScreenDrawVertexArray();
-    state_tracker.NotifyPolygonModes();
-    state_tracker.NotifyViewport0();
-    state_tracker.NotifyScissor0();
-    state_tracker.NotifyColorMask(0);
-    state_tracker.NotifyBlend0();
-    state_tracker.NotifyFramebuffer();
-    state_tracker.NotifyFrontFace();
-    state_tracker.NotifyCullTest();
-    state_tracker.NotifyDepthTest();
-    state_tracker.NotifyStencilTest();
-    state_tracker.NotifyPolygonOffset();
-    state_tracker.NotifyRasterizeEnable();
-    state_tracker.NotifyFramebufferSRGB();
-    state_tracker.NotifyLogicOp();
-    state_tracker.NotifyClipControl();
-    state_tracker.NotifyAlphaTest();
-
-    state_tracker.ClipControl(GL_LOWER_LEFT, GL_ZERO_TO_ONE);
-    glEnable(GL_CULL_FACE);
     if (screen_info.display_srgb) {
         glEnable(GL_FRAMEBUFFER_SRGB);
     } else {
         glDisable(GL_FRAMEBUFFER_SRGB);
     }
-    glDisable(GL_COLOR_LOGIC_OP);
-    glDisable(GL_DEPTH_TEST);
-    glDisable(GL_STENCIL_TEST);
-    glDisable(GL_POLYGON_OFFSET_FILL);
-    glDisable(GL_RASTERIZER_DISCARD);
-    glDisable(GL_ALPHA_TEST);
-    glDisablei(GL_BLEND, 0);
     glDisablei(GL_SCISSOR_TEST, 0);
-    glPolygonMode(GL_FRONT_AND_BACK, GL_FILL);
-    glCullFace(GL_BACK);
-    glFrontFace(GL_CW);
-    glColorMaski(0, GL_TRUE, GL_TRUE, GL_TRUE, GL_TRUE);
     glViewportIndexedf(0, 0.0f, 0.0f, static_cast<GLfloat>(layout.width),
                        static_cast<GLfloat>(layout.height));
-    glDepthRangeIndexed(0, 0.0, 0.0);
 
     glEnableVertexAttribArray(PositionLocation);
     glEnableVertexAttribArray(TexCoordLocation);
@@ -440,8 +546,11 @@ void RendererOpenGL::DrawScreen(const Layout::FramebufferLayout& layout) {
         glBindVertexBuffer(0, vertex_buffer.handle, 0, sizeof(ScreenRectVertex));
     }
 
-    glBindTextureUnit(0, screen_info.display_texture);
-    glBindSampler(0, present_sampler.handle);
+    if (Settings::values.scaling_filter.GetValue() != Settings::ScalingFilter::NearestNeighbor) {
+        glBindSampler(0, present_sampler.handle);
+    } else {
+        glBindSampler(0, present_sampler_nn.handle);
+    }
 
     glClear(GL_COLOR_BUFFER_BIT);
     glDrawArrays(GL_TRIANGLE_STRIP, 0, 4);
diff --git a/src/video_core/renderer_opengl/renderer_opengl.h b/src/video_core/renderer_opengl/renderer_opengl.h
index d455f572f..cda333cad 100644
--- a/src/video_core/renderer_opengl/renderer_opengl.h
+++ b/src/video_core/renderer_opengl/renderer_opengl.h
@@ -50,6 +50,7 @@ struct TextureInfo {
 /// Structure used for storing information about the display target for the Switch screen
 struct ScreenInfo {
     GLuint display_texture{};
+    bool was_accelerated = false;
     bool display_srgb{};
     const Common::Rectangle<float> display_texcoords{0.0f, 0.0f, 1.0f, 1.0f};
     TextureInfo texture;
@@ -109,9 +110,15 @@ private:
 
     // OpenGL object IDs
     OGLSampler present_sampler;
+    OGLSampler present_sampler_nn;
     OGLBuffer vertex_buffer;
+    OGLProgram fxaa_vertex;
+    OGLProgram fxaa_fragment;
     OGLProgram present_vertex;
-    OGLProgram present_fragment;
+    OGLProgram present_bilinear_fragment;
+    OGLProgram present_bicubic_fragment;
+    OGLProgram present_gaussian_fragment;
+    OGLProgram present_scaleforce_fragment;
     OGLFramebuffer screenshot_framebuffer;
 
     // GPU address of the vertex buffer
@@ -119,6 +126,8 @@ private:
 
     /// Display information for Switch screen
     ScreenInfo screen_info;
+    OGLTexture fxaa_texture;
+    OGLFramebuffer fxaa_framebuffer;
 
     /// OpenGL framebuffer data
     std::vector<u8> gl_framebuffer_data;
diff --git a/src/video_core/renderer_vulkan/blit_image.cpp b/src/video_core/renderer_vulkan/blit_image.cpp
index 6c1b2f063..b3884a4f5 100644
--- a/src/video_core/renderer_vulkan/blit_image.cpp
+++ b/src/video_core/renderer_vulkan/blit_image.cpp
@@ -363,7 +363,7 @@ BlitImageHelper::BlitImageHelper(const Device& device_, VKScheduler& scheduler_,
 
 BlitImageHelper::~BlitImageHelper() = default;
 
-void BlitImageHelper::BlitColor(const Framebuffer* dst_framebuffer, const ImageView& src_image_view,
+void BlitImageHelper::BlitColor(const Framebuffer* dst_framebuffer, VkImageView src_view,
                                 const Region2D& dst_region, const Region2D& src_region,
                                 Tegra::Engines::Fermi2D::Filter filter,
                                 Tegra::Engines::Fermi2D::Operation operation) {
@@ -373,9 +373,8 @@ void BlitImageHelper::BlitColor(const Framebuffer* dst_framebuffer, const ImageV
         .operation = operation,
     };
     const VkPipelineLayout layout = *one_texture_pipeline_layout;
-    const VkImageView src_view = src_image_view.Handle(Shader::TextureType::Color2D);
     const VkSampler sampler = is_linear ? *linear_sampler : *nearest_sampler;
-    const VkPipeline pipeline = FindOrEmplacePipeline(key);
+    const VkPipeline pipeline = FindOrEmplaceColorPipeline(key);
     scheduler.RequestRenderpass(dst_framebuffer);
     scheduler.Record([this, dst_region, src_region, pipeline, layout, sampler,
                       src_view](vk::CommandBuffer cmdbuf) {
@@ -398,10 +397,13 @@ void BlitImageHelper::BlitDepthStencil(const Framebuffer* dst_framebuffer,
                                        Tegra::Engines::Fermi2D::Operation operation) {
     ASSERT(filter == Tegra::Engines::Fermi2D::Filter::Point);
     ASSERT(operation == Tegra::Engines::Fermi2D::Operation::SrcCopy);
-
+    const BlitImagePipelineKey key{
+        .renderpass = dst_framebuffer->RenderPass(),
+        .operation = operation,
+    };
     const VkPipelineLayout layout = *two_textures_pipeline_layout;
     const VkSampler sampler = *nearest_sampler;
-    const VkPipeline pipeline = BlitDepthStencilPipeline(dst_framebuffer->RenderPass());
+    const VkPipeline pipeline = FindOrEmplaceDepthStencilPipeline(key);
     scheduler.RequestRenderpass(dst_framebuffer);
     scheduler.Record([dst_region, src_region, pipeline, layout, sampler, src_depth_view,
                       src_stencil_view, this](vk::CommandBuffer cmdbuf) {
@@ -419,40 +421,45 @@ void BlitImageHelper::BlitDepthStencil(const Framebuffer* dst_framebuffer,
 }
 
 void BlitImageHelper::ConvertD32ToR32(const Framebuffer* dst_framebuffer,
-                                      const ImageView& src_image_view) {
+                                      const ImageView& src_image_view, u32 up_scale,
+                                      u32 down_shift) {
     ConvertDepthToColorPipeline(convert_d32_to_r32_pipeline, dst_framebuffer->RenderPass());
-    Convert(*convert_d32_to_r32_pipeline, dst_framebuffer, src_image_view);
+    Convert(*convert_d32_to_r32_pipeline, dst_framebuffer, src_image_view, up_scale, down_shift);
 }
 
 void BlitImageHelper::ConvertR32ToD32(const Framebuffer* dst_framebuffer,
-                                      const ImageView& src_image_view) {
+                                      const ImageView& src_image_view, u32 up_scale,
+                                      u32 down_shift) {
     ConvertColorToDepthPipeline(convert_r32_to_d32_pipeline, dst_framebuffer->RenderPass());
-    Convert(*convert_r32_to_d32_pipeline, dst_framebuffer, src_image_view);
+    Convert(*convert_r32_to_d32_pipeline, dst_framebuffer, src_image_view, up_scale, down_shift);
 }
 
 void BlitImageHelper::ConvertD16ToR16(const Framebuffer* dst_framebuffer,
-                                      const ImageView& src_image_view) {
+                                      const ImageView& src_image_view, u32 up_scale,
+                                      u32 down_shift) {
     ConvertDepthToColorPipeline(convert_d16_to_r16_pipeline, dst_framebuffer->RenderPass());
-    Convert(*convert_d16_to_r16_pipeline, dst_framebuffer, src_image_view);
+    Convert(*convert_d16_to_r16_pipeline, dst_framebuffer, src_image_view, up_scale, down_shift);
 }
 
 void BlitImageHelper::ConvertR16ToD16(const Framebuffer* dst_framebuffer,
-                                      const ImageView& src_image_view) {
+                                      const ImageView& src_image_view, u32 up_scale,
+                                      u32 down_shift) {
     ConvertColorToDepthPipeline(convert_r16_to_d16_pipeline, dst_framebuffer->RenderPass());
-    Convert(*convert_r16_to_d16_pipeline, dst_framebuffer, src_image_view);
+    Convert(*convert_r16_to_d16_pipeline, dst_framebuffer, src_image_view, up_scale, down_shift);
 }
 
 void BlitImageHelper::Convert(VkPipeline pipeline, const Framebuffer* dst_framebuffer,
-                              const ImageView& src_image_view) {
+                              const ImageView& src_image_view, u32 up_scale, u32 down_shift) {
     const VkPipelineLayout layout = *one_texture_pipeline_layout;
     const VkImageView src_view = src_image_view.Handle(Shader::TextureType::Color2D);
     const VkSampler sampler = *nearest_sampler;
     const VkExtent2D extent{
-        .width = src_image_view.size.width,
-        .height = src_image_view.size.height,
+        .width = std::max((src_image_view.size.width * up_scale) >> down_shift, 1U),
+        .height = std::max((src_image_view.size.height * up_scale) >> down_shift, 1U),
     };
     scheduler.RequestRenderpass(dst_framebuffer);
-    scheduler.Record([pipeline, layout, sampler, src_view, extent, this](vk::CommandBuffer cmdbuf) {
+    scheduler.Record([pipeline, layout, sampler, src_view, extent, up_scale, down_shift,
+                      this](vk::CommandBuffer cmdbuf) {
         const VkOffset2D offset{
             .x = 0,
             .y = 0,
@@ -488,7 +495,7 @@ void BlitImageHelper::Convert(VkPipeline pipeline, const Framebuffer* dst_frameb
     scheduler.InvalidateState();
 }
 
-VkPipeline BlitImageHelper::FindOrEmplacePipeline(const BlitImagePipelineKey& key) {
+VkPipeline BlitImageHelper::FindOrEmplaceColorPipeline(const BlitImagePipelineKey& key) {
     const auto it = std::ranges::find(blit_color_keys, key);
     if (it != blit_color_keys.end()) {
         return *blit_color_pipelines[std::distance(blit_color_keys.begin(), it)];
@@ -542,12 +549,14 @@ VkPipeline BlitImageHelper::FindOrEmplacePipeline(const BlitImagePipelineKey& ke
     return *blit_color_pipelines.back();
 }
 
-VkPipeline BlitImageHelper::BlitDepthStencilPipeline(VkRenderPass renderpass) {
-    if (blit_depth_stencil_pipeline) {
-        return *blit_depth_stencil_pipeline;
+VkPipeline BlitImageHelper::FindOrEmplaceDepthStencilPipeline(const BlitImagePipelineKey& key) {
+    const auto it = std::ranges::find(blit_depth_stencil_keys, key);
+    if (it != blit_depth_stencil_keys.end()) {
+        return *blit_depth_stencil_pipelines[std::distance(blit_depth_stencil_keys.begin(), it)];
     }
+    blit_depth_stencil_keys.push_back(key);
     const std::array stages = MakeStages(*full_screen_vert, *blit_depth_stencil_frag);
-    blit_depth_stencil_pipeline = device.GetLogical().CreateGraphicsPipeline({
+    blit_depth_stencil_pipelines.push_back(device.GetLogical().CreateGraphicsPipeline({
         .sType = VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO,
         .pNext = nullptr,
         .flags = 0,
@@ -560,15 +569,15 @@ VkPipeline BlitImageHelper::BlitDepthStencilPipeline(VkRenderPass renderpass) {
         .pRasterizationState = &PIPELINE_RASTERIZATION_STATE_CREATE_INFO,
         .pMultisampleState = &PIPELINE_MULTISAMPLE_STATE_CREATE_INFO,
         .pDepthStencilState = &PIPELINE_DEPTH_STENCIL_STATE_CREATE_INFO,
-        .pColorBlendState = &PIPELINE_COLOR_BLEND_STATE_EMPTY_CREATE_INFO,
+        .pColorBlendState = &PIPELINE_COLOR_BLEND_STATE_GENERIC_CREATE_INFO,
         .pDynamicState = &PIPELINE_DYNAMIC_STATE_CREATE_INFO,
         .layout = *two_textures_pipeline_layout,
-        .renderPass = renderpass,
+        .renderPass = key.renderpass,
         .subpass = 0,
         .basePipelineHandle = VK_NULL_HANDLE,
         .basePipelineIndex = 0,
-    });
-    return *blit_depth_stencil_pipeline;
+    }));
+    return *blit_depth_stencil_pipelines.back();
 }
 
 void BlitImageHelper::ConvertDepthToColorPipeline(vk::Pipeline& pipeline, VkRenderPass renderpass) {
diff --git a/src/video_core/renderer_vulkan/blit_image.h b/src/video_core/renderer_vulkan/blit_image.h
index 33ee095c1..d77f76678 100644
--- a/src/video_core/renderer_vulkan/blit_image.h
+++ b/src/video_core/renderer_vulkan/blit_image.h
@@ -34,7 +34,7 @@ public:
                              StateTracker& state_tracker, DescriptorPool& descriptor_pool);
     ~BlitImageHelper();
 
-    void BlitColor(const Framebuffer* dst_framebuffer, const ImageView& src_image_view,
+    void BlitColor(const Framebuffer* dst_framebuffer, VkImageView src_image_view,
                    const Region2D& dst_region, const Region2D& src_region,
                    Tegra::Engines::Fermi2D::Filter filter,
                    Tegra::Engines::Fermi2D::Operation operation);
@@ -44,21 +44,25 @@ public:
                           const Region2D& src_region, Tegra::Engines::Fermi2D::Filter filter,
                           Tegra::Engines::Fermi2D::Operation operation);
 
-    void ConvertD32ToR32(const Framebuffer* dst_framebuffer, const ImageView& src_image_view);
+    void ConvertD32ToR32(const Framebuffer* dst_framebuffer, const ImageView& src_image_view,
+                         u32 up_scale, u32 down_shift);
 
-    void ConvertR32ToD32(const Framebuffer* dst_framebuffer, const ImageView& src_image_view);
+    void ConvertR32ToD32(const Framebuffer* dst_framebuffer, const ImageView& src_image_view,
+                         u32 up_scale, u32 down_shift);
 
-    void ConvertD16ToR16(const Framebuffer* dst_framebuffer, const ImageView& src_image_view);
+    void ConvertD16ToR16(const Framebuffer* dst_framebuffer, const ImageView& src_image_view,
+                         u32 up_scale, u32 down_shift);
 
-    void ConvertR16ToD16(const Framebuffer* dst_framebuffer, const ImageView& src_image_view);
+    void ConvertR16ToD16(const Framebuffer* dst_framebuffer, const ImageView& src_image_view,
+                         u32 up_scale, u32 down_shift);
 
 private:
     void Convert(VkPipeline pipeline, const Framebuffer* dst_framebuffer,
-                 const ImageView& src_image_view);
+                 const ImageView& src_image_view, u32 up_scale, u32 down_shift);
 
-    [[nodiscard]] VkPipeline FindOrEmplacePipeline(const BlitImagePipelineKey& key);
+    [[nodiscard]] VkPipeline FindOrEmplaceColorPipeline(const BlitImagePipelineKey& key);
 
-    [[nodiscard]] VkPipeline BlitDepthStencilPipeline(VkRenderPass renderpass);
+    [[nodiscard]] VkPipeline FindOrEmplaceDepthStencilPipeline(const BlitImagePipelineKey& key);
 
     void ConvertDepthToColorPipeline(vk::Pipeline& pipeline, VkRenderPass renderpass);
 
@@ -84,7 +88,8 @@ private:
 
     std::vector<BlitImagePipelineKey> blit_color_keys;
     std::vector<vk::Pipeline> blit_color_pipelines;
-    vk::Pipeline blit_depth_stencil_pipeline;
+    std::vector<BlitImagePipelineKey> blit_depth_stencil_keys;
+    std::vector<vk::Pipeline> blit_depth_stencil_pipelines;
     vk::Pipeline convert_d32_to_r32_pipeline;
     vk::Pipeline convert_r32_to_d32_pipeline;
     vk::Pipeline convert_d16_to_r16_pipeline;
diff --git a/src/video_core/renderer_vulkan/pipeline_helper.h b/src/video_core/renderer_vulkan/pipeline_helper.h
index 4847db6b6..11c160570 100644
--- a/src/video_core/renderer_vulkan/pipeline_helper.h
+++ b/src/video_core/renderer_vulkan/pipeline_helper.h
@@ -10,6 +10,7 @@
 
 #include "common/assert.h"
 #include "common/common_types.h"
+#include "shader_recompiler/backend/spirv/emit_spirv.h"
 #include "shader_recompiler/shader_info.h"
 #include "video_core/renderer_vulkan/vk_texture_cache.h"
 #include "video_core/renderer_vulkan/vk_update_descriptor.h"
@@ -20,6 +21,8 @@
 
 namespace Vulkan {
 
+using Shader::Backend::SPIRV::NUM_TEXTURE_AND_IMAGE_SCALING_WORDS;
+
 class DescriptorLayoutBuilder {
 public:
     DescriptorLayoutBuilder(const Device& device_) : device{&device_} {}
@@ -68,18 +71,28 @@ public:
     }
 
     vk::PipelineLayout CreatePipelineLayout(VkDescriptorSetLayout descriptor_set_layout) const {
+        using Shader::Backend::SPIRV::RescalingLayout;
+        const u32 size_offset = is_compute ? sizeof(RescalingLayout::down_factor) : 0u;
+        const VkPushConstantRange range{
+            .stageFlags = static_cast<VkShaderStageFlags>(
+                is_compute ? VK_SHADER_STAGE_COMPUTE_BIT : VK_SHADER_STAGE_ALL_GRAPHICS),
+            .offset = 0,
+            .size = static_cast<u32>(sizeof(RescalingLayout)) - size_offset,
+        };
         return device->GetLogical().CreatePipelineLayout({
             .sType = VK_STRUCTURE_TYPE_PIPELINE_LAYOUT_CREATE_INFO,
             .pNext = nullptr,
             .flags = 0,
             .setLayoutCount = descriptor_set_layout ? 1U : 0U,
             .pSetLayouts = bindings.empty() ? nullptr : &descriptor_set_layout,
-            .pushConstantRangeCount = 0,
-            .pPushConstantRanges = nullptr,
+            .pushConstantRangeCount = 1,
+            .pPushConstantRanges = &range,
         });
     }
 
     void Add(const Shader::Info& info, VkShaderStageFlags stage) {
+        is_compute |= (stage & VK_SHADER_STAGE_COMPUTE_BIT) != 0;
+
         Add(VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER, stage, info.constant_buffer_descriptors);
         Add(VK_DESCRIPTOR_TYPE_STORAGE_BUFFER, stage, info.storage_buffers_descriptors);
         Add(VK_DESCRIPTOR_TYPE_UNIFORM_TEXEL_BUFFER, stage, info.texture_buffer_descriptors);
@@ -115,6 +128,7 @@ private:
     }
 
     const Device* device{};
+    bool is_compute{};
     boost::container::small_vector<VkDescriptorSetLayoutBinding, 32> bindings;
     boost::container::small_vector<VkDescriptorUpdateTemplateEntryKHR, 32> entries;
     u32 binding{};
@@ -122,31 +136,68 @@ private:
     size_t offset{};
 };
 
-inline void PushImageDescriptors(const Shader::Info& info, const VkSampler*& samplers,
-                                 const ImageId*& image_view_ids, TextureCache& texture_cache,
-                                 VKUpdateDescriptorQueue& update_descriptor_queue) {
-    for (const auto& desc : info.texture_buffer_descriptors) {
-        image_view_ids += desc.count;
+class RescalingPushConstant {
+public:
+    explicit RescalingPushConstant() noexcept {}
+
+    void PushTexture(bool is_rescaled) noexcept {
+        *texture_ptr |= is_rescaled ? texture_bit : 0u;
+        texture_bit <<= 1u;
+        if (texture_bit == 0u) {
+            texture_bit = 1u;
+            ++texture_ptr;
+        }
     }
-    for (const auto& desc : info.image_buffer_descriptors) {
-        image_view_ids += desc.count;
+
+    void PushImage(bool is_rescaled) noexcept {
+        *image_ptr |= is_rescaled ? image_bit : 0u;
+        image_bit <<= 1u;
+        if (image_bit == 0u) {
+            image_bit = 1u;
+            ++image_ptr;
+        }
     }
+
+    const std::array<u32, NUM_TEXTURE_AND_IMAGE_SCALING_WORDS>& Data() const noexcept {
+        return words;
+    }
+
+private:
+    std::array<u32, NUM_TEXTURE_AND_IMAGE_SCALING_WORDS> words{};
+    u32* texture_ptr{words.data()};
+    u32* image_ptr{words.data() + Shader::Backend::SPIRV::NUM_TEXTURE_SCALING_WORDS};
+    u32 texture_bit{1u};
+    u32 image_bit{1u};
+};
+
+inline void PushImageDescriptors(TextureCache& texture_cache,
+                                 VKUpdateDescriptorQueue& update_descriptor_queue,
+                                 const Shader::Info& info, RescalingPushConstant& rescaling,
+                                 const VkSampler*& samplers,
+                                 const VideoCommon::ImageViewInOut*& views) {
+    const u32 num_texture_buffers = Shader::NumDescriptors(info.texture_buffer_descriptors);
+    const u32 num_image_buffers = Shader::NumDescriptors(info.image_buffer_descriptors);
+    views += num_texture_buffers;
+    views += num_image_buffers;
     for (const auto& desc : info.texture_descriptors) {
         for (u32 index = 0; index < desc.count; ++index) {
+            const VideoCommon::ImageViewId image_view_id{(views++)->id};
             const VkSampler sampler{*(samplers++)};
-            ImageView& image_view{texture_cache.GetImageView(*(image_view_ids++))};
+            ImageView& image_view{texture_cache.GetImageView(image_view_id)};
             const VkImageView vk_image_view{image_view.Handle(desc.type)};
             update_descriptor_queue.AddSampledImage(vk_image_view, sampler);
+            rescaling.PushTexture(texture_cache.IsRescaling(image_view));
         }
     }
     for (const auto& desc : info.image_descriptors) {
         for (u32 index = 0; index < desc.count; ++index) {
-            ImageView& image_view{texture_cache.GetImageView(*(image_view_ids++))};
+            ImageView& image_view{texture_cache.GetImageView((views++)->id)};
             if (desc.is_written) {
                 texture_cache.MarkModification(image_view.image_id);
             }
             const VkImageView vk_image_view{image_view.StorageView(desc.type, desc.format)};
             update_descriptor_queue.AddImage(vk_image_view);
+            rescaling.PushImage(texture_cache.IsRescaling(image_view));
         }
     }
 }
diff --git a/src/video_core/renderer_vulkan/vk_blit_screen.cpp b/src/video_core/renderer_vulkan/vk_blit_screen.cpp
index 888bc7392..1e447e621 100644
--- a/src/video_core/renderer_vulkan/vk_blit_screen.cpp
+++ b/src/video_core/renderer_vulkan/vk_blit_screen.cpp
@@ -12,14 +12,22 @@
 #include "common/assert.h"
 #include "common/common_types.h"
 #include "common/math_util.h"
+#include "common/settings.h"
 #include "core/core.h"
 #include "core/frontend/emu_window.h"
 #include "core/memory.h"
 #include "video_core/gpu.h"
+#include "video_core/host_shaders/fxaa_frag_spv.h"
+#include "video_core/host_shaders/fxaa_vert_spv.h"
+#include "video_core/host_shaders/present_bicubic_frag_spv.h"
+#include "video_core/host_shaders/present_gaussian_frag_spv.h"
 #include "video_core/host_shaders/vulkan_present_frag_spv.h"
+#include "video_core/host_shaders/vulkan_present_scaleforce_fp16_frag_spv.h"
+#include "video_core/host_shaders/vulkan_present_scaleforce_fp32_frag_spv.h"
 #include "video_core/host_shaders/vulkan_present_vert_spv.h"
 #include "video_core/renderer_vulkan/renderer_vulkan.h"
 #include "video_core/renderer_vulkan/vk_blit_screen.h"
+#include "video_core/renderer_vulkan/vk_fsr.h"
 #include "video_core/renderer_vulkan/vk_master_semaphore.h"
 #include "video_core/renderer_vulkan/vk_scheduler.h"
 #include "video_core/renderer_vulkan/vk_shader_util.h"
@@ -144,8 +152,8 @@ VkSemaphore VKBlitScreen::Draw(const Tegra::FramebufferConfig& framebuffer,
     scheduler.Wait(resource_ticks[image_index]);
     resource_ticks[image_index] = scheduler.CurrentTick();
 
-    UpdateDescriptorSet(image_index,
-                        use_accelerated ? screen_info.image_view : *raw_image_views[image_index]);
+    VkImageView source_image_view =
+        use_accelerated ? screen_info.image_view : *raw_image_views[image_index];
 
     BufferData data;
     SetUniformData(data, layout);
@@ -222,9 +230,134 @@ VkSemaphore VKBlitScreen::Draw(const Tegra::FramebufferConfig& framebuffer,
                                    read_barrier);
             cmdbuf.CopyBufferToImage(*buffer, image, VK_IMAGE_LAYOUT_GENERAL, copy);
             cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_TRANSFER_BIT,
-                                   VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, 0, write_barrier);
+                                   VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT |
+                                       VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT,
+                                   0, write_barrier);
         });
     }
+
+    const auto anti_alias_pass = Settings::values.anti_aliasing.GetValue();
+    if (use_accelerated && anti_alias_pass != Settings::AntiAliasing::None) {
+        UpdateAADescriptorSet(image_index, source_image_view, false);
+        const u32 up_scale = Settings::values.resolution_info.up_scale;
+        const u32 down_shift = Settings::values.resolution_info.down_shift;
+        VkExtent2D size{
+            .width = (up_scale * framebuffer.width) >> down_shift,
+            .height = (up_scale * framebuffer.height) >> down_shift,
+        };
+        scheduler.Record([this, image_index, size, anti_alias_pass](vk::CommandBuffer cmdbuf) {
+            const VkImageMemoryBarrier base_barrier{
+                .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER,
+                .pNext = nullptr,
+                .srcAccessMask = 0,
+                .dstAccessMask = 0,
+                .oldLayout = VK_IMAGE_LAYOUT_GENERAL,
+                .newLayout = VK_IMAGE_LAYOUT_GENERAL,
+                .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .image = {},
+                .subresourceRange =
+                    {
+                        .aspectMask = VK_IMAGE_ASPECT_COLOR_BIT,
+                        .baseMipLevel = 0,
+                        .levelCount = 1,
+                        .baseArrayLayer = 0,
+                        .layerCount = 1,
+                    },
+            };
+
+            {
+                VkImageMemoryBarrier fsr_write_barrier = base_barrier;
+                fsr_write_barrier.image = *aa_image;
+                fsr_write_barrier.oldLayout = VK_IMAGE_LAYOUT_UNDEFINED;
+                cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_ALL_COMMANDS_BIT,
+                                       VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, 0, fsr_write_barrier);
+            }
+
+            const f32 bg_red = Settings::values.bg_red.GetValue() / 255.0f;
+            const f32 bg_green = Settings::values.bg_green.GetValue() / 255.0f;
+            const f32 bg_blue = Settings::values.bg_blue.GetValue() / 255.0f;
+            const VkClearValue clear_color{
+                .color = {.float32 = {bg_red, bg_green, bg_blue, 1.0f}},
+            };
+            const VkRenderPassBeginInfo renderpass_bi{
+                .sType = VK_STRUCTURE_TYPE_RENDER_PASS_BEGIN_INFO,
+                .pNext = nullptr,
+                .renderPass = *aa_renderpass,
+                .framebuffer = *aa_framebuffer,
+                .renderArea =
+                    {
+                        .offset = {0, 0},
+                        .extent = size,
+                    },
+                .clearValueCount = 1,
+                .pClearValues = &clear_color,
+            };
+            const VkViewport viewport{
+                .x = 0.0f,
+                .y = 0.0f,
+                .width = static_cast<float>(size.width),
+                .height = static_cast<float>(size.height),
+                .minDepth = 0.0f,
+                .maxDepth = 1.0f,
+            };
+            const VkRect2D scissor{
+                .offset = {0, 0},
+                .extent = size,
+            };
+            cmdbuf.BeginRenderPass(renderpass_bi, VK_SUBPASS_CONTENTS_INLINE);
+            switch (anti_alias_pass) {
+            case Settings::AntiAliasing::Fxaa:
+                cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *aa_pipeline);
+                break;
+            default:
+                cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *aa_pipeline);
+                break;
+            }
+            cmdbuf.SetViewport(0, viewport);
+            cmdbuf.SetScissor(0, scissor);
+
+            cmdbuf.BindVertexBuffer(0, *buffer, offsetof(BufferData, vertices));
+            cmdbuf.BindDescriptorSets(VK_PIPELINE_BIND_POINT_GRAPHICS, *aa_pipeline_layout, 0,
+                                      aa_descriptor_sets[image_index], {});
+            cmdbuf.Draw(4, 1, 0, 0);
+            cmdbuf.EndRenderPass();
+
+            {
+                VkImageMemoryBarrier blit_read_barrier = base_barrier;
+                blit_read_barrier.image = *aa_image;
+                blit_read_barrier.srcAccessMask = VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT;
+                blit_read_barrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT;
+
+                cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT,
+                                       VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, 0, blit_read_barrier);
+            }
+        });
+        source_image_view = *aa_image_view;
+    }
+
+    if (fsr) {
+        auto crop_rect = framebuffer.crop_rect;
+        if (crop_rect.GetWidth() == 0) {
+            crop_rect.right = framebuffer.width;
+        }
+        if (crop_rect.GetHeight() == 0) {
+            crop_rect.bottom = framebuffer.height;
+        }
+        crop_rect = crop_rect.Scale(Settings::values.resolution_info.up_factor);
+        VkExtent2D fsr_input_size{
+            .width = Settings::values.resolution_info.ScaleUp(framebuffer.width),
+            .height = Settings::values.resolution_info.ScaleUp(framebuffer.height),
+        };
+        VkImageView fsr_image_view =
+            fsr->Draw(scheduler, image_index, source_image_view, fsr_input_size, crop_rect);
+        UpdateDescriptorSet(image_index, fsr_image_view, true);
+    } else {
+        const bool is_nn =
+            Settings::values.scaling_filter.GetValue() == Settings::ScalingFilter::NearestNeighbor;
+        UpdateDescriptorSet(image_index, source_image_view, is_nn);
+    }
+
     scheduler.Record(
         [this, host_framebuffer, image_index, size = render_area](vk::CommandBuffer cmdbuf) {
             const f32 bg_red = Settings::values.bg_red.GetValue() / 255.0f;
@@ -258,8 +391,28 @@ VkSemaphore VKBlitScreen::Draw(const Tegra::FramebufferConfig& framebuffer,
                 .offset = {0, 0},
                 .extent = size,
             };
+            const auto filter = Settings::values.scaling_filter.GetValue();
             cmdbuf.BeginRenderPass(renderpass_bi, VK_SUBPASS_CONTENTS_INLINE);
-            cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *pipeline);
+            switch (filter) {
+            case Settings::ScalingFilter::NearestNeighbor:
+                cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *bilinear_pipeline);
+                break;
+            case Settings::ScalingFilter::Bilinear:
+                cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *bilinear_pipeline);
+                break;
+            case Settings::ScalingFilter::Bicubic:
+                cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *bicubic_pipeline);
+                break;
+            case Settings::ScalingFilter::Gaussian:
+                cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *gaussian_pipeline);
+                break;
+            case Settings::ScalingFilter::ScaleForce:
+                cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *scaleforce_pipeline);
+                break;
+            default:
+                cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *bilinear_pipeline);
+                break;
+            }
             cmdbuf.SetViewport(0, viewport);
             cmdbuf.SetScissor(0, scissor);
 
@@ -281,11 +434,16 @@ VkSemaphore VKBlitScreen::DrawToSwapchain(const Tegra::FramebufferConfig& frameb
 }
 
 vk::Framebuffer VKBlitScreen::CreateFramebuffer(const VkImageView& image_view, VkExtent2D extent) {
+    return CreateFramebuffer(image_view, extent, renderpass);
+}
+
+vk::Framebuffer VKBlitScreen::CreateFramebuffer(const VkImageView& image_view, VkExtent2D extent,
+                                                vk::RenderPass& rd) {
     return device.GetLogical().CreateFramebuffer(VkFramebufferCreateInfo{
         .sType = VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO,
         .pNext = nullptr,
         .flags = 0,
-        .renderPass = *renderpass,
+        .renderPass = *rd,
         .attachmentCount = 1,
         .pAttachments = &image_view,
         .width = extent.width,
@@ -308,9 +466,21 @@ void VKBlitScreen::CreateDynamicResources() {
     CreateRenderPass();
     CreateFramebuffers();
     CreateGraphicsPipeline();
+    fsr.reset();
+    if (Settings::values.scaling_filter.GetValue() == Settings::ScalingFilter::Fsr) {
+        CreateFSR();
+    }
 }
 
 void VKBlitScreen::RefreshResources(const Tegra::FramebufferConfig& framebuffer) {
+    if (Settings::values.scaling_filter.GetValue() == Settings::ScalingFilter::Fsr) {
+        if (!fsr) {
+            CreateFSR();
+        }
+    } else {
+        fsr.reset();
+    }
+
     if (framebuffer.width == raw_width && framebuffer.height == raw_height && !raw_images.empty()) {
         return;
     }
@@ -324,7 +494,16 @@ void VKBlitScreen::RefreshResources(const Tegra::FramebufferConfig& framebuffer)
 
 void VKBlitScreen::CreateShaders() {
     vertex_shader = BuildShader(device, VULKAN_PRESENT_VERT_SPV);
-    fragment_shader = BuildShader(device, VULKAN_PRESENT_FRAG_SPV);
+    fxaa_vertex_shader = BuildShader(device, FXAA_VERT_SPV);
+    fxaa_fragment_shader = BuildShader(device, FXAA_FRAG_SPV);
+    bilinear_fragment_shader = BuildShader(device, VULKAN_PRESENT_FRAG_SPV);
+    bicubic_fragment_shader = BuildShader(device, PRESENT_BICUBIC_FRAG_SPV);
+    gaussian_fragment_shader = BuildShader(device, PRESENT_GAUSSIAN_FRAG_SPV);
+    if (device.IsFloat16Supported()) {
+        scaleforce_fragment_shader = BuildShader(device, VULKAN_PRESENT_SCALEFORCE_FP16_FRAG_SPV);
+    } else {
+        scaleforce_fragment_shader = BuildShader(device, VULKAN_PRESENT_SCALEFORCE_FP32_FRAG_SPV);
+    }
 }
 
 void VKBlitScreen::CreateSemaphores() {
@@ -344,6 +523,13 @@ void VKBlitScreen::CreateDescriptorPool() {
         },
     }};
 
+    const std::array<VkDescriptorPoolSize, 1> pool_sizes_aa{{
+        {
+            .type = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,
+            .descriptorCount = static_cast<u32>(image_count * 2),
+        },
+    }};
+
     const VkDescriptorPoolCreateInfo ci{
         .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_POOL_CREATE_INFO,
         .pNext = nullptr,
@@ -353,19 +539,33 @@ void VKBlitScreen::CreateDescriptorPool() {
         .pPoolSizes = pool_sizes.data(),
     };
     descriptor_pool = device.GetLogical().CreateDescriptorPool(ci);
+
+    const VkDescriptorPoolCreateInfo ci_aa{
+        .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_POOL_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT,
+        .maxSets = static_cast<u32>(image_count),
+        .poolSizeCount = static_cast<u32>(pool_sizes_aa.size()),
+        .pPoolSizes = pool_sizes_aa.data(),
+    };
+    aa_descriptor_pool = device.GetLogical().CreateDescriptorPool(ci_aa);
 }
 
 void VKBlitScreen::CreateRenderPass() {
+    renderpass = CreateRenderPassImpl(swapchain.GetImageViewFormat());
+}
+
+vk::RenderPass VKBlitScreen::CreateRenderPassImpl(VkFormat format, bool is_present) {
     const VkAttachmentDescription color_attachment{
         .flags = 0,
-        .format = swapchain.GetImageViewFormat(),
+        .format = format,
         .samples = VK_SAMPLE_COUNT_1_BIT,
         .loadOp = VK_ATTACHMENT_LOAD_OP_CLEAR,
         .storeOp = VK_ATTACHMENT_STORE_OP_STORE,
         .stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE,
         .stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE,
         .initialLayout = VK_IMAGE_LAYOUT_UNDEFINED,
-        .finalLayout = VK_IMAGE_LAYOUT_PRESENT_SRC_KHR,
+        .finalLayout = is_present ? VK_IMAGE_LAYOUT_PRESENT_SRC_KHR : VK_IMAGE_LAYOUT_GENERAL,
     };
 
     const VkAttachmentReference color_attachment_ref{
@@ -408,7 +608,7 @@ void VKBlitScreen::CreateRenderPass() {
         .pDependencies = &dependency,
     };
 
-    renderpass = device.GetLogical().CreateRenderPass(renderpass_ci);
+    return device.GetLogical().CreateRenderPass(renderpass_ci);
 }
 
 void VKBlitScreen::CreateDescriptorSetLayout() {
@@ -429,6 +629,23 @@ void VKBlitScreen::CreateDescriptorSetLayout() {
         },
     }};
 
+    const std::array<VkDescriptorSetLayoutBinding, 2> layout_bindings_aa{{
+        {
+            .binding = 0,
+            .descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,
+            .descriptorCount = 1,
+            .stageFlags = VK_SHADER_STAGE_VERTEX_BIT,
+            .pImmutableSamplers = nullptr,
+        },
+        {
+            .binding = 1,
+            .descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,
+            .descriptorCount = 1,
+            .stageFlags = VK_SHADER_STAGE_FRAGMENT_BIT,
+            .pImmutableSamplers = nullptr,
+        },
+    }};
+
     const VkDescriptorSetLayoutCreateInfo ci{
         .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO,
         .pNext = nullptr,
@@ -437,11 +654,21 @@ void VKBlitScreen::CreateDescriptorSetLayout() {
         .pBindings = layout_bindings.data(),
     };
 
+    const VkDescriptorSetLayoutCreateInfo ci_aa{
+        .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .bindingCount = static_cast<u32>(layout_bindings_aa.size()),
+        .pBindings = layout_bindings_aa.data(),
+    };
+
     descriptor_set_layout = device.GetLogical().CreateDescriptorSetLayout(ci);
+    aa_descriptor_set_layout = device.GetLogical().CreateDescriptorSetLayout(ci_aa);
 }
 
 void VKBlitScreen::CreateDescriptorSets() {
     const std::vector layouts(image_count, *descriptor_set_layout);
+    const std::vector layouts_aa(image_count, *aa_descriptor_set_layout);
 
     const VkDescriptorSetAllocateInfo ai{
         .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO,
@@ -451,7 +678,16 @@ void VKBlitScreen::CreateDescriptorSets() {
         .pSetLayouts = layouts.data(),
     };
 
+    const VkDescriptorSetAllocateInfo ai_aa{
+        .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO,
+        .pNext = nullptr,
+        .descriptorPool = *aa_descriptor_pool,
+        .descriptorSetCount = static_cast<u32>(image_count),
+        .pSetLayouts = layouts_aa.data(),
+    };
+
     descriptor_sets = descriptor_pool.Allocate(ai);
+    aa_descriptor_sets = aa_descriptor_pool.Allocate(ai_aa);
 }
 
 void VKBlitScreen::CreatePipelineLayout() {
@@ -464,11 +700,63 @@ void VKBlitScreen::CreatePipelineLayout() {
         .pushConstantRangeCount = 0,
         .pPushConstantRanges = nullptr,
     };
+    const VkPipelineLayoutCreateInfo ci_aa{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_LAYOUT_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .setLayoutCount = 1,
+        .pSetLayouts = aa_descriptor_set_layout.address(),
+        .pushConstantRangeCount = 0,
+        .pPushConstantRanges = nullptr,
+    };
     pipeline_layout = device.GetLogical().CreatePipelineLayout(ci);
+    aa_pipeline_layout = device.GetLogical().CreatePipelineLayout(ci_aa);
 }
 
 void VKBlitScreen::CreateGraphicsPipeline() {
-    const std::array<VkPipelineShaderStageCreateInfo, 2> shader_stages{{
+    const std::array<VkPipelineShaderStageCreateInfo, 2> bilinear_shader_stages{{
+        {
+            .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .stage = VK_SHADER_STAGE_VERTEX_BIT,
+            .module = *vertex_shader,
+            .pName = "main",
+            .pSpecializationInfo = nullptr,
+        },
+        {
+            .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .stage = VK_SHADER_STAGE_FRAGMENT_BIT,
+            .module = *bilinear_fragment_shader,
+            .pName = "main",
+            .pSpecializationInfo = nullptr,
+        },
+    }};
+
+    const std::array<VkPipelineShaderStageCreateInfo, 2> bicubic_shader_stages{{
+        {
+            .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .stage = VK_SHADER_STAGE_VERTEX_BIT,
+            .module = *vertex_shader,
+            .pName = "main",
+            .pSpecializationInfo = nullptr,
+        },
+        {
+            .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .stage = VK_SHADER_STAGE_FRAGMENT_BIT,
+            .module = *bicubic_fragment_shader,
+            .pName = "main",
+            .pSpecializationInfo = nullptr,
+        },
+    }};
+
+    const std::array<VkPipelineShaderStageCreateInfo, 2> gaussian_shader_stages{{
         {
             .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
             .pNext = nullptr,
@@ -483,7 +771,28 @@ void VKBlitScreen::CreateGraphicsPipeline() {
             .pNext = nullptr,
             .flags = 0,
             .stage = VK_SHADER_STAGE_FRAGMENT_BIT,
-            .module = *fragment_shader,
+            .module = *gaussian_fragment_shader,
+            .pName = "main",
+            .pSpecializationInfo = nullptr,
+        },
+    }};
+
+    const std::array<VkPipelineShaderStageCreateInfo, 2> scaleforce_shader_stages{{
+        {
+            .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .stage = VK_SHADER_STAGE_VERTEX_BIT,
+            .module = *vertex_shader,
+            .pName = "main",
+            .pSpecializationInfo = nullptr,
+        },
+        {
+            .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .stage = VK_SHADER_STAGE_FRAGMENT_BIT,
+            .module = *scaleforce_fragment_shader,
             .pName = "main",
             .pSpecializationInfo = nullptr,
         },
@@ -583,12 +892,12 @@ void VKBlitScreen::CreateGraphicsPipeline() {
         .pDynamicStates = dynamic_states.data(),
     };
 
-    const VkGraphicsPipelineCreateInfo pipeline_ci{
+    const VkGraphicsPipelineCreateInfo bilinear_pipeline_ci{
         .sType = VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO,
         .pNext = nullptr,
         .flags = 0,
-        .stageCount = static_cast<u32>(shader_stages.size()),
-        .pStages = shader_stages.data(),
+        .stageCount = static_cast<u32>(bilinear_shader_stages.size()),
+        .pStages = bilinear_shader_stages.data(),
         .pVertexInputState = &vertex_input_ci,
         .pInputAssemblyState = &input_assembly_ci,
         .pTessellationState = nullptr,
@@ -605,7 +914,76 @@ void VKBlitScreen::CreateGraphicsPipeline() {
         .basePipelineIndex = 0,
     };
 
-    pipeline = device.GetLogical().CreateGraphicsPipeline(pipeline_ci);
+    const VkGraphicsPipelineCreateInfo bicubic_pipeline_ci{
+        .sType = VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .stageCount = static_cast<u32>(bicubic_shader_stages.size()),
+        .pStages = bicubic_shader_stages.data(),
+        .pVertexInputState = &vertex_input_ci,
+        .pInputAssemblyState = &input_assembly_ci,
+        .pTessellationState = nullptr,
+        .pViewportState = &viewport_state_ci,
+        .pRasterizationState = &rasterization_ci,
+        .pMultisampleState = &multisampling_ci,
+        .pDepthStencilState = nullptr,
+        .pColorBlendState = &color_blend_ci,
+        .pDynamicState = &dynamic_state_ci,
+        .layout = *pipeline_layout,
+        .renderPass = *renderpass,
+        .subpass = 0,
+        .basePipelineHandle = 0,
+        .basePipelineIndex = 0,
+    };
+
+    const VkGraphicsPipelineCreateInfo gaussian_pipeline_ci{
+        .sType = VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .stageCount = static_cast<u32>(gaussian_shader_stages.size()),
+        .pStages = gaussian_shader_stages.data(),
+        .pVertexInputState = &vertex_input_ci,
+        .pInputAssemblyState = &input_assembly_ci,
+        .pTessellationState = nullptr,
+        .pViewportState = &viewport_state_ci,
+        .pRasterizationState = &rasterization_ci,
+        .pMultisampleState = &multisampling_ci,
+        .pDepthStencilState = nullptr,
+        .pColorBlendState = &color_blend_ci,
+        .pDynamicState = &dynamic_state_ci,
+        .layout = *pipeline_layout,
+        .renderPass = *renderpass,
+        .subpass = 0,
+        .basePipelineHandle = 0,
+        .basePipelineIndex = 0,
+    };
+
+    const VkGraphicsPipelineCreateInfo scaleforce_pipeline_ci{
+        .sType = VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .stageCount = static_cast<u32>(scaleforce_shader_stages.size()),
+        .pStages = scaleforce_shader_stages.data(),
+        .pVertexInputState = &vertex_input_ci,
+        .pInputAssemblyState = &input_assembly_ci,
+        .pTessellationState = nullptr,
+        .pViewportState = &viewport_state_ci,
+        .pRasterizationState = &rasterization_ci,
+        .pMultisampleState = &multisampling_ci,
+        .pDepthStencilState = nullptr,
+        .pColorBlendState = &color_blend_ci,
+        .pDynamicState = &dynamic_state_ci,
+        .layout = *pipeline_layout,
+        .renderPass = *renderpass,
+        .subpass = 0,
+        .basePipelineHandle = 0,
+        .basePipelineIndex = 0,
+    };
+
+    bilinear_pipeline = device.GetLogical().CreateGraphicsPipeline(bilinear_pipeline_ci);
+    bicubic_pipeline = device.GetLogical().CreateGraphicsPipeline(bicubic_pipeline_ci);
+    gaussian_pipeline = device.GetLogical().CreateGraphicsPipeline(gaussian_pipeline_ci);
+    scaleforce_pipeline = device.GetLogical().CreateGraphicsPipeline(scaleforce_pipeline_ci);
 }
 
 void VKBlitScreen::CreateSampler() {
@@ -614,8 +992,29 @@ void VKBlitScreen::CreateSampler() {
         .pNext = nullptr,
         .flags = 0,
         .magFilter = VK_FILTER_LINEAR,
+        .minFilter = VK_FILTER_LINEAR,
+        .mipmapMode = VK_SAMPLER_MIPMAP_MODE_NEAREST,
+        .addressModeU = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER,
+        .addressModeV = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER,
+        .addressModeW = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER,
+        .mipLodBias = 0.0f,
+        .anisotropyEnable = VK_FALSE,
+        .maxAnisotropy = 0.0f,
+        .compareEnable = VK_FALSE,
+        .compareOp = VK_COMPARE_OP_NEVER,
+        .minLod = 0.0f,
+        .maxLod = 0.0f,
+        .borderColor = VK_BORDER_COLOR_FLOAT_OPAQUE_BLACK,
+        .unnormalizedCoordinates = VK_FALSE,
+    };
+
+    const VkSamplerCreateInfo ci_nn{
+        .sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .magFilter = VK_FILTER_NEAREST,
         .minFilter = VK_FILTER_NEAREST,
-        .mipmapMode = VK_SAMPLER_MIPMAP_MODE_LINEAR,
+        .mipmapMode = VK_SAMPLER_MIPMAP_MODE_NEAREST,
         .addressModeU = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER,
         .addressModeV = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER,
         .addressModeW = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_BORDER,
@@ -631,6 +1030,7 @@ void VKBlitScreen::CreateSampler() {
     };
 
     sampler = device.GetLogical().CreateSampler(ci);
+    nn_sampler = device.GetLogical().CreateSampler(ci_nn);
 }
 
 void VKBlitScreen::CreateFramebuffers() {
@@ -639,7 +1039,7 @@ void VKBlitScreen::CreateFramebuffers() {
 
     for (std::size_t i = 0; i < image_count; ++i) {
         const VkImageView image_view{swapchain.GetImageViewIndex(i)};
-        framebuffers[i] = CreateFramebuffer(image_view, size);
+        framebuffers[i] = CreateFramebuffer(image_view, size, renderpass);
     }
 }
 
@@ -649,6 +1049,11 @@ void VKBlitScreen::ReleaseRawImages() {
     }
     raw_images.clear();
     raw_buffer_commits.clear();
+
+    aa_image_view.reset();
+    aa_image.reset();
+    aa_commit = MemoryCommit{};
+
     buffer.reset();
     buffer_commit = MemoryCommit{};
 }
@@ -675,8 +1080,11 @@ void VKBlitScreen::CreateRawImages(const Tegra::FramebufferConfig& framebuffer)
     raw_image_views.resize(image_count);
     raw_buffer_commits.resize(image_count);
 
-    for (size_t i = 0; i < image_count; ++i) {
-        raw_images[i] = device.GetLogical().CreateImage(VkImageCreateInfo{
+    const auto create_image = [&](bool used_on_framebuffer = false, u32 up_scale = 1,
+                                  u32 down_shift = 0) {
+        u32 extra_usages = used_on_framebuffer ? VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT
+                                               : VK_IMAGE_USAGE_TRANSFER_DST_BIT;
+        return device.GetLogical().CreateImage(VkImageCreateInfo{
             .sType = VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO,
             .pNext = nullptr,
             .flags = 0,
@@ -684,26 +1092,30 @@ void VKBlitScreen::CreateRawImages(const Tegra::FramebufferConfig& framebuffer)
             .format = GetFormat(framebuffer),
             .extent =
                 {
-                    .width = framebuffer.width,
-                    .height = framebuffer.height,
+                    .width = (up_scale * framebuffer.width) >> down_shift,
+                    .height = (up_scale * framebuffer.height) >> down_shift,
                     .depth = 1,
                 },
             .mipLevels = 1,
             .arrayLayers = 1,
             .samples = VK_SAMPLE_COUNT_1_BIT,
-            .tiling = VK_IMAGE_TILING_LINEAR,
-            .usage = VK_IMAGE_USAGE_TRANSFER_DST_BIT | VK_IMAGE_USAGE_SAMPLED_BIT,
+            .tiling = used_on_framebuffer ? VK_IMAGE_TILING_OPTIMAL : VK_IMAGE_TILING_LINEAR,
+            .usage = VK_IMAGE_USAGE_TRANSFER_DST_BIT | VK_IMAGE_USAGE_SAMPLED_BIT | extra_usages,
             .sharingMode = VK_SHARING_MODE_EXCLUSIVE,
             .queueFamilyIndexCount = 0,
             .pQueueFamilyIndices = nullptr,
             .initialLayout = VK_IMAGE_LAYOUT_UNDEFINED,
         });
-        raw_buffer_commits[i] = memory_allocator.Commit(raw_images[i], MemoryUsage::DeviceLocal);
-        raw_image_views[i] = device.GetLogical().CreateImageView(VkImageViewCreateInfo{
+    };
+    const auto create_commit = [&](vk::Image& image) {
+        return memory_allocator.Commit(image, MemoryUsage::DeviceLocal);
+    };
+    const auto create_image_view = [&](vk::Image& image) {
+        return device.GetLogical().CreateImageView(VkImageViewCreateInfo{
             .sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO,
             .pNext = nullptr,
             .flags = 0,
-            .image = *raw_images[i],
+            .image = *image,
             .viewType = VK_IMAGE_VIEW_TYPE_2D,
             .format = GetFormat(framebuffer),
             .components =
@@ -722,10 +1134,211 @@ void VKBlitScreen::CreateRawImages(const Tegra::FramebufferConfig& framebuffer)
                     .layerCount = 1,
                 },
         });
+    };
+
+    for (size_t i = 0; i < image_count; ++i) {
+        raw_images[i] = create_image();
+        raw_buffer_commits[i] = create_commit(raw_images[i]);
+        raw_image_views[i] = create_image_view(raw_images[i]);
     }
+
+    // AA Resources
+    const u32 up_scale = Settings::values.resolution_info.up_scale;
+    const u32 down_shift = Settings::values.resolution_info.down_shift;
+    aa_image = create_image(true, up_scale, down_shift);
+    aa_commit = create_commit(aa_image);
+    aa_image_view = create_image_view(aa_image);
+    VkExtent2D size{
+        .width = (up_scale * framebuffer.width) >> down_shift,
+        .height = (up_scale * framebuffer.height) >> down_shift,
+    };
+    if (aa_renderpass) {
+        aa_framebuffer = CreateFramebuffer(*aa_image_view, size, aa_renderpass);
+        return;
+    }
+    aa_renderpass = CreateRenderPassImpl(GetFormat(framebuffer), false);
+    aa_framebuffer = CreateFramebuffer(*aa_image_view, size, aa_renderpass);
+
+    const std::array<VkPipelineShaderStageCreateInfo, 2> fxaa_shader_stages{{
+        {
+            .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .stage = VK_SHADER_STAGE_VERTEX_BIT,
+            .module = *fxaa_vertex_shader,
+            .pName = "main",
+            .pSpecializationInfo = nullptr,
+        },
+        {
+            .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .stage = VK_SHADER_STAGE_FRAGMENT_BIT,
+            .module = *fxaa_fragment_shader,
+            .pName = "main",
+            .pSpecializationInfo = nullptr,
+        },
+    }};
+
+    const auto vertex_binding_description = ScreenRectVertex::GetDescription();
+    const auto vertex_attrs_description = ScreenRectVertex::GetAttributes();
+
+    const VkPipelineVertexInputStateCreateInfo vertex_input_ci{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_VERTEX_INPUT_STATE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .vertexBindingDescriptionCount = 1,
+        .pVertexBindingDescriptions = &vertex_binding_description,
+        .vertexAttributeDescriptionCount = u32{vertex_attrs_description.size()},
+        .pVertexAttributeDescriptions = vertex_attrs_description.data(),
+    };
+
+    const VkPipelineInputAssemblyStateCreateInfo input_assembly_ci{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_INPUT_ASSEMBLY_STATE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .topology = VK_PRIMITIVE_TOPOLOGY_TRIANGLE_STRIP,
+        .primitiveRestartEnable = VK_FALSE,
+    };
+
+    const VkPipelineViewportStateCreateInfo viewport_state_ci{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_STATE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .viewportCount = 1,
+        .pViewports = nullptr,
+        .scissorCount = 1,
+        .pScissors = nullptr,
+    };
+
+    const VkPipelineRasterizationStateCreateInfo rasterization_ci{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_STATE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .depthClampEnable = VK_FALSE,
+        .rasterizerDiscardEnable = VK_FALSE,
+        .polygonMode = VK_POLYGON_MODE_FILL,
+        .cullMode = VK_CULL_MODE_NONE,
+        .frontFace = VK_FRONT_FACE_CLOCKWISE,
+        .depthBiasEnable = VK_FALSE,
+        .depthBiasConstantFactor = 0.0f,
+        .depthBiasClamp = 0.0f,
+        .depthBiasSlopeFactor = 0.0f,
+        .lineWidth = 1.0f,
+    };
+
+    const VkPipelineMultisampleStateCreateInfo multisampling_ci{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_MULTISAMPLE_STATE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .rasterizationSamples = VK_SAMPLE_COUNT_1_BIT,
+        .sampleShadingEnable = VK_FALSE,
+        .minSampleShading = 0.0f,
+        .pSampleMask = nullptr,
+        .alphaToCoverageEnable = VK_FALSE,
+        .alphaToOneEnable = VK_FALSE,
+    };
+
+    const VkPipelineColorBlendAttachmentState color_blend_attachment{
+        .blendEnable = VK_FALSE,
+        .srcColorBlendFactor = VK_BLEND_FACTOR_ZERO,
+        .dstColorBlendFactor = VK_BLEND_FACTOR_ZERO,
+        .colorBlendOp = VK_BLEND_OP_ADD,
+        .srcAlphaBlendFactor = VK_BLEND_FACTOR_ZERO,
+        .dstAlphaBlendFactor = VK_BLEND_FACTOR_ZERO,
+        .alphaBlendOp = VK_BLEND_OP_ADD,
+        .colorWriteMask = VK_COLOR_COMPONENT_R_BIT | VK_COLOR_COMPONENT_G_BIT |
+                          VK_COLOR_COMPONENT_B_BIT | VK_COLOR_COMPONENT_A_BIT,
+    };
+
+    const VkPipelineColorBlendStateCreateInfo color_blend_ci{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_COLOR_BLEND_STATE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .logicOpEnable = VK_FALSE,
+        .logicOp = VK_LOGIC_OP_COPY,
+        .attachmentCount = 1,
+        .pAttachments = &color_blend_attachment,
+        .blendConstants = {0.0f, 0.0f, 0.0f, 0.0f},
+    };
+
+    static constexpr std::array dynamic_states{
+        VK_DYNAMIC_STATE_VIEWPORT,
+        VK_DYNAMIC_STATE_SCISSOR,
+    };
+    const VkPipelineDynamicStateCreateInfo dynamic_state_ci{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_DYNAMIC_STATE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .dynamicStateCount = static_cast<u32>(dynamic_states.size()),
+        .pDynamicStates = dynamic_states.data(),
+    };
+
+    const VkGraphicsPipelineCreateInfo fxaa_pipeline_ci{
+        .sType = VK_STRUCTURE_TYPE_GRAPHICS_PIPELINE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .stageCount = static_cast<u32>(fxaa_shader_stages.size()),
+        .pStages = fxaa_shader_stages.data(),
+        .pVertexInputState = &vertex_input_ci,
+        .pInputAssemblyState = &input_assembly_ci,
+        .pTessellationState = nullptr,
+        .pViewportState = &viewport_state_ci,
+        .pRasterizationState = &rasterization_ci,
+        .pMultisampleState = &multisampling_ci,
+        .pDepthStencilState = nullptr,
+        .pColorBlendState = &color_blend_ci,
+        .pDynamicState = &dynamic_state_ci,
+        .layout = *aa_pipeline_layout,
+        .renderPass = *aa_renderpass,
+        .subpass = 0,
+        .basePipelineHandle = 0,
+        .basePipelineIndex = 0,
+    };
+
+    // AA
+    aa_pipeline = device.GetLogical().CreateGraphicsPipeline(fxaa_pipeline_ci);
 }
 
-void VKBlitScreen::UpdateDescriptorSet(std::size_t image_index, VkImageView image_view) const {
+void VKBlitScreen::UpdateAADescriptorSet(std::size_t image_index, VkImageView image_view,
+                                         bool nn) const {
+    const VkDescriptorImageInfo image_info{
+        .sampler = nn ? *nn_sampler : *sampler,
+        .imageView = image_view,
+        .imageLayout = VK_IMAGE_LAYOUT_GENERAL,
+    };
+
+    const VkWriteDescriptorSet sampler_write{
+        .sType = VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET,
+        .pNext = nullptr,
+        .dstSet = aa_descriptor_sets[image_index],
+        .dstBinding = 0,
+        .dstArrayElement = 0,
+        .descriptorCount = 1,
+        .descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,
+        .pImageInfo = &image_info,
+        .pBufferInfo = nullptr,
+        .pTexelBufferView = nullptr,
+    };
+
+    const VkWriteDescriptorSet sampler_write_2{
+        .sType = VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET,
+        .pNext = nullptr,
+        .dstSet = aa_descriptor_sets[image_index],
+        .dstBinding = 1,
+        .dstArrayElement = 0,
+        .descriptorCount = 1,
+        .descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,
+        .pImageInfo = &image_info,
+        .pBufferInfo = nullptr,
+        .pTexelBufferView = nullptr,
+    };
+
+    device.GetLogical().UpdateDescriptorSets(std::array{sampler_write, sampler_write_2}, {});
+}
+
+void VKBlitScreen::UpdateDescriptorSet(std::size_t image_index, VkImageView image_view,
+                                       bool nn) const {
     const VkDescriptorBufferInfo buffer_info{
         .buffer = *buffer,
         .offset = offsetof(BufferData, uniform),
@@ -746,7 +1359,7 @@ void VKBlitScreen::UpdateDescriptorSet(std::size_t image_index, VkImageView imag
     };
 
     const VkDescriptorImageInfo image_info{
-        .sampler = *sampler,
+        .sampler = nn ? *nn_sampler : *sampler,
         .imageView = image_view,
         .imageLayout = VK_IMAGE_LAYOUT_GENERAL,
     };
@@ -798,17 +1411,19 @@ void VKBlitScreen::SetVertexData(BufferData& data, const Tegra::FramebufferConfi
     UNIMPLEMENTED_IF(framebuffer_crop_rect.top != 0);
     UNIMPLEMENTED_IF(framebuffer_crop_rect.left != 0);
 
-    // Scale the output by the crop width/height. This is commonly used with 1280x720 rendering
-    // (e.g. handheld mode) on a 1920x1080 framebuffer.
     f32 scale_u = 1.0f;
     f32 scale_v = 1.0f;
-    if (framebuffer_crop_rect.GetWidth() > 0) {
-        scale_u = static_cast<f32>(framebuffer_crop_rect.GetWidth()) /
-                  static_cast<f32>(screen_info.width);
-    }
-    if (framebuffer_crop_rect.GetHeight() > 0) {
-        scale_v = static_cast<f32>(framebuffer_crop_rect.GetHeight()) /
-                  static_cast<f32>(screen_info.height);
+    // Scale the output by the crop width/height. This is commonly used with 1280x720 rendering
+    // (e.g. handheld mode) on a 1920x1080 framebuffer.
+    if (!fsr) {
+        if (framebuffer_crop_rect.GetWidth() > 0) {
+            scale_u = static_cast<f32>(framebuffer_crop_rect.GetWidth()) /
+                      static_cast<f32>(screen_info.width);
+        }
+        if (framebuffer_crop_rect.GetHeight() > 0) {
+            scale_v = static_cast<f32>(framebuffer_crop_rect.GetHeight()) /
+                      static_cast<f32>(screen_info.height);
+        }
     }
 
     const auto& screen = layout.screen;
@@ -822,6 +1437,15 @@ void VKBlitScreen::SetVertexData(BufferData& data, const Tegra::FramebufferConfi
     data.vertices[3] = ScreenRectVertex(x + w, y + h, texcoords.bottom * scale_u, right * scale_v);
 }
 
+void VKBlitScreen::CreateFSR() {
+    const auto& layout = render_window.GetFramebufferLayout();
+    const VkExtent2D fsr_size{
+        .width = layout.screen.GetWidth(),
+        .height = layout.screen.GetHeight(),
+    };
+    fsr = std::make_unique<FSR>(device, memory_allocator, image_count, fsr_size);
+}
+
 u64 VKBlitScreen::CalculateBufferSize(const Tegra::FramebufferConfig& framebuffer) const {
     return sizeof(BufferData) + GetSizeInBytes(framebuffer) * image_count;
 }
diff --git a/src/video_core/renderer_vulkan/vk_blit_screen.h b/src/video_core/renderer_vulkan/vk_blit_screen.h
index 430bcfbca..bbca71af3 100644
--- a/src/video_core/renderer_vulkan/vk_blit_screen.h
+++ b/src/video_core/renderer_vulkan/vk_blit_screen.h
@@ -34,6 +34,7 @@ namespace Vulkan {
 struct ScreenInfo;
 
 class Device;
+class FSR;
 class RasterizerVulkan;
 class VKScheduler;
 class VKSwapchain;
@@ -66,6 +67,9 @@ public:
     [[nodiscard]] vk::Framebuffer CreateFramebuffer(const VkImageView& image_view,
                                                     VkExtent2D extent);
 
+    [[nodiscard]] vk::Framebuffer CreateFramebuffer(const VkImageView& image_view,
+                                                    VkExtent2D extent, vk::RenderPass& rd);
+
 private:
     struct BufferData;
 
@@ -74,6 +78,7 @@ private:
     void CreateSemaphores();
     void CreateDescriptorPool();
     void CreateRenderPass();
+    vk::RenderPass CreateRenderPassImpl(VkFormat, bool is_present = true);
     void CreateDescriptorSetLayout();
     void CreateDescriptorSets();
     void CreatePipelineLayout();
@@ -88,11 +93,14 @@ private:
     void CreateStagingBuffer(const Tegra::FramebufferConfig& framebuffer);
     void CreateRawImages(const Tegra::FramebufferConfig& framebuffer);
 
-    void UpdateDescriptorSet(std::size_t image_index, VkImageView image_view) const;
+    void UpdateDescriptorSet(std::size_t image_index, VkImageView image_view, bool nn) const;
+    void UpdateAADescriptorSet(std::size_t image_index, VkImageView image_view, bool nn) const;
     void SetUniformData(BufferData& data, const Layout::FramebufferLayout layout) const;
     void SetVertexData(BufferData& data, const Tegra::FramebufferConfig& framebuffer,
                        const Layout::FramebufferLayout layout) const;
 
+    void CreateFSR();
+
     u64 CalculateBufferSize(const Tegra::FramebufferConfig& framebuffer) const;
     u64 GetRawImageOffset(const Tegra::FramebufferConfig& framebuffer,
                           std::size_t image_index) const;
@@ -107,14 +115,24 @@ private:
     const VKScreenInfo& screen_info;
 
     vk::ShaderModule vertex_shader;
-    vk::ShaderModule fragment_shader;
+    vk::ShaderModule fxaa_vertex_shader;
+    vk::ShaderModule fxaa_fragment_shader;
+    vk::ShaderModule bilinear_fragment_shader;
+    vk::ShaderModule bicubic_fragment_shader;
+    vk::ShaderModule gaussian_fragment_shader;
+    vk::ShaderModule scaleforce_fragment_shader;
     vk::DescriptorPool descriptor_pool;
     vk::DescriptorSetLayout descriptor_set_layout;
     vk::PipelineLayout pipeline_layout;
-    vk::Pipeline pipeline;
+    vk::Pipeline nearest_neightbor_pipeline;
+    vk::Pipeline bilinear_pipeline;
+    vk::Pipeline bicubic_pipeline;
+    vk::Pipeline gaussian_pipeline;
+    vk::Pipeline scaleforce_pipeline;
     vk::RenderPass renderpass;
     std::vector<vk::Framebuffer> framebuffers;
     vk::DescriptorSets descriptor_sets;
+    vk::Sampler nn_sampler;
     vk::Sampler sampler;
 
     vk::Buffer buffer;
@@ -126,8 +144,22 @@ private:
     std::vector<vk::Image> raw_images;
     std::vector<vk::ImageView> raw_image_views;
     std::vector<MemoryCommit> raw_buffer_commits;
+
+    vk::DescriptorPool aa_descriptor_pool;
+    vk::DescriptorSetLayout aa_descriptor_set_layout;
+    vk::PipelineLayout aa_pipeline_layout;
+    vk::Pipeline aa_pipeline;
+    vk::RenderPass aa_renderpass;
+    vk::Framebuffer aa_framebuffer;
+    vk::DescriptorSets aa_descriptor_sets;
+    vk::Image aa_image;
+    vk::ImageView aa_image_view;
+    MemoryCommit aa_commit;
+
     u32 raw_width = 0;
     u32 raw_height = 0;
+
+    std::unique_ptr<FSR> fsr;
 };
 
 } // namespace Vulkan
diff --git a/src/video_core/renderer_vulkan/vk_buffer_cache.cpp b/src/video_core/renderer_vulkan/vk_buffer_cache.cpp
index 8ac58bc2f..5ffd93499 100644
--- a/src/video_core/renderer_vulkan/vk_buffer_cache.cpp
+++ b/src/video_core/renderer_vulkan/vk_buffer_cache.cpp
@@ -146,7 +146,7 @@ void BufferCacheRuntime::Finish() {
 }
 
 void BufferCacheRuntime::CopyBuffer(VkBuffer dst_buffer, VkBuffer src_buffer,
-                                    std::span<const VideoCommon::BufferCopy> copies) {
+                                    std::span<const VideoCommon::BufferCopy> copies, bool barrier) {
     static constexpr VkMemoryBarrier READ_BARRIER{
         .sType = VK_STRUCTURE_TYPE_MEMORY_BARRIER,
         .pNext = nullptr,
@@ -163,10 +163,42 @@ void BufferCacheRuntime::CopyBuffer(VkBuffer dst_buffer, VkBuffer src_buffer,
     boost::container::small_vector<VkBufferCopy, 3> vk_copies(copies.size());
     std::ranges::transform(copies, vk_copies.begin(), MakeBufferCopy);
     scheduler.RequestOutsideRenderPassOperationContext();
-    scheduler.Record([src_buffer, dst_buffer, vk_copies](vk::CommandBuffer cmdbuf) {
+    scheduler.Record([src_buffer, dst_buffer, vk_copies, barrier](vk::CommandBuffer cmdbuf) {
+        if (barrier) {
+            cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_ALL_COMMANDS_BIT,
+                                   VK_PIPELINE_STAGE_TRANSFER_BIT, 0, READ_BARRIER);
+        }
+        cmdbuf.CopyBuffer(src_buffer, dst_buffer, vk_copies);
+        if (barrier) {
+            cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_TRANSFER_BIT,
+                                   VK_PIPELINE_STAGE_ALL_COMMANDS_BIT, 0, WRITE_BARRIER);
+        }
+    });
+}
+
+void BufferCacheRuntime::PreCopyBarrier() {
+    static constexpr VkMemoryBarrier READ_BARRIER{
+        .sType = VK_STRUCTURE_TYPE_MEMORY_BARRIER,
+        .pNext = nullptr,
+        .srcAccessMask = VK_ACCESS_MEMORY_WRITE_BIT,
+        .dstAccessMask = VK_ACCESS_TRANSFER_READ_BIT | VK_ACCESS_TRANSFER_WRITE_BIT,
+    };
+    scheduler.RequestOutsideRenderPassOperationContext();
+    scheduler.Record([](vk::CommandBuffer cmdbuf) {
         cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_ALL_COMMANDS_BIT, VK_PIPELINE_STAGE_TRANSFER_BIT,
                                0, READ_BARRIER);
-        cmdbuf.CopyBuffer(src_buffer, dst_buffer, vk_copies);
+    });
+}
+
+void BufferCacheRuntime::PostCopyBarrier() {
+    static constexpr VkMemoryBarrier WRITE_BARRIER{
+        .sType = VK_STRUCTURE_TYPE_MEMORY_BARRIER,
+        .pNext = nullptr,
+        .srcAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT,
+        .dstAccessMask = VK_ACCESS_MEMORY_READ_BIT | VK_ACCESS_MEMORY_WRITE_BIT,
+    };
+    scheduler.RequestOutsideRenderPassOperationContext();
+    scheduler.Record([](vk::CommandBuffer cmdbuf) {
         cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_TRANSFER_BIT, VK_PIPELINE_STAGE_ALL_COMMANDS_BIT,
                                0, WRITE_BARRIER);
     });
diff --git a/src/video_core/renderer_vulkan/vk_buffer_cache.h b/src/video_core/renderer_vulkan/vk_buffer_cache.h
index c27402ff0..1ee0d8420 100644
--- a/src/video_core/renderer_vulkan/vk_buffer_cache.h
+++ b/src/video_core/renderer_vulkan/vk_buffer_cache.h
@@ -69,8 +69,12 @@ public:
 
     [[nodiscard]] StagingBufferRef DownloadStagingBuffer(size_t size);
 
+    void PreCopyBarrier();
+
     void CopyBuffer(VkBuffer src_buffer, VkBuffer dst_buffer,
-                    std::span<const VideoCommon::BufferCopy> copies);
+                    std::span<const VideoCommon::BufferCopy> copies, bool barrier = true);
+
+    void PostCopyBarrier();
 
     void ClearBuffer(VkBuffer dest_buffer, u32 offset, size_t size, u32 value);
 
diff --git a/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp b/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp
index 44faf626a..de36bcdb7 100644
--- a/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp
+++ b/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp
@@ -22,6 +22,7 @@
 namespace Vulkan {
 
 using Shader::ImageBufferDescriptor;
+using Shader::Backend::SPIRV::RESCALING_LAYOUT_WORDS_OFFSET;
 using Tegra::Texture::TexturePair;
 
 ComputePipeline::ComputePipeline(const Device& device_, DescriptorPool& descriptor_pool,
@@ -108,8 +109,7 @@ void ComputePipeline::Configure(Tegra::Engines::KeplerCompute& kepler_compute,
     texture_cache.SynchronizeComputeDescriptors();
 
     static constexpr size_t max_elements = 64;
-    std::array<ImageId, max_elements> image_view_ids;
-    boost::container::static_vector<u32, max_elements> image_view_indices;
+    boost::container::static_vector<VideoCommon::ImageViewInOut, max_elements> views;
     boost::container::static_vector<VkSampler, max_elements> samplers;
 
     const auto& qmd{kepler_compute.launch_description};
@@ -134,30 +134,37 @@ void ComputePipeline::Configure(Tegra::Engines::KeplerCompute& kepler_compute,
         }
         return TexturePair(gpu_memory.Read<u32>(addr), via_header_index);
     }};
-    const auto add_image{[&](const auto& desc) {
+    const auto add_image{[&](const auto& desc, bool blacklist) {
         for (u32 index = 0; index < desc.count; ++index) {
             const auto handle{read_handle(desc, index)};
-            image_view_indices.push_back(handle.first);
+            views.push_back({
+                .index = handle.first,
+                .blacklist = blacklist,
+                .id = {},
+            });
         }
     }};
-    std::ranges::for_each(info.texture_buffer_descriptors, add_image);
-    std::ranges::for_each(info.image_buffer_descriptors, add_image);
+    for (const auto& desc : info.texture_buffer_descriptors) {
+        add_image(desc, false);
+    }
+    for (const auto& desc : info.image_buffer_descriptors) {
+        add_image(desc, false);
+    }
     for (const auto& desc : info.texture_descriptors) {
         for (u32 index = 0; index < desc.count; ++index) {
             const auto handle{read_handle(desc, index)};
-            image_view_indices.push_back(handle.first);
+            views.push_back({handle.first});
 
             Sampler* const sampler = texture_cache.GetComputeSampler(handle.second);
             samplers.push_back(sampler->Handle());
         }
     }
-    std::ranges::for_each(info.image_descriptors, add_image);
-
-    const std::span indices_span(image_view_indices.data(), image_view_indices.size());
-    texture_cache.FillComputeImageViews(indices_span, image_view_ids);
+    for (const auto& desc : info.image_descriptors) {
+        add_image(desc, desc.is_written);
+    }
+    texture_cache.FillComputeImageViews(std::span(views.data(), views.size()));
 
     buffer_cache.UnbindComputeTextureBuffers();
-    ImageId* texture_buffer_ids{image_view_ids.data()};
     size_t index{};
     const auto add_buffer{[&](const auto& desc) {
         constexpr bool is_image = std::is_same_v<decltype(desc), const ImageBufferDescriptor&>;
@@ -166,11 +173,10 @@ void ComputePipeline::Configure(Tegra::Engines::KeplerCompute& kepler_compute,
             if constexpr (is_image) {
                 is_written = desc.is_written;
             }
-            ImageView& image_view = texture_cache.GetImageView(*texture_buffer_ids);
+            ImageView& image_view = texture_cache.GetImageView(views[index].id);
             buffer_cache.BindComputeTextureBuffer(index, image_view.GpuAddr(),
                                                   image_view.BufferSize(), image_view.format,
                                                   is_written, is_image);
-            ++texture_buffer_ids;
             ++index;
         }
     }};
@@ -180,9 +186,11 @@ void ComputePipeline::Configure(Tegra::Engines::KeplerCompute& kepler_compute,
     buffer_cache.UpdateComputeBuffers();
     buffer_cache.BindHostComputeBuffers();
 
+    RescalingPushConstant rescaling;
     const VkSampler* samplers_it{samplers.data()};
-    const ImageId* views_it{image_view_ids.data()};
-    PushImageDescriptors(info, samplers_it, views_it, texture_cache, update_descriptor_queue);
+    const VideoCommon::ImageViewInOut* views_it{views.data()};
+    PushImageDescriptors(texture_cache, update_descriptor_queue, info, rescaling, samplers_it,
+                         views_it);
 
     if (!is_built.load(std::memory_order::relaxed)) {
         // Wait for the pipeline to be built
@@ -192,11 +200,18 @@ void ComputePipeline::Configure(Tegra::Engines::KeplerCompute& kepler_compute,
         });
     }
     const void* const descriptor_data{update_descriptor_queue.UpdateData()};
-    scheduler.Record([this, descriptor_data](vk::CommandBuffer cmdbuf) {
+    const bool is_rescaling = !info.texture_descriptors.empty() || !info.image_descriptors.empty();
+    scheduler.Record([this, descriptor_data, is_rescaling,
+                      rescaling_data = rescaling.Data()](vk::CommandBuffer cmdbuf) {
         cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_COMPUTE, *pipeline);
         if (!descriptor_set_layout) {
             return;
         }
+        if (is_rescaling) {
+            cmdbuf.PushConstants(*pipeline_layout, VK_SHADER_STAGE_COMPUTE_BIT,
+                                 RESCALING_LAYOUT_WORDS_OFFSET, sizeof(rescaling_data),
+                                 rescaling_data.data());
+        }
         const VkDescriptorSet descriptor_set{descriptor_allocator.Commit()};
         const vk::Device& dev{device.GetLogical()};
         dev.UpdateDescriptorSet(descriptor_set, *descriptor_update_template, descriptor_data);
diff --git a/src/video_core/renderer_vulkan/vk_fsr.cpp b/src/video_core/renderer_vulkan/vk_fsr.cpp
new file mode 100644
index 000000000..73629d229
--- /dev/null
+++ b/src/video_core/renderer_vulkan/vk_fsr.cpp
@@ -0,0 +1,553 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <cmath>
+#include "common/bit_cast.h"
+#include "common/common_types.h"
+#include "common/div_ceil.h"
+
+#include "video_core/host_shaders/vulkan_fidelityfx_fsr_easu_fp16_comp_spv.h"
+#include "video_core/host_shaders/vulkan_fidelityfx_fsr_easu_fp32_comp_spv.h"
+#include "video_core/host_shaders/vulkan_fidelityfx_fsr_rcas_fp16_comp_spv.h"
+#include "video_core/host_shaders/vulkan_fidelityfx_fsr_rcas_fp32_comp_spv.h"
+#include "video_core/renderer_vulkan/vk_fsr.h"
+#include "video_core/renderer_vulkan/vk_scheduler.h"
+#include "video_core/renderer_vulkan/vk_shader_util.h"
+#include "video_core/vulkan_common/vulkan_device.h"
+
+namespace Vulkan {
+namespace {
+// Reimplementations of the constant generating functions in ffx_fsr1.h
+// GCC generated a lot of warnings when using the official header.
+u32 AU1_AH1_AF1(f32 f) {
+    static constexpr u32 base[512]{
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+        0x0000, 0x0000, 0x0000, 0x0000, 0x0001, 0x0002, 0x0004, 0x0008, 0x0010, 0x0020, 0x0040,
+        0x0080, 0x0100, 0x0200, 0x0400, 0x0800, 0x0c00, 0x1000, 0x1400, 0x1800, 0x1c00, 0x2000,
+        0x2400, 0x2800, 0x2c00, 0x3000, 0x3400, 0x3800, 0x3c00, 0x4000, 0x4400, 0x4800, 0x4c00,
+        0x5000, 0x5400, 0x5800, 0x5c00, 0x6000, 0x6400, 0x6800, 0x6c00, 0x7000, 0x7400, 0x7800,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff, 0x7bff,
+        0x7bff, 0x7bff, 0x7bff, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000,
+        0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8000, 0x8001, 0x8002, 0x8004, 0x8008,
+        0x8010, 0x8020, 0x8040, 0x8080, 0x8100, 0x8200, 0x8400, 0x8800, 0x8c00, 0x9000, 0x9400,
+        0x9800, 0x9c00, 0xa000, 0xa400, 0xa800, 0xac00, 0xb000, 0xb400, 0xb800, 0xbc00, 0xc000,
+        0xc400, 0xc800, 0xcc00, 0xd000, 0xd400, 0xd800, 0xdc00, 0xe000, 0xe400, 0xe800, 0xec00,
+        0xf000, 0xf400, 0xf800, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+        0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff, 0xfbff,
+    };
+    static constexpr s8 shift[512]{
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x17, 0x16,
+        0x15, 0x14, 0x13, 0x12, 0x11, 0x10, 0x0f, 0x0e, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d,
+        0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d,
+        0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x17,
+        0x16, 0x15, 0x14, 0x13, 0x12, 0x11, 0x10, 0x0f, 0x0e, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d,
+        0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d,
+        0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+        0x18, 0x18,
+    };
+    const u32 u = Common::BitCast<u32>(f);
+    const u32 i = u >> 23;
+    return base[i] + ((u & 0x7fffff) >> shift[i]);
+}
+
+u32 AU1_AH2_AF2(f32 a[2]) {
+    return AU1_AH1_AF1(a[0]) + (AU1_AH1_AF1(a[1]) << 16);
+}
+
+void FsrEasuCon(u32 con0[4], u32 con1[4], u32 con2[4], u32 con3[4], f32 inputViewportInPixelsX,
+                f32 inputViewportInPixelsY, f32 inputSizeInPixelsX, f32 inputSizeInPixelsY,
+                f32 outputSizeInPixelsX, f32 outputSizeInPixelsY) {
+    con0[0] = Common::BitCast<u32>(inputViewportInPixelsX / outputSizeInPixelsX);
+    con0[1] = Common::BitCast<u32>(inputViewportInPixelsY / outputSizeInPixelsY);
+    con0[2] = Common::BitCast<u32>(0.5f * inputViewportInPixelsX / outputSizeInPixelsX - 0.5f);
+    con0[3] = Common::BitCast<u32>(0.5f * inputViewportInPixelsY / outputSizeInPixelsY - 0.5f);
+    con1[0] = Common::BitCast<u32>(1.0f / inputSizeInPixelsX);
+    con1[1] = Common::BitCast<u32>(1.0f / inputSizeInPixelsY);
+    con1[2] = Common::BitCast<u32>(1.0f / inputSizeInPixelsX);
+    con1[3] = Common::BitCast<u32>(-1.0f / inputSizeInPixelsY);
+    con2[0] = Common::BitCast<u32>(-1.0f / inputSizeInPixelsX);
+    con2[1] = Common::BitCast<u32>(2.0f / inputSizeInPixelsY);
+    con2[2] = Common::BitCast<u32>(1.0f / inputSizeInPixelsX);
+    con2[3] = Common::BitCast<u32>(2.0f / inputSizeInPixelsY);
+    con3[0] = Common::BitCast<u32>(0.0f / inputSizeInPixelsX);
+    con3[1] = Common::BitCast<u32>(4.0f / inputSizeInPixelsY);
+    con3[2] = con3[3] = 0;
+}
+
+void FsrEasuConOffset(u32 con0[4], u32 con1[4], u32 con2[4], u32 con3[4],
+                      f32 inputViewportInPixelsX, f32 inputViewportInPixelsY,
+                      f32 inputSizeInPixelsX, f32 inputSizeInPixelsY, f32 outputSizeInPixelsX,
+                      f32 outputSizeInPixelsY, f32 inputOffsetInPixelsX, f32 inputOffsetInPixelsY) {
+    FsrEasuCon(con0, con1, con2, con3, inputViewportInPixelsX, inputViewportInPixelsY,
+               inputSizeInPixelsX, inputSizeInPixelsY, outputSizeInPixelsX, outputSizeInPixelsY);
+    con0[2] = Common::BitCast<u32>(0.5f * inputViewportInPixelsX / outputSizeInPixelsX - 0.5f +
+                                   inputOffsetInPixelsX);
+    con0[3] = Common::BitCast<u32>(0.5f * inputViewportInPixelsY / outputSizeInPixelsY - 0.5f +
+                                   inputOffsetInPixelsY);
+}
+
+void FsrRcasCon(u32* con, f32 sharpness) {
+    sharpness = std::exp2f(-sharpness);
+    f32 hSharp[2]{sharpness, sharpness};
+    con[0] = Common::BitCast<u32>(sharpness);
+    con[1] = AU1_AH2_AF2(hSharp);
+    con[2] = 0;
+    con[3] = 0;
+}
+} // Anonymous namespace
+
+FSR::FSR(const Device& device_, MemoryAllocator& memory_allocator_, size_t image_count_,
+         VkExtent2D output_size_)
+    : device{device_}, memory_allocator{memory_allocator_}, image_count{image_count_},
+      output_size{output_size_} {
+
+    CreateImages();
+    CreateSampler();
+    CreateShaders();
+    CreateDescriptorPool();
+    CreateDescriptorSetLayout();
+    CreateDescriptorSets();
+    CreatePipelineLayout();
+    CreatePipeline();
+}
+
+VkImageView FSR::Draw(VKScheduler& scheduler, size_t image_index, VkImageView image_view,
+                      VkExtent2D input_image_extent, const Common::Rectangle<int>& crop_rect) {
+
+    UpdateDescriptorSet(image_index, image_view);
+
+    scheduler.Record([this, image_index, input_image_extent, crop_rect](vk::CommandBuffer cmdbuf) {
+        const VkImageMemoryBarrier base_barrier{
+            .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER,
+            .pNext = nullptr,
+            .srcAccessMask = 0,
+            .dstAccessMask = 0,
+            .oldLayout = VK_IMAGE_LAYOUT_GENERAL,
+            .newLayout = VK_IMAGE_LAYOUT_GENERAL,
+            .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+            .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+            .image = {},
+            .subresourceRange =
+                {
+                    .aspectMask = VK_IMAGE_ASPECT_COLOR_BIT,
+                    .baseMipLevel = 0,
+                    .levelCount = 1,
+                    .baseArrayLayer = 0,
+                    .layerCount = 1,
+                },
+        };
+
+        cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_COMPUTE, *easu_pipeline);
+
+        std::array<u32, 4 * 4> push_constants;
+        FsrEasuConOffset(
+            push_constants.data() + 0, push_constants.data() + 4, push_constants.data() + 8,
+            push_constants.data() + 12,
+
+            static_cast<f32>(crop_rect.GetWidth()), static_cast<f32>(crop_rect.GetHeight()),
+            static_cast<f32>(input_image_extent.width), static_cast<f32>(input_image_extent.height),
+            static_cast<f32>(output_size.width), static_cast<f32>(output_size.height),
+            static_cast<f32>(crop_rect.left), static_cast<f32>(crop_rect.top));
+        cmdbuf.PushConstants(*pipeline_layout, VK_SHADER_STAGE_COMPUTE_BIT, push_constants);
+
+        {
+            VkImageMemoryBarrier fsr_write_barrier = base_barrier;
+            fsr_write_barrier.image = *images[image_index],
+            fsr_write_barrier.oldLayout = VK_IMAGE_LAYOUT_UNDEFINED;
+
+            cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_ALL_COMMANDS_BIT,
+                                   VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT, 0, fsr_write_barrier);
+        }
+
+        cmdbuf.BindDescriptorSets(VK_PIPELINE_BIND_POINT_COMPUTE, *pipeline_layout, 0,
+                                  descriptor_sets[image_index * 2], {});
+        cmdbuf.Dispatch(Common::DivCeil(output_size.width, 16u),
+                        Common::DivCeil(output_size.height, 16u), 1);
+
+        cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_COMPUTE, *rcas_pipeline);
+
+        FsrRcasCon(push_constants.data(), 0.25f);
+        cmdbuf.PushConstants(*pipeline_layout, VK_SHADER_STAGE_COMPUTE_BIT, push_constants);
+
+        {
+            std::array<VkImageMemoryBarrier, 2> barriers;
+            auto& fsr_read_barrier = barriers[0];
+            auto& blit_write_barrier = barriers[1];
+
+            fsr_read_barrier = base_barrier;
+            fsr_read_barrier.image = *images[image_index];
+            fsr_read_barrier.srcAccessMask = VK_ACCESS_SHADER_WRITE_BIT;
+            fsr_read_barrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT;
+
+            blit_write_barrier = base_barrier;
+            blit_write_barrier.image = *images[image_count + image_index];
+            blit_write_barrier.oldLayout = VK_IMAGE_LAYOUT_UNDEFINED;
+            blit_write_barrier.newLayout = VK_IMAGE_LAYOUT_GENERAL;
+
+            cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT,
+                                   VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT, 0, {}, {}, barriers);
+        }
+
+        cmdbuf.BindDescriptorSets(VK_PIPELINE_BIND_POINT_COMPUTE, *pipeline_layout, 0,
+                                  descriptor_sets[image_index * 2 + 1], {});
+        cmdbuf.Dispatch(Common::DivCeil(output_size.width, 16u),
+                        Common::DivCeil(output_size.height, 16u), 1);
+
+        {
+            std::array<VkImageMemoryBarrier, 1> barriers;
+            auto& blit_read_barrier = barriers[0];
+
+            blit_read_barrier = base_barrier;
+            blit_read_barrier.image = *images[image_count + image_index];
+            blit_read_barrier.srcAccessMask = VK_ACCESS_SHADER_WRITE_BIT;
+            blit_read_barrier.dstAccessMask = VK_ACCESS_SHADER_READ_BIT;
+
+            cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT,
+                                   VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT, 0, {}, {}, barriers);
+        }
+    });
+
+    return *image_views[image_count + image_index];
+}
+
+void FSR::CreateDescriptorPool() {
+    const std::array<VkDescriptorPoolSize, 2> pool_sizes{{
+        {
+            .type = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,
+            .descriptorCount = static_cast<u32>(image_count * 2),
+        },
+        {
+            .type = VK_DESCRIPTOR_TYPE_STORAGE_IMAGE,
+            .descriptorCount = static_cast<u32>(image_count * 2),
+        },
+    }};
+
+    const VkDescriptorPoolCreateInfo ci{
+        .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_POOL_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT,
+        .maxSets = static_cast<u32>(image_count * 2),
+        .poolSizeCount = static_cast<u32>(pool_sizes.size()),
+        .pPoolSizes = pool_sizes.data(),
+    };
+    descriptor_pool = device.GetLogical().CreateDescriptorPool(ci);
+}
+
+void FSR::CreateDescriptorSetLayout() {
+    const std::array<VkDescriptorSetLayoutBinding, 2> layout_bindings{{
+        {
+            .binding = 0,
+            .descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,
+            .descriptorCount = 1,
+            .stageFlags = VK_SHADER_STAGE_COMPUTE_BIT,
+            .pImmutableSamplers = sampler.address(),
+        },
+        {
+            .binding = 1,
+            .descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_IMAGE,
+            .descriptorCount = 1,
+            .stageFlags = VK_SHADER_STAGE_COMPUTE_BIT,
+            .pImmutableSamplers = sampler.address(),
+        },
+    }};
+
+    const VkDescriptorSetLayoutCreateInfo ci{
+        .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .bindingCount = static_cast<u32>(layout_bindings.size()),
+        .pBindings = layout_bindings.data(),
+    };
+
+    descriptor_set_layout = device.GetLogical().CreateDescriptorSetLayout(ci);
+}
+
+void FSR::CreateDescriptorSets() {
+    const u32 sets = static_cast<u32>(image_count * 2);
+    const std::vector layouts(sets, *descriptor_set_layout);
+
+    const VkDescriptorSetAllocateInfo ai{
+        .sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_ALLOCATE_INFO,
+        .pNext = nullptr,
+        .descriptorPool = *descriptor_pool,
+        .descriptorSetCount = sets,
+        .pSetLayouts = layouts.data(),
+    };
+
+    descriptor_sets = descriptor_pool.Allocate(ai);
+}
+
+void FSR::CreateImages() {
+    images.resize(image_count * 2);
+    image_views.resize(image_count * 2);
+    buffer_commits.resize(image_count * 2);
+
+    for (size_t i = 0; i < image_count * 2; ++i) {
+        images[i] = device.GetLogical().CreateImage(VkImageCreateInfo{
+            .sType = VK_STRUCTURE_TYPE_IMAGE_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .imageType = VK_IMAGE_TYPE_2D,
+            .format = VK_FORMAT_R16G16B16A16_SFLOAT,
+            .extent =
+                {
+                    .width = output_size.width,
+                    .height = output_size.height,
+                    .depth = 1,
+                },
+            .mipLevels = 1,
+            .arrayLayers = 1,
+            .samples = VK_SAMPLE_COUNT_1_BIT,
+            .tiling = VK_IMAGE_TILING_OPTIMAL,
+            .usage = VK_IMAGE_USAGE_TRANSFER_SRC_BIT | VK_IMAGE_USAGE_STORAGE_BIT |
+                     VK_IMAGE_USAGE_SAMPLED_BIT,
+            .sharingMode = VK_SHARING_MODE_EXCLUSIVE,
+            .queueFamilyIndexCount = 0,
+            .pQueueFamilyIndices = nullptr,
+            .initialLayout = VK_IMAGE_LAYOUT_UNDEFINED,
+        });
+        buffer_commits[i] = memory_allocator.Commit(images[i], MemoryUsage::DeviceLocal);
+        image_views[i] = device.GetLogical().CreateImageView(VkImageViewCreateInfo{
+            .sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO,
+            .pNext = nullptr,
+            .flags = 0,
+            .image = *images[i],
+            .viewType = VK_IMAGE_VIEW_TYPE_2D,
+            .format = VK_FORMAT_R16G16B16A16_SFLOAT,
+            .components =
+                {
+                    .r = VK_COMPONENT_SWIZZLE_IDENTITY,
+                    .g = VK_COMPONENT_SWIZZLE_IDENTITY,
+                    .b = VK_COMPONENT_SWIZZLE_IDENTITY,
+                    .a = VK_COMPONENT_SWIZZLE_IDENTITY,
+                },
+            .subresourceRange =
+                {
+                    .aspectMask = VK_IMAGE_ASPECT_COLOR_BIT,
+                    .baseMipLevel = 0,
+                    .levelCount = 1,
+                    .baseArrayLayer = 0,
+                    .layerCount = 1,
+                },
+        });
+    }
+}
+
+void FSR::CreatePipelineLayout() {
+    VkPushConstantRange push_const{
+        .stageFlags = VK_SHADER_STAGE_COMPUTE_BIT,
+        .offset = 0,
+        .size = sizeof(std::array<u32, 4 * 4>),
+    };
+    VkPipelineLayoutCreateInfo ci{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_LAYOUT_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .setLayoutCount = 1,
+        .pSetLayouts = descriptor_set_layout.address(),
+        .pushConstantRangeCount = 1,
+        .pPushConstantRanges = &push_const,
+    };
+
+    pipeline_layout = device.GetLogical().CreatePipelineLayout(ci);
+}
+
+void FSR::UpdateDescriptorSet(std::size_t image_index, VkImageView image_view) const {
+    const auto fsr_image_view = *image_views[image_index];
+    const auto blit_image_view = *image_views[image_count + image_index];
+
+    const VkDescriptorImageInfo image_info{
+        .sampler = VK_NULL_HANDLE,
+        .imageView = image_view,
+        .imageLayout = VK_IMAGE_LAYOUT_GENERAL,
+    };
+    const VkDescriptorImageInfo fsr_image_info{
+        .sampler = VK_NULL_HANDLE,
+        .imageView = fsr_image_view,
+        .imageLayout = VK_IMAGE_LAYOUT_GENERAL,
+    };
+    const VkDescriptorImageInfo blit_image_info{
+        .sampler = VK_NULL_HANDLE,
+        .imageView = blit_image_view,
+        .imageLayout = VK_IMAGE_LAYOUT_GENERAL,
+    };
+
+    VkWriteDescriptorSet sampler_write{
+        .sType = VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET,
+        .pNext = nullptr,
+        .dstSet = descriptor_sets[image_index * 2],
+        .dstBinding = 0,
+        .dstArrayElement = 0,
+        .descriptorCount = 1,
+        .descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER,
+        .pImageInfo = &image_info,
+        .pBufferInfo = nullptr,
+        .pTexelBufferView = nullptr,
+    };
+
+    VkWriteDescriptorSet output_write{
+        .sType = VK_STRUCTURE_TYPE_WRITE_DESCRIPTOR_SET,
+        .pNext = nullptr,
+        .dstSet = descriptor_sets[image_index * 2],
+        .dstBinding = 1,
+        .dstArrayElement = 0,
+        .descriptorCount = 1,
+        .descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_IMAGE,
+        .pImageInfo = &fsr_image_info,
+        .pBufferInfo = nullptr,
+        .pTexelBufferView = nullptr,
+    };
+
+    device.GetLogical().UpdateDescriptorSets(std::array{sampler_write, output_write}, {});
+
+    sampler_write.dstSet = descriptor_sets[image_index * 2 + 1];
+    sampler_write.pImageInfo = &fsr_image_info;
+    output_write.dstSet = descriptor_sets[image_index * 2 + 1];
+    output_write.pImageInfo = &blit_image_info;
+
+    device.GetLogical().UpdateDescriptorSets(std::array{sampler_write, output_write}, {});
+}
+
+void FSR::CreateSampler() {
+    const VkSamplerCreateInfo ci{
+        .sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .magFilter = VK_FILTER_LINEAR,
+        .minFilter = VK_FILTER_LINEAR,
+        .mipmapMode = VK_SAMPLER_MIPMAP_MODE_LINEAR,
+        .addressModeU = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE,
+        .addressModeV = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE,
+        .addressModeW = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE,
+        .mipLodBias = 0.0f,
+        .anisotropyEnable = VK_FALSE,
+        .maxAnisotropy = 0.0f,
+        .compareEnable = VK_FALSE,
+        .compareOp = VK_COMPARE_OP_NEVER,
+        .minLod = 0.0f,
+        .maxLod = 0.0f,
+        .borderColor = VK_BORDER_COLOR_FLOAT_OPAQUE_BLACK,
+        .unnormalizedCoordinates = VK_FALSE,
+    };
+
+    sampler = device.GetLogical().CreateSampler(ci);
+}
+
+void FSR::CreateShaders() {
+    if (device.IsFloat16Supported()) {
+        easu_shader = BuildShader(device, VULKAN_FIDELITYFX_FSR_EASU_FP16_COMP_SPV);
+        rcas_shader = BuildShader(device, VULKAN_FIDELITYFX_FSR_RCAS_FP16_COMP_SPV);
+    } else {
+        easu_shader = BuildShader(device, VULKAN_FIDELITYFX_FSR_EASU_FP32_COMP_SPV);
+        rcas_shader = BuildShader(device, VULKAN_FIDELITYFX_FSR_RCAS_FP32_COMP_SPV);
+    }
+}
+
+void FSR::CreatePipeline() {
+    VkPipelineShaderStageCreateInfo shader_stage_easu{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .stage = VK_SHADER_STAGE_COMPUTE_BIT,
+        .module = *easu_shader,
+        .pName = "main",
+        .pSpecializationInfo = nullptr,
+    };
+
+    VkPipelineShaderStageCreateInfo shader_stage_rcas{
+        .sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .stage = VK_SHADER_STAGE_COMPUTE_BIT,
+        .module = *rcas_shader,
+        .pName = "main",
+        .pSpecializationInfo = nullptr,
+    };
+
+    VkComputePipelineCreateInfo pipeline_ci_easu{
+        .sType = VK_STRUCTURE_TYPE_COMPUTE_PIPELINE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .stage = shader_stage_easu,
+        .layout = *pipeline_layout,
+        .basePipelineHandle = VK_NULL_HANDLE,
+        .basePipelineIndex = 0,
+    };
+
+    VkComputePipelineCreateInfo pipeline_ci_rcas{
+        .sType = VK_STRUCTURE_TYPE_COMPUTE_PIPELINE_CREATE_INFO,
+        .pNext = nullptr,
+        .flags = 0,
+        .stage = shader_stage_rcas,
+        .layout = *pipeline_layout,
+        .basePipelineHandle = VK_NULL_HANDLE,
+        .basePipelineIndex = 0,
+    };
+
+    easu_pipeline = device.GetLogical().CreateComputePipeline(pipeline_ci_easu);
+    rcas_pipeline = device.GetLogical().CreateComputePipeline(pipeline_ci_rcas);
+}
+
+} // namespace Vulkan
diff --git a/src/video_core/renderer_vulkan/vk_fsr.h b/src/video_core/renderer_vulkan/vk_fsr.h
new file mode 100644
index 000000000..6bbec3d36
--- /dev/null
+++ b/src/video_core/renderer_vulkan/vk_fsr.h
@@ -0,0 +1,54 @@
+// Copyright 2021 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include "common/math_util.h"
+#include "video_core/vulkan_common/vulkan_memory_allocator.h"
+#include "video_core/vulkan_common/vulkan_wrapper.h"
+
+namespace Vulkan {
+
+class Device;
+class VKScheduler;
+
+class FSR {
+public:
+    explicit FSR(const Device& device, MemoryAllocator& memory_allocator, size_t image_count,
+                 VkExtent2D output_size);
+    VkImageView Draw(VKScheduler& scheduler, size_t image_index, VkImageView image_view,
+                     VkExtent2D input_image_extent, const Common::Rectangle<int>& crop_rect);
+
+private:
+    void CreateDescriptorPool();
+    void CreateDescriptorSetLayout();
+    void CreateDescriptorSets();
+    void CreateImages();
+    void CreateSampler();
+    void CreateShaders();
+    void CreatePipeline();
+    void CreatePipelineLayout();
+
+    void UpdateDescriptorSet(std::size_t image_index, VkImageView image_view) const;
+
+    const Device& device;
+    MemoryAllocator& memory_allocator;
+    size_t image_count;
+    VkExtent2D output_size;
+
+    vk::DescriptorPool descriptor_pool;
+    vk::DescriptorSetLayout descriptor_set_layout;
+    vk::DescriptorSets descriptor_sets;
+    vk::PipelineLayout pipeline_layout;
+    vk::ShaderModule easu_shader;
+    vk::ShaderModule rcas_shader;
+    vk::Pipeline easu_pipeline;
+    vk::Pipeline rcas_pipeline;
+    vk::Sampler sampler;
+    std::vector<vk::Image> images;
+    std::vector<vk::ImageView> image_views;
+    std::vector<MemoryCommit> buffer_commits;
+};
+
+} // namespace Vulkan
diff --git a/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp b/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp
index 8634c3316..616a7b457 100644
--- a/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp
+++ b/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp
@@ -32,6 +32,8 @@ namespace {
 using boost::container::small_vector;
 using boost::container::static_vector;
 using Shader::ImageBufferDescriptor;
+using Shader::Backend::SPIRV::RESCALING_LAYOUT_DOWN_FACTOR_OFFSET;
+using Shader::Backend::SPIRV::RESCALING_LAYOUT_WORDS_OFFSET;
 using Tegra::Texture::TexturePair;
 using VideoCore::Surface::PixelFormat;
 using VideoCore::Surface::PixelFormatFromDepthFormat;
@@ -235,6 +237,7 @@ GraphicsPipeline::GraphicsPipeline(
         stage_infos[stage] = *info;
         enabled_uniform_buffer_masks[stage] = info->constant_buffer_mask;
         std::ranges::copy(info->constant_buffer_used_sizes, uniform_buffer_sizes[stage].begin());
+        num_textures += Shader::NumDescriptors(info->texture_descriptors);
     }
     auto func{[this, shader_notify, &render_pass_cache, &descriptor_pool, pipeline_statistics] {
         DescriptorLayoutBuilder builder{MakeBuilder(device, stage_infos)};
@@ -277,11 +280,10 @@ void GraphicsPipeline::AddTransition(GraphicsPipeline* transition) {
 
 template <typename Spec>
 void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
-    std::array<ImageId, MAX_IMAGE_ELEMENTS> image_view_ids;
-    std::array<u32, MAX_IMAGE_ELEMENTS> image_view_indices;
+    std::array<VideoCommon::ImageViewInOut, MAX_IMAGE_ELEMENTS> views;
     std::array<VkSampler, MAX_IMAGE_ELEMENTS> samplers;
     size_t sampler_index{};
-    size_t image_index{};
+    size_t view_index{};
 
     texture_cache.SynchronizeGraphicsDescriptors();
 
@@ -322,26 +324,30 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
             }
             return TexturePair(gpu_memory.Read<u32>(addr), via_header_index);
         }};
-        const auto add_image{[&](const auto& desc) {
+        const auto add_image{[&](const auto& desc, bool blacklist) LAMBDA_FORCEINLINE {
             for (u32 index = 0; index < desc.count; ++index) {
                 const auto handle{read_handle(desc, index)};
-                image_view_indices[image_index++] = handle.first;
+                views[view_index++] = {
+                    .index = handle.first,
+                    .blacklist = blacklist,
+                    .id = {},
+                };
             }
         }};
         if constexpr (Spec::has_texture_buffers) {
             for (const auto& desc : info.texture_buffer_descriptors) {
-                add_image(desc);
+                add_image(desc, false);
             }
         }
         if constexpr (Spec::has_image_buffers) {
             for (const auto& desc : info.image_buffer_descriptors) {
-                add_image(desc);
+                add_image(desc, false);
             }
         }
         for (const auto& desc : info.texture_descriptors) {
             for (u32 index = 0; index < desc.count; ++index) {
                 const auto handle{read_handle(desc, index)};
-                image_view_indices[image_index++] = handle.first;
+                views[view_index++] = {handle.first};
 
                 Sampler* const sampler{texture_cache.GetGraphicsSampler(handle.second)};
                 samplers[sampler_index++] = sampler->Handle();
@@ -349,7 +355,7 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
         }
         if constexpr (Spec::has_images) {
             for (const auto& desc : info.image_descriptors) {
-                add_image(desc);
+                add_image(desc, desc.is_written);
             }
         }
     }};
@@ -368,10 +374,9 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
     if constexpr (Spec::enabled_stages[4]) {
         config_stage(4);
     }
-    const std::span indices_span(image_view_indices.data(), image_index);
-    texture_cache.FillGraphicsImageViews(indices_span, image_view_ids);
+    texture_cache.FillGraphicsImageViews<Spec::has_images>(std::span(views.data(), view_index));
 
-    ImageId* texture_buffer_index{image_view_ids.data()};
+    VideoCommon::ImageViewInOut* texture_buffer_it{views.data()};
     const auto bind_stage_info{[&](size_t stage) LAMBDA_FORCEINLINE {
         size_t index{};
         const auto add_buffer{[&](const auto& desc) {
@@ -381,12 +386,12 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
                 if constexpr (is_image) {
                     is_written = desc.is_written;
                 }
-                ImageView& image_view{texture_cache.GetImageView(*texture_buffer_index)};
+                ImageView& image_view{texture_cache.GetImageView(texture_buffer_it->id)};
                 buffer_cache.BindGraphicsTextureBuffer(stage, index, image_view.GpuAddr(),
                                                        image_view.BufferSize(), image_view.format,
                                                        is_written, is_image);
                 ++index;
-                ++texture_buffer_index;
+                ++texture_buffer_it;
             }
         }};
         buffer_cache.UnbindGraphicsTextureBuffers(stage);
@@ -402,13 +407,9 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
                 add_buffer(desc);
             }
         }
-        for (const auto& desc : info.texture_descriptors) {
-            texture_buffer_index += desc.count;
-        }
+        texture_buffer_it += Shader::NumDescriptors(info.texture_descriptors);
         if constexpr (Spec::has_images) {
-            for (const auto& desc : info.image_descriptors) {
-                texture_buffer_index += desc.count;
-            }
+            texture_buffer_it += Shader::NumDescriptors(info.image_descriptors);
         }
     }};
     if constexpr (Spec::enabled_stages[0]) {
@@ -432,12 +433,13 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
 
     update_descriptor_queue.Acquire();
 
+    RescalingPushConstant rescaling;
     const VkSampler* samplers_it{samplers.data()};
-    const ImageId* views_it{image_view_ids.data()};
+    const VideoCommon::ImageViewInOut* views_it{views.data()};
     const auto prepare_stage{[&](size_t stage) LAMBDA_FORCEINLINE {
         buffer_cache.BindHostStageBuffers(stage);
-        PushImageDescriptors(stage_infos[stage], samplers_it, views_it, texture_cache,
-                             update_descriptor_queue);
+        PushImageDescriptors(texture_cache, update_descriptor_queue, stage_infos[stage], rescaling,
+                             samplers_it, views_it);
     }};
     if constexpr (Spec::enabled_stages[0]) {
         prepare_stage(0);
@@ -454,10 +456,10 @@ void GraphicsPipeline::ConfigureImpl(bool is_indexed) {
     if constexpr (Spec::enabled_stages[4]) {
         prepare_stage(4);
     }
-    ConfigureDraw();
+    ConfigureDraw(rescaling);
 }
 
-void GraphicsPipeline::ConfigureDraw() {
+void GraphicsPipeline::ConfigureDraw(const RescalingPushConstant& rescaling) {
     texture_cache.UpdateRenderTargets(false);
     scheduler.RequestRenderpass(texture_cache.GetFramebuffer());
 
@@ -468,12 +470,25 @@ void GraphicsPipeline::ConfigureDraw() {
             build_condvar.wait(lock, [this] { return is_built.load(std::memory_order::relaxed); });
         });
     }
+    const bool is_rescaling{texture_cache.IsRescaling()};
+    const bool update_rescaling{scheduler.UpdateRescaling(is_rescaling)};
     const bool bind_pipeline{scheduler.UpdateGraphicsPipeline(this)};
     const void* const descriptor_data{update_descriptor_queue.UpdateData()};
-    scheduler.Record([this, descriptor_data, bind_pipeline](vk::CommandBuffer cmdbuf) {
+    scheduler.Record([this, descriptor_data, bind_pipeline, rescaling_data = rescaling.Data(),
+                      is_rescaling, update_rescaling](vk::CommandBuffer cmdbuf) {
         if (bind_pipeline) {
             cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, *pipeline);
         }
+        cmdbuf.PushConstants(*pipeline_layout, VK_SHADER_STAGE_ALL_GRAPHICS,
+                             RESCALING_LAYOUT_WORDS_OFFSET, sizeof(rescaling_data),
+                             rescaling_data.data());
+        if (update_rescaling) {
+            const f32 config_down_factor{Settings::values.resolution_info.down_factor};
+            const f32 scale_down_factor{is_rescaling ? config_down_factor : 1.0f};
+            cmdbuf.PushConstants(*pipeline_layout, VK_SHADER_STAGE_ALL_GRAPHICS,
+                                 RESCALING_LAYOUT_DOWN_FACTOR_OFFSET, sizeof(scale_down_factor),
+                                 &scale_down_factor);
+        }
         if (!descriptor_set_layout) {
             return;
         }
@@ -826,18 +841,10 @@ void GraphicsPipeline::MakePipeline(VkRenderPass render_pass) {
 void GraphicsPipeline::Validate() {
     size_t num_images{};
     for (const auto& info : stage_infos) {
-        for (const auto& desc : info.texture_buffer_descriptors) {
-            num_images += desc.count;
-        }
-        for (const auto& desc : info.image_buffer_descriptors) {
-            num_images += desc.count;
-        }
-        for (const auto& desc : info.texture_descriptors) {
-            num_images += desc.count;
-        }
-        for (const auto& desc : info.image_descriptors) {
-            num_images += desc.count;
-        }
+        num_images += Shader::NumDescriptors(info.texture_buffer_descriptors);
+        num_images += Shader::NumDescriptors(info.image_buffer_descriptors);
+        num_images += Shader::NumDescriptors(info.texture_descriptors);
+        num_images += Shader::NumDescriptors(info.image_descriptors);
     }
     ASSERT(num_images <= MAX_IMAGE_ELEMENTS);
 }
diff --git a/src/video_core/renderer_vulkan/vk_graphics_pipeline.h b/src/video_core/renderer_vulkan/vk_graphics_pipeline.h
index 1c780e944..a0c1d8f07 100644
--- a/src/video_core/renderer_vulkan/vk_graphics_pipeline.h
+++ b/src/video_core/renderer_vulkan/vk_graphics_pipeline.h
@@ -62,6 +62,7 @@ namespace Vulkan {
 class Device;
 class PipelineStatistics;
 class RenderPassCache;
+class RescalingPushConstant;
 class VKScheduler;
 class VKUpdateDescriptorQueue;
 
@@ -113,7 +114,7 @@ private:
     template <typename Spec>
     void ConfigureImpl(bool is_indexed);
 
-    void ConfigureDraw();
+    void ConfigureDraw(const RescalingPushConstant& rescaling);
 
     void MakePipeline(VkRenderPass render_pass);
 
@@ -138,6 +139,7 @@ private:
     std::array<Shader::Info, NUM_STAGES> stage_infos;
     std::array<u32, 5> enabled_uniform_buffer_masks{};
     VideoCommon::UniformBufferSizes uniform_buffer_sizes{};
+    u32 num_textures{};
 
     vk::DescriptorSetLayout descriptor_set_layout;
     DescriptorAllocator descriptor_allocator;
diff --git a/src/video_core/renderer_vulkan/vk_master_semaphore.h b/src/video_core/renderer_vulkan/vk_master_semaphore.h
index 0886b7da8..9be9c9bed 100644
--- a/src/video_core/renderer_vulkan/vk_master_semaphore.h
+++ b/src/video_core/renderer_vulkan/vk_master_semaphore.h
@@ -70,7 +70,9 @@ public:
             return;
         }
         // If none of the above is hit, fallback to a regular wait
-        semaphore.Wait(tick);
+        while (!semaphore.Wait(tick)) {
+        }
+        Refresh();
     }
 
 private:
diff --git a/src/video_core/renderer_vulkan/vk_rasterizer.cpp b/src/video_core/renderer_vulkan/vk_rasterizer.cpp
index 30b47a7a0..fd334a146 100644
--- a/src/video_core/renderer_vulkan/vk_rasterizer.cpp
+++ b/src/video_core/renderer_vulkan/vk_rasterizer.cpp
@@ -58,18 +58,28 @@ struct DrawParams {
     bool is_indexed;
 };
 
-VkViewport GetViewportState(const Device& device, const Maxwell& regs, size_t index) {
+VkViewport GetViewportState(const Device& device, const Maxwell& regs, size_t index, float scale) {
     const auto& src = regs.viewport_transform[index];
-    const float width = src.scale_x * 2.0f;
-    float y = src.translate_y - src.scale_y;
-    float height = src.scale_y * 2.0f;
+    const auto conv = [scale](float value) {
+        float new_value = value * scale;
+        if (scale < 1.0f) {
+            const bool sign = std::signbit(value);
+            new_value = std::round(std::abs(new_value));
+            new_value = sign ? -new_value : new_value;
+        }
+        return new_value;
+    };
+    const float x = conv(src.translate_x - src.scale_x);
+    const float width = conv(src.scale_x * 2.0f);
+    float y = conv(src.translate_y - src.scale_y);
+    float height = conv(src.scale_y * 2.0f);
     if (regs.screen_y_control.y_negate) {
         y += height;
         height = -height;
     }
     const float reduce_z = regs.depth_mode == Maxwell::DepthMode::MinusOneToOne ? 1.0f : 0.0f;
     VkViewport viewport{
-        .x = src.translate_x - src.scale_x,
+        .x = x,
         .y = y,
         .width = width != 0.0f ? width : 1.0f,
         .height = height != 0.0f ? height : 1.0f,
@@ -83,14 +93,27 @@ VkViewport GetViewportState(const Device& device, const Maxwell& regs, size_t in
     return viewport;
 }
 
-VkRect2D GetScissorState(const Maxwell& regs, size_t index) {
+VkRect2D GetScissorState(const Maxwell& regs, size_t index, u32 up_scale = 1, u32 down_shift = 0) {
     const auto& src = regs.scissor_test[index];
     VkRect2D scissor;
+    const auto scale_up = [&](s32 value) -> s32 {
+        if (value == 0) {
+            return 0U;
+        }
+        const s32 upset = value * up_scale;
+        s32 acumm = 0;
+        if ((up_scale >> down_shift) == 0) {
+            acumm = upset % 2;
+        }
+        const s32 converted_value = (value * up_scale) >> down_shift;
+        return value < 0 ? std::min<s32>(converted_value - acumm, -1)
+                         : std::max<s32>(converted_value + acumm, 1);
+    };
     if (src.enable) {
-        scissor.offset.x = static_cast<s32>(src.min_x);
-        scissor.offset.y = static_cast<s32>(src.min_y);
-        scissor.extent.width = src.max_x - src.min_x;
-        scissor.extent.height = src.max_y - src.min_y;
+        scissor.offset.x = scale_up(static_cast<s32>(src.min_x));
+        scissor.offset.y = scale_up(static_cast<s32>(src.min_y));
+        scissor.extent.width = scale_up(src.max_x - src.min_x);
+        scissor.extent.height = scale_up(src.max_y - src.min_y);
     } else {
         scissor.offset.x = 0;
         scissor.offset.y = 0;
@@ -199,7 +222,7 @@ void RasterizerVulkan::Clear() {
 
     query_cache.UpdateCounters();
 
-    const auto& regs = maxwell3d.regs;
+    auto& regs = maxwell3d.regs;
     const bool use_color = regs.clear_buffers.R || regs.clear_buffers.G || regs.clear_buffers.B ||
                            regs.clear_buffers.A;
     const bool use_depth = regs.clear_buffers.Z;
@@ -214,8 +237,16 @@ void RasterizerVulkan::Clear() {
     const VkExtent2D render_area = framebuffer->RenderArea();
     scheduler.RequestRenderpass(framebuffer);
 
+    u32 up_scale = 1;
+    u32 down_shift = 0;
+    if (texture_cache.IsRescaling()) {
+        up_scale = Settings::values.resolution_info.up_scale;
+        down_shift = Settings::values.resolution_info.down_shift;
+    }
+    UpdateViewportsState(regs);
+
     VkClearRect clear_rect{
-        .rect = GetScissorState(regs, 0),
+        .rect = GetScissorState(regs, 0, up_scale, down_shift),
         .baseArrayLayer = regs.clear_buffers.layer,
         .layerCount = 1,
     };
@@ -230,7 +261,38 @@ void RasterizerVulkan::Clear() {
     const u32 color_attachment = regs.clear_buffers.RT;
     if (use_color && framebuffer->HasAspectColorBit(color_attachment)) {
         VkClearValue clear_value;
-        std::memcpy(clear_value.color.float32, regs.clear_color, sizeof(regs.clear_color));
+        bool is_integer = false;
+        bool is_signed = false;
+        size_t int_size = 8;
+        for (std::size_t i = 0; i < Tegra::Engines::Maxwell3D::Regs::NumRenderTargets; ++i) {
+            const auto& this_rt = regs.rt[i];
+            if (this_rt.Address() == 0) {
+                continue;
+            }
+            if (this_rt.format == Tegra::RenderTargetFormat::NONE) {
+                continue;
+            }
+            const auto format =
+                VideoCore::Surface::PixelFormatFromRenderTargetFormat(this_rt.format);
+            is_integer = IsPixelFormatInteger(format);
+            is_signed = IsPixelFormatSignedInteger(format);
+            int_size = PixelComponentSizeBitsInteger(format);
+            break;
+        }
+        if (!is_integer) {
+            std::memcpy(clear_value.color.float32, regs.clear_color, sizeof(regs.clear_color));
+        } else if (!is_signed) {
+            for (size_t i = 0; i < 4; i++) {
+                clear_value.color.uint32[i] = static_cast<u32>(
+                    static_cast<f32>(static_cast<u64>(int_size) << 1U) * regs.clear_color[i]);
+            }
+        } else {
+            for (size_t i = 0; i < 4; i++) {
+                clear_value.color.int32[i] =
+                    static_cast<s32>(static_cast<f32>(static_cast<s64>(int_size - 1) << 1) *
+                                     (regs.clear_color[i] - 0.5f));
+            }
+        }
 
         scheduler.Record([color_attachment, clear_value, clear_rect](vk::CommandBuffer cmdbuf) {
             const VkClearAttachment attachment{
@@ -595,15 +657,17 @@ void RasterizerVulkan::UpdateViewportsState(Tegra::Engines::Maxwell3D::Regs& reg
     if (!state_tracker.TouchViewports()) {
         return;
     }
+    const bool is_rescaling{texture_cache.IsRescaling()};
+    const float scale = is_rescaling ? Settings::values.resolution_info.up_factor : 1.0f;
     const std::array viewports{
-        GetViewportState(device, regs, 0),  GetViewportState(device, regs, 1),
-        GetViewportState(device, regs, 2),  GetViewportState(device, regs, 3),
-        GetViewportState(device, regs, 4),  GetViewportState(device, regs, 5),
-        GetViewportState(device, regs, 6),  GetViewportState(device, regs, 7),
-        GetViewportState(device, regs, 8),  GetViewportState(device, regs, 9),
-        GetViewportState(device, regs, 10), GetViewportState(device, regs, 11),
-        GetViewportState(device, regs, 12), GetViewportState(device, regs, 13),
-        GetViewportState(device, regs, 14), GetViewportState(device, regs, 15),
+        GetViewportState(device, regs, 0, scale),  GetViewportState(device, regs, 1, scale),
+        GetViewportState(device, regs, 2, scale),  GetViewportState(device, regs, 3, scale),
+        GetViewportState(device, regs, 4, scale),  GetViewportState(device, regs, 5, scale),
+        GetViewportState(device, regs, 6, scale),  GetViewportState(device, regs, 7, scale),
+        GetViewportState(device, regs, 8, scale),  GetViewportState(device, regs, 9, scale),
+        GetViewportState(device, regs, 10, scale), GetViewportState(device, regs, 11, scale),
+        GetViewportState(device, regs, 12, scale), GetViewportState(device, regs, 13, scale),
+        GetViewportState(device, regs, 14, scale), GetViewportState(device, regs, 15, scale),
     };
     scheduler.Record([viewports](vk::CommandBuffer cmdbuf) { cmdbuf.SetViewport(0, viewports); });
 }
@@ -612,13 +676,29 @@ void RasterizerVulkan::UpdateScissorsState(Tegra::Engines::Maxwell3D::Regs& regs
     if (!state_tracker.TouchScissors()) {
         return;
     }
+    u32 up_scale = 1;
+    u32 down_shift = 0;
+    if (texture_cache.IsRescaling()) {
+        up_scale = Settings::values.resolution_info.up_scale;
+        down_shift = Settings::values.resolution_info.down_shift;
+    }
     const std::array scissors{
-        GetScissorState(regs, 0),  GetScissorState(regs, 1),  GetScissorState(regs, 2),
-        GetScissorState(regs, 3),  GetScissorState(regs, 4),  GetScissorState(regs, 5),
-        GetScissorState(regs, 6),  GetScissorState(regs, 7),  GetScissorState(regs, 8),
-        GetScissorState(regs, 9),  GetScissorState(regs, 10), GetScissorState(regs, 11),
-        GetScissorState(regs, 12), GetScissorState(regs, 13), GetScissorState(regs, 14),
-        GetScissorState(regs, 15),
+        GetScissorState(regs, 0, up_scale, down_shift),
+        GetScissorState(regs, 1, up_scale, down_shift),
+        GetScissorState(regs, 2, up_scale, down_shift),
+        GetScissorState(regs, 3, up_scale, down_shift),
+        GetScissorState(regs, 4, up_scale, down_shift),
+        GetScissorState(regs, 5, up_scale, down_shift),
+        GetScissorState(regs, 6, up_scale, down_shift),
+        GetScissorState(regs, 7, up_scale, down_shift),
+        GetScissorState(regs, 8, up_scale, down_shift),
+        GetScissorState(regs, 9, up_scale, down_shift),
+        GetScissorState(regs, 10, up_scale, down_shift),
+        GetScissorState(regs, 11, up_scale, down_shift),
+        GetScissorState(regs, 12, up_scale, down_shift),
+        GetScissorState(regs, 13, up_scale, down_shift),
+        GetScissorState(regs, 14, up_scale, down_shift),
+        GetScissorState(regs, 15, up_scale, down_shift),
     };
     scheduler.Record([scissors](vk::CommandBuffer cmdbuf) { cmdbuf.SetScissor(0, scissors); });
 }
diff --git a/src/video_core/renderer_vulkan/vk_scheduler.cpp b/src/video_core/renderer_vulkan/vk_scheduler.cpp
index 0c11c814f..3bfdf41ba 100644
--- a/src/video_core/renderer_vulkan/vk_scheduler.cpp
+++ b/src/video_core/renderer_vulkan/vk_scheduler.cpp
@@ -128,6 +128,15 @@ bool VKScheduler::UpdateGraphicsPipeline(GraphicsPipeline* pipeline) {
     return true;
 }
 
+bool VKScheduler::UpdateRescaling(bool is_rescaling) {
+    if (state.rescaling_defined && is_rescaling == state.is_rescaling) {
+        return false;
+    }
+    state.rescaling_defined = true;
+    state.is_rescaling = is_rescaling;
+    return true;
+}
+
 void VKScheduler::WorkerThread(std::stop_token stop_token) {
     Common::SetCurrentThreadName("yuzu:VulkanWorker");
     do {
@@ -227,6 +236,7 @@ void VKScheduler::AllocateNewContext() {
 
 void VKScheduler::InvalidateState() {
     state.graphics_pipeline = nullptr;
+    state.rescaling_defined = false;
     state_tracker.InvalidateCommandBufferState();
 }
 
diff --git a/src/video_core/renderer_vulkan/vk_scheduler.h b/src/video_core/renderer_vulkan/vk_scheduler.h
index 85fc1712f..1b06c9296 100644
--- a/src/video_core/renderer_vulkan/vk_scheduler.h
+++ b/src/video_core/renderer_vulkan/vk_scheduler.h
@@ -56,6 +56,9 @@ public:
     /// Update the pipeline to the current execution context.
     bool UpdateGraphicsPipeline(GraphicsPipeline* pipeline);
 
+    /// Update the rescaling state. Returns true if the state has to be updated.
+    bool UpdateRescaling(bool is_rescaling);
+
     /// Invalidates current command buffer state except for render passes
     void InvalidateState();
 
@@ -185,6 +188,8 @@ private:
         VkFramebuffer framebuffer = nullptr;
         VkExtent2D render_area = {0, 0};
         GraphicsPipeline* graphics_pipeline = nullptr;
+        bool is_rescaling = false;
+        bool rescaling_defined = false;
     };
 
     void WorkerThread(std::stop_token stop_token);
diff --git a/src/video_core/renderer_vulkan/vk_state_tracker.h b/src/video_core/renderer_vulkan/vk_state_tracker.h
index 2f2d6b31f..40a149832 100644
--- a/src/video_core/renderer_vulkan/vk_state_tracker.h
+++ b/src/video_core/renderer_vulkan/vk_state_tracker.h
@@ -71,11 +71,15 @@ public:
     }
 
     bool TouchViewports() {
-        return Exchange(Dirty::Viewports, false);
+        const bool dirty_viewports = Exchange(Dirty::Viewports, false);
+        const bool rescale_viewports = Exchange(VideoCommon::Dirty::RescaleViewports, false);
+        return dirty_viewports || rescale_viewports;
     }
 
     bool TouchScissors() {
-        return Exchange(Dirty::Scissors, false);
+        const bool dirty_scissors = Exchange(Dirty::Scissors, false);
+        const bool rescale_scissors = Exchange(VideoCommon::Dirty::RescaleScissors, false);
+        return dirty_scissors || rescale_scissors;
     }
 
     bool TouchDepthBias() {
diff --git a/src/video_core/renderer_vulkan/vk_texture_cache.cpp b/src/video_core/renderer_vulkan/vk_texture_cache.cpp
index 06c5fb867..407fd2a15 100644
--- a/src/video_core/renderer_vulkan/vk_texture_cache.cpp
+++ b/src/video_core/renderer_vulkan/vk_texture_cache.cpp
@@ -32,10 +32,12 @@ using Tegra::Engines::Fermi2D;
 using Tegra::Texture::SwizzleSource;
 using Tegra::Texture::TextureMipmapFilter;
 using VideoCommon::BufferImageCopy;
+using VideoCommon::ImageFlagBits;
 using VideoCommon::ImageInfo;
 using VideoCommon::ImageType;
 using VideoCommon::SubresourceRange;
 using VideoCore::Surface::IsPixelFormatASTC;
+using VideoCore::Surface::IsPixelFormatInteger;
 
 namespace {
 constexpr VkBorderColor ConvertBorderColor(const std::array<float, 4>& color) {
@@ -588,8 +590,158 @@ struct RangedBarrierRange {
     UNREACHABLE_MSG("Invalid image format={}", format);
     return VK_FORMAT_R32_UINT;
 }
+
+void BlitScale(VKScheduler& scheduler, VkImage src_image, VkImage dst_image, const ImageInfo& info,
+               VkImageAspectFlags aspect_mask, const Settings::ResolutionScalingInfo& resolution,
+               bool up_scaling = true) {
+    const bool is_2d = info.type == ImageType::e2D;
+    const auto resources = info.resources;
+    const VkExtent2D extent{
+        .width = info.size.width,
+        .height = info.size.height,
+    };
+    // Depth and integer formats must use NEAREST filter for blits.
+    const bool is_color{aspect_mask == VK_IMAGE_ASPECT_COLOR_BIT};
+    const bool is_bilinear{is_color && !IsPixelFormatInteger(info.format)};
+    const VkFilter vk_filter = is_bilinear ? VK_FILTER_LINEAR : VK_FILTER_NEAREST;
+
+    scheduler.RequestOutsideRenderPassOperationContext();
+    scheduler.Record([dst_image, src_image, extent, resources, aspect_mask, resolution, is_2d,
+                      vk_filter, up_scaling](vk::CommandBuffer cmdbuf) {
+        const VkOffset2D src_size{
+            .x = static_cast<s32>(up_scaling ? extent.width : resolution.ScaleUp(extent.width)),
+            .y = static_cast<s32>(is_2d && up_scaling ? extent.height
+                                                      : resolution.ScaleUp(extent.height)),
+        };
+        const VkOffset2D dst_size{
+            .x = static_cast<s32>(up_scaling ? resolution.ScaleUp(extent.width) : extent.width),
+            .y = static_cast<s32>(is_2d && up_scaling ? resolution.ScaleUp(extent.height)
+                                                      : extent.height),
+        };
+        boost::container::small_vector<VkImageBlit, 4> regions;
+        regions.reserve(resources.levels);
+        for (s32 level = 0; level < resources.levels; level++) {
+            regions.push_back({
+                .srcSubresource{
+                    .aspectMask = aspect_mask,
+                    .mipLevel = static_cast<u32>(level),
+                    .baseArrayLayer = 0,
+                    .layerCount = static_cast<u32>(resources.layers),
+                },
+                .srcOffsets{
+                    {
+                        .x = 0,
+                        .y = 0,
+                        .z = 0,
+                    },
+                    {
+                        .x = std::max(1, src_size.x >> level),
+                        .y = std::max(1, src_size.y >> level),
+                        .z = 1,
+                    },
+                },
+                .dstSubresource{
+                    .aspectMask = aspect_mask,
+                    .mipLevel = static_cast<u32>(level),
+                    .baseArrayLayer = 0,
+                    .layerCount = static_cast<u32>(resources.layers),
+                },
+                .dstOffsets{
+                    {
+                        .x = 0,
+                        .y = 0,
+                        .z = 0,
+                    },
+                    {
+                        .x = std::max(1, dst_size.x >> level),
+                        .y = std::max(1, dst_size.y >> level),
+                        .z = 1,
+                    },
+                },
+            });
+        }
+        const VkImageSubresourceRange subresource_range{
+            .aspectMask = aspect_mask,
+            .baseMipLevel = 0,
+            .levelCount = VK_REMAINING_MIP_LEVELS,
+            .baseArrayLayer = 0,
+            .layerCount = VK_REMAINING_ARRAY_LAYERS,
+        };
+        const std::array read_barriers{
+            VkImageMemoryBarrier{
+                .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER,
+                .pNext = nullptr,
+                .srcAccessMask = VK_ACCESS_MEMORY_WRITE_BIT,
+                .dstAccessMask = VK_ACCESS_TRANSFER_READ_BIT,
+                .oldLayout = VK_IMAGE_LAYOUT_GENERAL,
+                .newLayout = VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL,
+                .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .image = src_image,
+                .subresourceRange = subresource_range,
+            },
+            VkImageMemoryBarrier{
+                .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER,
+                .pNext = nullptr,
+                .srcAccessMask = VK_ACCESS_SHADER_WRITE_BIT |
+                                 VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT |
+                                 VK_ACCESS_TRANSFER_WRITE_BIT,
+                .dstAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT,
+                .oldLayout = VK_IMAGE_LAYOUT_UNDEFINED, // Discard contents
+                .newLayout = VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
+                .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .image = dst_image,
+                .subresourceRange = subresource_range,
+            },
+        };
+        const std::array write_barriers{
+            VkImageMemoryBarrier{
+                .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER,
+                .pNext = nullptr,
+                .srcAccessMask = 0,
+                .dstAccessMask = VK_ACCESS_MEMORY_WRITE_BIT | VK_ACCESS_MEMORY_READ_BIT,
+                .oldLayout = VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL,
+                .newLayout = VK_IMAGE_LAYOUT_GENERAL,
+                .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .image = src_image,
+                .subresourceRange = subresource_range,
+            },
+            VkImageMemoryBarrier{
+                .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER,
+                .pNext = nullptr,
+                .srcAccessMask = VK_ACCESS_TRANSFER_WRITE_BIT,
+                .dstAccessMask = VK_ACCESS_MEMORY_WRITE_BIT | VK_ACCESS_MEMORY_READ_BIT,
+                .oldLayout = VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
+                .newLayout = VK_IMAGE_LAYOUT_GENERAL,
+                .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
+                .image = dst_image,
+                .subresourceRange = subresource_range,
+            },
+        };
+        cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_ALL_COMMANDS_BIT, VK_PIPELINE_STAGE_TRANSFER_BIT,
+                               0, nullptr, nullptr, read_barriers);
+        cmdbuf.BlitImage(src_image, VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL, dst_image,
+                         VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL, regions, vk_filter);
+        cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_TRANSFER_BIT, VK_PIPELINE_STAGE_ALL_COMMANDS_BIT,
+                               0, nullptr, nullptr, write_barriers);
+    });
+}
 } // Anonymous namespace
 
+TextureCacheRuntime::TextureCacheRuntime(const Device& device_, VKScheduler& scheduler_,
+                                         MemoryAllocator& memory_allocator_,
+                                         StagingBufferPool& staging_buffer_pool_,
+                                         BlitImageHelper& blit_image_helper_,
+                                         ASTCDecoderPass& astc_decoder_pass_,
+                                         RenderPassCache& render_pass_cache_)
+    : device{device_}, scheduler{scheduler_}, memory_allocator{memory_allocator_},
+      staging_buffer_pool{staging_buffer_pool_}, blit_image_helper{blit_image_helper_},
+      astc_decoder_pass{astc_decoder_pass_}, render_pass_cache{render_pass_cache_},
+      resolution{Settings::values.resolution_info} {}
+
 void TextureCacheRuntime::Finish() {
     scheduler.Finish();
 }
@@ -614,8 +766,8 @@ void TextureCacheRuntime::BlitImage(Framebuffer* dst_framebuffer, ImageView& dst
         return;
     }
     if (aspect_mask == VK_IMAGE_ASPECT_COLOR_BIT && !is_src_msaa && !is_dst_msaa) {
-        blit_image_helper.BlitColor(dst_framebuffer, src, dst_region, src_region, filter,
-                                    operation);
+        blit_image_helper.BlitColor(dst_framebuffer, src.Handle(Shader::TextureType::Color2D),
+                                    dst_region, src_region, filter, operation);
         return;
     }
     if (aspect_mask == (VK_IMAGE_ASPECT_DEPTH_BIT | VK_IMAGE_ASPECT_STENCIL_BIT)) {
@@ -719,26 +871,29 @@ void TextureCacheRuntime::BlitImage(Framebuffer* dst_framebuffer, ImageView& dst
     });
 }
 
-void TextureCacheRuntime::ConvertImage(Framebuffer* dst, ImageView& dst_view, ImageView& src_view) {
+void TextureCacheRuntime::ConvertImage(Framebuffer* dst, ImageView& dst_view, ImageView& src_view,
+                                       bool rescaled) {
+    const u32 up_scale = rescaled ? resolution.up_scale : 1;
+    const u32 down_shift = rescaled ? resolution.down_shift : 0;
     switch (dst_view.format) {
     case PixelFormat::R16_UNORM:
         if (src_view.format == PixelFormat::D16_UNORM) {
-            return blit_image_helper.ConvertD16ToR16(dst, src_view);
+            return blit_image_helper.ConvertD16ToR16(dst, src_view, up_scale, down_shift);
         }
         break;
     case PixelFormat::R32_FLOAT:
         if (src_view.format == PixelFormat::D32_FLOAT) {
-            return blit_image_helper.ConvertD32ToR32(dst, src_view);
+            return blit_image_helper.ConvertD32ToR32(dst, src_view, up_scale, down_shift);
         }
         break;
     case PixelFormat::D16_UNORM:
         if (src_view.format == PixelFormat::R16_UNORM) {
-            return blit_image_helper.ConvertR16ToD16(dst, src_view);
+            return blit_image_helper.ConvertR16ToD16(dst, src_view, up_scale, down_shift);
         }
         break;
     case PixelFormat::D32_FLOAT:
         if (src_view.format == PixelFormat::R32_FLOAT) {
-            return blit_image_helper.ConvertR32ToD32(dst, src_view);
+            return blit_image_helper.ConvertR32ToD32(dst, src_view, up_scale, down_shift);
         }
         break;
     default:
@@ -840,36 +995,39 @@ u64 TextureCacheRuntime::GetDeviceLocalMemory() const {
     return device.GetDeviceLocalMemory();
 }
 
-Image::Image(TextureCacheRuntime& runtime, const ImageInfo& info_, GPUVAddr gpu_addr_,
+void TextureCacheRuntime::TickFrame() {}
+
+Image::Image(TextureCacheRuntime& runtime_, const ImageInfo& info_, GPUVAddr gpu_addr_,
              VAddr cpu_addr_)
-    : VideoCommon::ImageBase(info_, gpu_addr_, cpu_addr_), scheduler{&runtime.scheduler},
-      image(MakeImage(runtime.device, info)),
-      commit(runtime.memory_allocator.Commit(image, MemoryUsage::DeviceLocal)),
+    : VideoCommon::ImageBase(info_, gpu_addr_, cpu_addr_), scheduler{&runtime_.scheduler},
+      runtime{&runtime_}, original_image(MakeImage(runtime_.device, info)),
+      commit(runtime_.memory_allocator.Commit(original_image, MemoryUsage::DeviceLocal)),
       aspect_mask(ImageAspectMask(info.format)) {
-    if (IsPixelFormatASTC(info.format) && !runtime.device.IsOptimalAstcSupported()) {
+    if (IsPixelFormatASTC(info.format) && !runtime->device.IsOptimalAstcSupported()) {
         if (Settings::values.accelerate_astc.GetValue()) {
             flags |= VideoCommon::ImageFlagBits::AcceleratedUpload;
         } else {
             flags |= VideoCommon::ImageFlagBits::Converted;
         }
     }
-    if (runtime.device.HasDebuggingToolAttached()) {
-        image.SetObjectNameEXT(VideoCommon::Name(*this).c_str());
+    if (runtime->device.HasDebuggingToolAttached()) {
+        original_image.SetObjectNameEXT(VideoCommon::Name(*this).c_str());
     }
     static constexpr VkImageViewUsageCreateInfo storage_image_view_usage_create_info{
         .sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_USAGE_CREATE_INFO,
         .pNext = nullptr,
         .usage = VK_IMAGE_USAGE_STORAGE_BIT,
     };
-    if (IsPixelFormatASTC(info.format) && !runtime.device.IsOptimalAstcSupported()) {
-        const auto& device = runtime.device.GetLogical();
+    current_image = *original_image;
+    if (IsPixelFormatASTC(info.format) && !runtime->device.IsOptimalAstcSupported()) {
+        const auto& device = runtime->device.GetLogical();
         storage_image_views.reserve(info.resources.levels);
         for (s32 level = 0; level < info.resources.levels; ++level) {
             storage_image_views.push_back(device.CreateImageView(VkImageViewCreateInfo{
                 .sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO,
                 .pNext = &storage_image_view_usage_create_info,
                 .flags = 0,
-                .image = *image,
+                .image = *original_image,
                 .viewType = VK_IMAGE_VIEW_TYPE_2D_ARRAY,
                 .format = VK_FORMAT_A8B8G8R8_UNORM_PACK32,
                 .components{
@@ -890,26 +1048,39 @@ Image::Image(TextureCacheRuntime& runtime, const ImageInfo& info_, GPUVAddr gpu_
     }
 }
 
+Image::Image(const VideoCommon::NullImageParams& params) : VideoCommon::ImageBase{params} {}
+
 Image::~Image() = default;
 
 void Image::UploadMemory(const StagingBufferRef& map, std::span<const BufferImageCopy> copies) {
     // TODO: Move this to another API
+    const bool is_rescaled = True(flags & ImageFlagBits::Rescaled);
+    if (is_rescaled) {
+        ScaleDown(true);
+    }
     scheduler->RequestOutsideRenderPassOperationContext();
     std::vector vk_copies = TransformBufferImageCopies(copies, map.offset, aspect_mask);
     const VkBuffer src_buffer = map.buffer;
-    const VkImage vk_image = *image;
+    const VkImage vk_image = *original_image;
     const VkImageAspectFlags vk_aspect_mask = aspect_mask;
     const bool is_initialized = std::exchange(initialized, true);
     scheduler->Record([src_buffer, vk_image, vk_aspect_mask, is_initialized,
                        vk_copies](vk::CommandBuffer cmdbuf) {
         CopyBufferToImage(cmdbuf, src_buffer, vk_image, vk_aspect_mask, is_initialized, vk_copies);
     });
+    if (is_rescaled) {
+        ScaleUp();
+    }
 }
 
 void Image::DownloadMemory(const StagingBufferRef& map, std::span<const BufferImageCopy> copies) {
+    const bool is_rescaled = True(flags & ImageFlagBits::Rescaled);
+    if (is_rescaled) {
+        ScaleDown();
+    }
     std::vector vk_copies = TransformBufferImageCopies(copies, map.offset, aspect_mask);
     scheduler->RequestOutsideRenderPassOperationContext();
-    scheduler->Record([buffer = map.buffer, image = *image, aspect_mask = aspect_mask,
+    scheduler->Record([buffer = map.buffer, image = *original_image, aspect_mask = aspect_mask,
                        vk_copies](vk::CommandBuffer cmdbuf) {
         const VkImageMemoryBarrier read_barrier{
             .sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER,
@@ -959,6 +1130,146 @@ void Image::DownloadMemory(const StagingBufferRef& map, std::span<const BufferIm
         cmdbuf.PipelineBarrier(VK_PIPELINE_STAGE_TRANSFER_BIT, VK_PIPELINE_STAGE_ALL_COMMANDS_BIT,
                                0, memory_write_barrier, nullptr, image_write_barrier);
     });
+    if (is_rescaled) {
+        ScaleUp(true);
+    }
+}
+
+bool Image::ScaleUp(bool ignore) {
+    if (True(flags & ImageFlagBits::Rescaled)) {
+        return false;
+    }
+    ASSERT(info.type != ImageType::Linear);
+    flags |= ImageFlagBits::Rescaled;
+    const auto& resolution = runtime->resolution;
+    if (!resolution.active) {
+        return false;
+    }
+    has_scaled = true;
+    const auto& device = runtime->device;
+    if (!scaled_image) {
+        const bool is_2d = info.type == ImageType::e2D;
+        const u32 scaled_width = resolution.ScaleUp(info.size.width);
+        const u32 scaled_height = is_2d ? resolution.ScaleUp(info.size.height) : info.size.height;
+        auto scaled_info = info;
+        scaled_info.size.width = scaled_width;
+        scaled_info.size.height = scaled_height;
+        scaled_image = MakeImage(device, scaled_info);
+        auto& allocator = runtime->memory_allocator;
+        scaled_commit = MemoryCommit(allocator.Commit(scaled_image, MemoryUsage::DeviceLocal));
+        ignore = false;
+    }
+    current_image = *scaled_image;
+    if (ignore) {
+        return true;
+    }
+
+    if (aspect_mask == 0) {
+        aspect_mask = ImageAspectMask(info.format);
+    }
+    static constexpr auto OPTIMAL_FORMAT = FormatType::Optimal;
+    const PixelFormat format = StorageFormat(info.format);
+    const auto vk_format = MaxwellToVK::SurfaceFormat(device, OPTIMAL_FORMAT, false, format).format;
+    const auto blit_usage = VK_FORMAT_FEATURE_BLIT_SRC_BIT | VK_FORMAT_FEATURE_BLIT_DST_BIT;
+    if (device.IsFormatSupported(vk_format, blit_usage, OPTIMAL_FORMAT)) {
+        BlitScale(*scheduler, *original_image, *scaled_image, info, aspect_mask, resolution);
+    } else {
+        return BlitScaleHelper(true);
+    }
+    return true;
+}
+
+bool Image::ScaleDown(bool ignore) {
+    if (False(flags & ImageFlagBits::Rescaled)) {
+        return false;
+    }
+    ASSERT(info.type != ImageType::Linear);
+    flags &= ~ImageFlagBits::Rescaled;
+    const auto& resolution = runtime->resolution;
+    if (!resolution.active) {
+        return false;
+    }
+    current_image = *original_image;
+    if (ignore) {
+        return true;
+    }
+    if (aspect_mask == 0) {
+        aspect_mask = ImageAspectMask(info.format);
+    }
+    static constexpr auto OPTIMAL_FORMAT = FormatType::Optimal;
+    const PixelFormat format = StorageFormat(info.format);
+    const auto& device = runtime->device;
+    const auto vk_format = MaxwellToVK::SurfaceFormat(device, OPTIMAL_FORMAT, false, format).format;
+    const auto blit_usage = VK_FORMAT_FEATURE_BLIT_SRC_BIT | VK_FORMAT_FEATURE_BLIT_DST_BIT;
+    if (device.IsFormatSupported(vk_format, blit_usage, OPTIMAL_FORMAT)) {
+        BlitScale(*scheduler, *scaled_image, *original_image, info, aspect_mask, resolution, false);
+    } else {
+        return BlitScaleHelper(false);
+    }
+    return true;
+}
+
+bool Image::BlitScaleHelper(bool scale_up) {
+    using namespace VideoCommon;
+    static constexpr auto BLIT_OPERATION = Tegra::Engines::Fermi2D::Operation::SrcCopy;
+    const bool is_color{aspect_mask == VK_IMAGE_ASPECT_COLOR_BIT};
+    const bool is_bilinear{is_color && !IsPixelFormatInteger(info.format)};
+    const auto operation = is_bilinear ? Tegra::Engines::Fermi2D::Filter::Bilinear
+                                       : Tegra::Engines::Fermi2D::Filter::Point;
+
+    const bool is_2d = info.type == ImageType::e2D;
+    const auto& resolution = runtime->resolution;
+    const u32 scaled_width = resolution.ScaleUp(info.size.width);
+    const u32 scaled_height = is_2d ? resolution.ScaleUp(info.size.height) : info.size.height;
+    std::unique_ptr<ImageView>& blit_view = scale_up ? scale_view : normal_view;
+    std::unique_ptr<Framebuffer>& blit_framebuffer =
+        scale_up ? scale_framebuffer : normal_framebuffer;
+    if (!blit_view) {
+        const auto view_info = ImageViewInfo(ImageViewType::e2D, info.format);
+        blit_view = std::make_unique<ImageView>(*runtime, view_info, NULL_IMAGE_ID, *this);
+    }
+
+    const u32 src_width = scale_up ? info.size.width : scaled_width;
+    const u32 src_height = scale_up ? info.size.height : scaled_height;
+    const u32 dst_width = scale_up ? scaled_width : info.size.width;
+    const u32 dst_height = scale_up ? scaled_height : info.size.height;
+    const Region2D src_region{
+        .start = {0, 0},
+        .end = {static_cast<s32>(src_width), static_cast<s32>(src_height)},
+    };
+    const Region2D dst_region{
+        .start = {0, 0},
+        .end = {static_cast<s32>(dst_width), static_cast<s32>(dst_height)},
+    };
+    const VkExtent2D extent{
+        .width = std::max(scaled_width, info.size.width),
+        .height = std::max(scaled_height, info.size.width),
+    };
+
+    auto* view_ptr = blit_view.get();
+    if (aspect_mask == VK_IMAGE_ASPECT_COLOR_BIT) {
+        if (!blit_framebuffer) {
+            blit_framebuffer = std::make_unique<Framebuffer>(*runtime, view_ptr, nullptr, extent);
+        }
+        const auto color_view = blit_view->Handle(Shader::TextureType::Color2D);
+
+        runtime->blit_image_helper.BlitColor(blit_framebuffer.get(), color_view, dst_region,
+                                             src_region, operation, BLIT_OPERATION);
+    } else if (!runtime->device.IsBlitDepthStencilSupported() &&
+               aspect_mask == (VK_IMAGE_ASPECT_DEPTH_BIT | VK_IMAGE_ASPECT_STENCIL_BIT)) {
+        if (!blit_framebuffer) {
+            blit_framebuffer = std::make_unique<Framebuffer>(*runtime, nullptr, view_ptr, extent);
+        }
+        runtime->blit_image_helper.BlitDepthStencil(blit_framebuffer.get(), blit_view->DepthView(),
+                                                    blit_view->StencilView(), dst_region,
+                                                    src_region, operation, BLIT_OPERATION);
+    } else {
+        // TODO: Use helper blits where applicable
+        flags &= ~ImageFlagBits::Rescaled;
+        LOG_ERROR(Render_Vulkan, "Device does not support scaling format {}", info.format);
+        return false;
+    }
+    return true;
 }
 
 ImageView::ImageView(TextureCacheRuntime& runtime, const VideoCommon::ImageViewInfo& info,
@@ -1052,9 +1363,11 @@ ImageView::ImageView(TextureCacheRuntime&, const VideoCommon::ImageInfo& info,
     : VideoCommon::ImageViewBase{info, view_info}, gpu_addr{gpu_addr_},
       buffer_size{VideoCommon::CalculateGuestSizeInBytes(info)} {}
 
-ImageView::ImageView(TextureCacheRuntime&, const VideoCommon::NullImageParams& params)
+ImageView::ImageView(TextureCacheRuntime&, const VideoCommon::NullImageViewParams& params)
     : VideoCommon::ImageViewBase{params} {}
 
+ImageView::~ImageView() = default;
+
 VkImageView ImageView::DepthView() {
     if (depth_view) {
         return *depth_view;
@@ -1137,7 +1450,8 @@ Sampler::Sampler(TextureCacheRuntime& runtime, const Tegra::Texture::TSCEntry& t
         LOG_WARNING(Render_Vulkan, "VK_EXT_sampler_filter_minmax is required");
     }
     // Some games have samplers with garbage. Sanitize them here.
-    const float max_anisotropy = std::clamp(tsc.MaxAnisotropy(), 1.0f, 16.0f);
+    const f32 max_anisotropy = std::clamp(tsc.MaxAnisotropy(), 1.0f, 16.0f);
+
     sampler = device.GetLogical().CreateSampler(VkSamplerCreateInfo{
         .sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO,
         .pNext = pnext,
@@ -1162,7 +1476,29 @@ Sampler::Sampler(TextureCacheRuntime& runtime, const Tegra::Texture::TSCEntry& t
 }
 
 Framebuffer::Framebuffer(TextureCacheRuntime& runtime, std::span<ImageView*, NUM_RT> color_buffers,
-                         ImageView* depth_buffer, const VideoCommon::RenderTargets& key) {
+                         ImageView* depth_buffer, const VideoCommon::RenderTargets& key)
+    : render_area{VkExtent2D{
+          .width = key.size.width,
+          .height = key.size.height,
+      }} {
+    CreateFramebuffer(runtime, color_buffers, depth_buffer);
+    if (runtime.device.HasDebuggingToolAttached()) {
+        framebuffer.SetObjectNameEXT(VideoCommon::Name(key).c_str());
+    }
+}
+
+Framebuffer::Framebuffer(TextureCacheRuntime& runtime, ImageView* color_buffer,
+                         ImageView* depth_buffer, VkExtent2D extent)
+    : render_area{extent} {
+    std::array<ImageView*, NUM_RT> color_buffers{color_buffer};
+    CreateFramebuffer(runtime, color_buffers, depth_buffer);
+}
+
+Framebuffer::~Framebuffer() = default;
+
+void Framebuffer::CreateFramebuffer(TextureCacheRuntime& runtime,
+                                    std::span<ImageView*, NUM_RT> color_buffers,
+                                    ImageView* depth_buffer) {
     std::vector<VkImageView> attachments;
     RenderPassKey renderpass_key{};
     s32 num_layers = 1;
@@ -1200,10 +1536,6 @@ Framebuffer::Framebuffer(TextureCacheRuntime& runtime, std::span<ImageView*, NUM
 
     renderpass = runtime.render_pass_cache.Get(renderpass_key);
 
-    render_area = VkExtent2D{
-        .width = key.size.width,
-        .height = key.size.height,
-    };
     num_color_buffers = static_cast<u32>(num_colors);
     framebuffer = runtime.device.GetLogical().CreateFramebuffer({
         .sType = VK_STRUCTURE_TYPE_FRAMEBUFFER_CREATE_INFO,
@@ -1212,13 +1544,10 @@ Framebuffer::Framebuffer(TextureCacheRuntime& runtime, std::span<ImageView*, NUM
         .renderPass = renderpass,
         .attachmentCount = static_cast<u32>(attachments.size()),
         .pAttachments = attachments.data(),
-        .width = key.size.width,
-        .height = key.size.height,
+        .width = render_area.width,
+        .height = render_area.height,
         .layers = static_cast<u32>(std::max(num_layers, 1)),
     });
-    if (runtime.device.HasDebuggingToolAttached()) {
-        framebuffer.SetObjectNameEXT(VideoCommon::Name(key).c_str());
-    }
 }
 
 void TextureCacheRuntime::AccelerateImageUpload(
diff --git a/src/video_core/renderer_vulkan/vk_texture_cache.h b/src/video_core/renderer_vulkan/vk_texture_cache.h
index b09c468e4..ff28b4e96 100644
--- a/src/video_core/renderer_vulkan/vk_texture_cache.h
+++ b/src/video_core/renderer_vulkan/vk_texture_cache.h
@@ -13,6 +13,10 @@
 #include "video_core/vulkan_common/vulkan_memory_allocator.h"
 #include "video_core/vulkan_common/vulkan_wrapper.h"
 
+namespace Settings {
+struct ResolutionScalingInfo;
+}
+
 namespace Vulkan {
 
 using VideoCommon::ImageId;
@@ -31,14 +35,14 @@ class RenderPassCache;
 class StagingBufferPool;
 class VKScheduler;
 
-struct TextureCacheRuntime {
-    const Device& device;
-    VKScheduler& scheduler;
-    MemoryAllocator& memory_allocator;
-    StagingBufferPool& staging_buffer_pool;
-    BlitImageHelper& blit_image_helper;
-    ASTCDecoderPass& astc_decoder_pass;
-    RenderPassCache& render_pass_cache;
+class TextureCacheRuntime {
+public:
+    explicit TextureCacheRuntime(const Device& device_, VKScheduler& scheduler_,
+                                 MemoryAllocator& memory_allocator_,
+                                 StagingBufferPool& staging_buffer_pool_,
+                                 BlitImageHelper& blit_image_helper_,
+                                 ASTCDecoderPass& astc_decoder_pass_,
+                                 RenderPassCache& render_pass_cache_);
 
     void Finish();
 
@@ -46,6 +50,10 @@ struct TextureCacheRuntime {
 
     StagingBufferRef DownloadStagingBuffer(size_t size);
 
+    void TickFrame();
+
+    u64 GetDeviceLocalMemory() const;
+
     void BlitImage(Framebuffer* dst_framebuffer, ImageView& dst, ImageView& src,
                    const Region2D& dst_region, const Region2D& src_region,
                    Tegra::Engines::Fermi2D::Filter filter,
@@ -53,7 +61,7 @@ struct TextureCacheRuntime {
 
     void CopyImage(Image& dst, Image& src, std::span<const VideoCommon::ImageCopy> copies);
 
-    void ConvertImage(Framebuffer* dst, ImageView& dst_view, ImageView& src_view);
+    void ConvertImage(Framebuffer* dst, ImageView& dst_view, ImageView& src_view, bool rescaled);
 
     bool CanAccelerateImageUpload(Image&) const noexcept {
         return false;
@@ -74,13 +82,21 @@ struct TextureCacheRuntime {
         return true;
     }
 
-    u64 GetDeviceLocalMemory() const;
+    const Device& device;
+    VKScheduler& scheduler;
+    MemoryAllocator& memory_allocator;
+    StagingBufferPool& staging_buffer_pool;
+    BlitImageHelper& blit_image_helper;
+    ASTCDecoderPass& astc_decoder_pass;
+    RenderPassCache& render_pass_cache;
+    const Settings::ResolutionScalingInfo& resolution;
 };
 
 class Image : public VideoCommon::ImageBase {
 public:
     explicit Image(TextureCacheRuntime&, const VideoCommon::ImageInfo& info, GPUVAddr gpu_addr,
                    VAddr cpu_addr);
+    explicit Image(const VideoCommon::NullImageParams&);
 
     ~Image();
 
@@ -97,7 +113,7 @@ public:
                         std::span<const VideoCommon::BufferImageCopy> copies);
 
     [[nodiscard]] VkImage Handle() const noexcept {
-        return *image;
+        return current_image;
     }
 
     [[nodiscard]] VkImageAspectFlags AspectMask() const noexcept {
@@ -113,14 +129,30 @@ public:
         return std::exchange(initialized, true);
     }
 
+    bool ScaleUp(bool ignore = false);
+
+    bool ScaleDown(bool ignore = false);
+
 private:
-    VKScheduler* scheduler;
-    vk::Image image;
+    bool BlitScaleHelper(bool scale_up);
+
+    VKScheduler* scheduler{};
+    TextureCacheRuntime* runtime{};
+
+    vk::Image original_image;
     MemoryCommit commit;
-    vk::ImageView image_view;
     std::vector<vk::ImageView> storage_image_views;
     VkImageAspectFlags aspect_mask = 0;
     bool initialized = false;
+    vk::Image scaled_image{};
+    MemoryCommit scaled_commit{};
+    VkImage current_image{};
+
+    std::unique_ptr<Framebuffer> scale_framebuffer;
+    std::unique_ptr<ImageView> scale_view;
+
+    std::unique_ptr<Framebuffer> normal_framebuffer;
+    std::unique_ptr<ImageView> normal_view;
 };
 
 class ImageView : public VideoCommon::ImageViewBase {
@@ -128,7 +160,15 @@ public:
     explicit ImageView(TextureCacheRuntime&, const VideoCommon::ImageViewInfo&, ImageId, Image&);
     explicit ImageView(TextureCacheRuntime&, const VideoCommon::ImageInfo&,
                        const VideoCommon::ImageViewInfo&, GPUVAddr);
-    explicit ImageView(TextureCacheRuntime&, const VideoCommon::NullImageParams&);
+    explicit ImageView(TextureCacheRuntime&, const VideoCommon::NullImageViewParams&);
+
+    ~ImageView();
+
+    ImageView(const ImageView&) = delete;
+    ImageView& operator=(const ImageView&) = delete;
+
+    ImageView(ImageView&&) = default;
+    ImageView& operator=(ImageView&&) = default;
 
     [[nodiscard]] VkImageView DepthView();
 
@@ -197,9 +237,23 @@ private:
 
 class Framebuffer {
 public:
-    explicit Framebuffer(TextureCacheRuntime&, std::span<ImageView*, NUM_RT> color_buffers,
+    explicit Framebuffer(TextureCacheRuntime& runtime, std::span<ImageView*, NUM_RT> color_buffers,
                          ImageView* depth_buffer, const VideoCommon::RenderTargets& key);
 
+    explicit Framebuffer(TextureCacheRuntime& runtime, ImageView* color_buffer,
+                         ImageView* depth_buffer, VkExtent2D extent);
+
+    ~Framebuffer();
+
+    Framebuffer(const Framebuffer&) = delete;
+    Framebuffer& operator=(const Framebuffer&) = delete;
+
+    Framebuffer(Framebuffer&&) = default;
+    Framebuffer& operator=(Framebuffer&&) = default;
+
+    void CreateFramebuffer(TextureCacheRuntime& runtime,
+                           std::span<ImageView*, NUM_RT> color_buffers, ImageView* depth_buffer);
+
     [[nodiscard]] VkFramebuffer Handle() const noexcept {
         return *framebuffer;
     }
diff --git a/src/video_core/surface.cpp b/src/video_core/surface.cpp
index eb1746265..58d262446 100644
--- a/src/video_core/surface.cpp
+++ b/src/video_core/surface.cpp
@@ -279,6 +279,80 @@ bool IsPixelFormatSRGB(PixelFormat format) {
     }
 }
 
+bool IsPixelFormatInteger(PixelFormat format) {
+    switch (format) {
+    case PixelFormat::A8B8G8R8_SINT:
+    case PixelFormat::A8B8G8R8_UINT:
+    case PixelFormat::A2B10G10R10_UINT:
+    case PixelFormat::R8_SINT:
+    case PixelFormat::R8_UINT:
+    case PixelFormat::R16G16B16A16_SINT:
+    case PixelFormat::R16G16B16A16_UINT:
+    case PixelFormat::R32G32B32A32_UINT:
+    case PixelFormat::R32G32B32A32_SINT:
+    case PixelFormat::R32G32_SINT:
+    case PixelFormat::R16_UINT:
+    case PixelFormat::R16_SINT:
+    case PixelFormat::R16G16_UINT:
+    case PixelFormat::R16G16_SINT:
+    case PixelFormat::R8G8_SINT:
+    case PixelFormat::R8G8_UINT:
+    case PixelFormat::R32G32_UINT:
+    case PixelFormat::R32_UINT:
+    case PixelFormat::R32_SINT:
+        return true;
+    default:
+        return false;
+    }
+}
+
+bool IsPixelFormatSignedInteger(PixelFormat format) {
+    switch (format) {
+    case PixelFormat::A8B8G8R8_SINT:
+    case PixelFormat::R8_SINT:
+    case PixelFormat::R16G16B16A16_SINT:
+    case PixelFormat::R32G32B32A32_SINT:
+    case PixelFormat::R32G32_SINT:
+    case PixelFormat::R16_SINT:
+    case PixelFormat::R16G16_SINT:
+    case PixelFormat::R8G8_SINT:
+    case PixelFormat::R32_SINT:
+        return true;
+    default:
+        return false;
+    }
+}
+
+size_t PixelComponentSizeBitsInteger(PixelFormat format) {
+    switch (format) {
+    case PixelFormat::A8B8G8R8_SINT:
+    case PixelFormat::A8B8G8R8_UINT:
+    case PixelFormat::R8_SINT:
+    case PixelFormat::R8_UINT:
+    case PixelFormat::R8G8_SINT:
+    case PixelFormat::R8G8_UINT:
+        return 8;
+    case PixelFormat::A2B10G10R10_UINT:
+        return 10;
+    case PixelFormat::R16G16B16A16_SINT:
+    case PixelFormat::R16G16B16A16_UINT:
+    case PixelFormat::R16_UINT:
+    case PixelFormat::R16_SINT:
+    case PixelFormat::R16G16_UINT:
+    case PixelFormat::R16G16_SINT:
+        return 16;
+    case PixelFormat::R32G32B32A32_UINT:
+    case PixelFormat::R32G32B32A32_SINT:
+    case PixelFormat::R32G32_SINT:
+    case PixelFormat::R32G32_UINT:
+    case PixelFormat::R32_UINT:
+    case PixelFormat::R32_SINT:
+        return 32;
+    default:
+        return 0;
+    }
+}
+
 std::pair<u32, u32> GetASTCBlockSize(PixelFormat format) {
     return {DefaultBlockWidth(format), DefaultBlockHeight(format)};
 }
diff --git a/src/video_core/surface.h b/src/video_core/surface.h
index 1503db81f..2ce7c7d33 100644
--- a/src/video_core/surface.h
+++ b/src/video_core/surface.h
@@ -460,6 +460,12 @@ bool IsPixelFormatASTC(PixelFormat format);
 
 bool IsPixelFormatSRGB(PixelFormat format);
 
+bool IsPixelFormatInteger(PixelFormat format);
+
+bool IsPixelFormatSignedInteger(PixelFormat format);
+
+size_t PixelComponentSizeBitsInteger(PixelFormat format);
+
 std::pair<u32, u32> GetASTCBlockSize(PixelFormat format);
 
 u64 EstimatedDecompressedSize(u64 base_size, PixelFormat format);
diff --git a/src/video_core/texture_cache/image_base.cpp b/src/video_core/texture_cache/image_base.cpp
index 6052d148a..3db2fdf34 100644
--- a/src/video_core/texture_cache/image_base.cpp
+++ b/src/video_core/texture_cache/image_base.cpp
@@ -60,15 +60,17 @@ namespace {
 ImageBase::ImageBase(const ImageInfo& info_, GPUVAddr gpu_addr_, VAddr cpu_addr_)
     : info{info_}, guest_size_bytes{CalculateGuestSizeInBytes(info)},
       unswizzled_size_bytes{CalculateUnswizzledSizeBytes(info)},
-      converted_size_bytes{CalculateConvertedSizeBytes(info)}, gpu_addr{gpu_addr_},
-      cpu_addr{cpu_addr_}, cpu_addr_end{cpu_addr + guest_size_bytes},
-      mip_level_offsets{CalculateMipLevelOffsets(info)} {
+      converted_size_bytes{CalculateConvertedSizeBytes(info)}, scale_rating{}, scale_tick{},
+      has_scaled{}, gpu_addr{gpu_addr_}, cpu_addr{cpu_addr_},
+      cpu_addr_end{cpu_addr + guest_size_bytes}, mip_level_offsets{CalculateMipLevelOffsets(info)} {
     if (info.type == ImageType::e3D) {
         slice_offsets = CalculateSliceOffsets(info);
         slice_subresources = CalculateSliceSubresources(info);
     }
 }
 
+ImageBase::ImageBase(const NullImageParams&) {}
+
 ImageMapView::ImageMapView(GPUVAddr gpu_addr_, VAddr cpu_addr_, size_t size_, ImageId image_id_)
     : gpu_addr{gpu_addr_}, cpu_addr{cpu_addr_}, size{size_}, image_id{image_id_} {}
 
@@ -254,6 +256,8 @@ void AddImageAlias(ImageBase& lhs, ImageBase& rhs, ImageId lhs_id, ImageId rhs_i
     }
     lhs.aliased_images.push_back(std::move(lhs_alias));
     rhs.aliased_images.push_back(std::move(rhs_alias));
+    lhs.flags &= ~ImageFlagBits::IsRescalable;
+    rhs.flags &= ~ImageFlagBits::IsRescalable;
 }
 
 } // namespace VideoCommon
diff --git a/src/video_core/texture_cache/image_base.h b/src/video_core/texture_cache/image_base.h
index 0c17a791b..89c111c00 100644
--- a/src/video_core/texture_cache/image_base.h
+++ b/src/video_core/texture_cache/image_base.h
@@ -33,6 +33,11 @@ enum class ImageFlagBits : u32 {
                           ///< garbage collection priority
     Alias = 1 << 11,      ///< This image has aliases and has priority on garbage
                           ///< collection
+
+    // Rescaler
+    Rescaled = 1 << 12,
+    CheckingRescalable = 1 << 13,
+    IsRescalable = 1 << 14,
 };
 DECLARE_ENUM_FLAG_OPERATORS(ImageFlagBits)
 
@@ -43,8 +48,11 @@ struct AliasedImage {
     ImageId id;
 };
 
+struct NullImageParams {};
+
 struct ImageBase {
     explicit ImageBase(const ImageInfo& info, GPUVAddr gpu_addr, VAddr cpu_addr);
+    explicit ImageBase(const NullImageParams&);
 
     [[nodiscard]] std::optional<SubresourceBase> TryFindBase(GPUVAddr other_addr) const noexcept;
 
@@ -68,11 +76,18 @@ struct ImageBase {
     void CheckBadOverlapState();
     void CheckAliasState();
 
+    bool HasScaled() const {
+        return has_scaled;
+    }
+
     ImageInfo info;
 
     u32 guest_size_bytes = 0;
     u32 unswizzled_size_bytes = 0;
     u32 converted_size_bytes = 0;
+    u32 scale_rating = 0;
+    u64 scale_tick = 0;
+    bool has_scaled = false;
     ImageFlagBits flags = ImageFlagBits::CpuModified;
 
     GPUVAddr gpu_addr = 0;
diff --git a/src/video_core/texture_cache/image_info.cpp b/src/video_core/texture_cache/image_info.cpp
index 64fd7010a..afb94082b 100644
--- a/src/video_core/texture_cache/image_info.cpp
+++ b/src/video_core/texture_cache/image_info.cpp
@@ -16,6 +16,7 @@ namespace VideoCommon {
 using Tegra::Texture::TextureType;
 using Tegra::Texture::TICEntry;
 using VideoCore::Surface::PixelFormat;
+using VideoCore::Surface::SurfaceType;
 
 ImageInfo::ImageInfo(const TICEntry& config) noexcept {
     format = PixelFormatFromTextureInfo(config.format, config.r_type, config.g_type, config.b_type,
@@ -31,6 +32,7 @@ ImageInfo::ImageInfo(const TICEntry& config) noexcept {
             .depth = config.block_depth,
         };
     }
+    rescaleable = false;
     tile_width_spacing = config.tile_width_spacing;
     if (config.texture_type != TextureType::Texture2D &&
         config.texture_type != TextureType::Texture2DNoMipmap) {
@@ -41,6 +43,7 @@ ImageInfo::ImageInfo(const TICEntry& config) noexcept {
         ASSERT(config.BaseLayer() == 0);
         type = ImageType::e1D;
         size.width = config.Width();
+        resources.layers = 1;
         break;
     case TextureType::Texture1DArray:
         UNIMPLEMENTED_IF(config.BaseLayer() != 0);
@@ -52,12 +55,14 @@ ImageInfo::ImageInfo(const TICEntry& config) noexcept {
     case TextureType::Texture2DNoMipmap:
         ASSERT(config.Depth() == 1);
         type = config.IsPitchLinear() ? ImageType::Linear : ImageType::e2D;
+        rescaleable = !config.IsPitchLinear();
         size.width = config.Width();
         size.height = config.Height();
         resources.layers = config.BaseLayer() + 1;
         break;
     case TextureType::Texture2DArray:
         type = ImageType::e2D;
+        rescaleable = true;
         size.width = config.Width();
         size.height = config.Height();
         resources.layers = config.BaseLayer() + config.Depth();
@@ -82,10 +87,12 @@ ImageInfo::ImageInfo(const TICEntry& config) noexcept {
         size.width = config.Width();
         size.height = config.Height();
         size.depth = config.Depth();
+        resources.layers = 1;
         break;
     case TextureType::Texture1DBuffer:
         type = ImageType::Buffer;
         size.width = config.Width();
+        resources.layers = 1;
         break;
     default:
         UNREACHABLE_MSG("Invalid texture_type={}", static_cast<int>(config.texture_type.Value()));
@@ -95,12 +102,16 @@ ImageInfo::ImageInfo(const TICEntry& config) noexcept {
         // FIXME: Call this without passing *this
         layer_stride = CalculateLayerStride(*this);
         maybe_unaligned_layer_stride = CalculateLayerSize(*this);
+        rescaleable &= (block.depth == 0) && resources.levels == 1;
+        rescaleable &= size.height > 256 || GetFormatType(format) != SurfaceType::ColorTexture;
+        downscaleable = size.height > 512;
     }
 }
 
 ImageInfo::ImageInfo(const Tegra::Engines::Maxwell3D::Regs& regs, size_t index) noexcept {
     const auto& rt = regs.rt[index];
     format = VideoCore::Surface::PixelFormatFromRenderTargetFormat(rt.format);
+    rescaleable = false;
     if (rt.tile_mode.is_pitch_linear) {
         ASSERT(rt.tile_mode.is_3d == 0);
         type = ImageType::Linear;
@@ -126,6 +137,9 @@ ImageInfo::ImageInfo(const Tegra::Engines::Maxwell3D::Regs& regs, size_t index)
         type = ImageType::e3D;
         size.depth = rt.depth;
     } else {
+        rescaleable = block.depth == 0;
+        rescaleable &= size.height > 256;
+        downscaleable = size.height > 512;
         type = ImageType::e2D;
         resources.layers = rt.depth;
     }
@@ -135,6 +149,7 @@ ImageInfo::ImageInfo(const Tegra::Engines::Maxwell3D::Regs& regs) noexcept {
     format = VideoCore::Surface::PixelFormatFromDepthFormat(regs.zeta.format);
     size.width = regs.zeta_width;
     size.height = regs.zeta_height;
+    rescaleable = false;
     resources.levels = 1;
     layer_stride = regs.zeta.layer_stride * 4;
     maybe_unaligned_layer_stride = layer_stride;
@@ -153,6 +168,8 @@ ImageInfo::ImageInfo(const Tegra::Engines::Maxwell3D::Regs& regs) noexcept {
         type = ImageType::e3D;
         size.depth = regs.zeta_depth;
     } else {
+        rescaleable = block.depth == 0;
+        downscaleable = size.height > 512;
         type = ImageType::e2D;
         resources.layers = regs.zeta_depth;
     }
@@ -161,6 +178,7 @@ ImageInfo::ImageInfo(const Tegra::Engines::Maxwell3D::Regs& regs) noexcept {
 ImageInfo::ImageInfo(const Tegra::Engines::Fermi2D::Surface& config) noexcept {
     UNIMPLEMENTED_IF_MSG(config.layer != 0, "Surface layer is not zero");
     format = VideoCore::Surface::PixelFormatFromRenderTargetFormat(config.format);
+    rescaleable = false;
     if (config.linear == Tegra::Engines::Fermi2D::MemoryLayout::Pitch) {
         type = ImageType::Linear;
         size = Extent3D{
@@ -171,6 +189,7 @@ ImageInfo::ImageInfo(const Tegra::Engines::Fermi2D::Surface& config) noexcept {
         pitch = config.pitch;
     } else {
         type = config.block_depth > 0 ? ImageType::e3D : ImageType::e2D;
+
         block = Extent3D{
             .width = config.block_width,
             .height = config.block_height,
@@ -183,6 +202,9 @@ ImageInfo::ImageInfo(const Tegra::Engines::Fermi2D::Surface& config) noexcept {
             .height = config.height,
             .depth = 1,
         };
+        rescaleable = block.depth == 0;
+        rescaleable &= size.height > 256;
+        downscaleable = size.height > 512;
     }
 }
 
diff --git a/src/video_core/texture_cache/image_info.h b/src/video_core/texture_cache/image_info.h
index 5049fc36e..5932dcaba 100644
--- a/src/video_core/texture_cache/image_info.h
+++ b/src/video_core/texture_cache/image_info.h
@@ -15,7 +15,7 @@ using Tegra::Texture::TICEntry;
 using VideoCore::Surface::PixelFormat;
 
 struct ImageInfo {
-    explicit ImageInfo() = default;
+    ImageInfo() = default;
     explicit ImageInfo(const TICEntry& config) noexcept;
     explicit ImageInfo(const Tegra::Engines::Maxwell3D::Regs& regs, size_t index) noexcept;
     explicit ImageInfo(const Tegra::Engines::Maxwell3D::Regs& regs) noexcept;
@@ -33,6 +33,8 @@ struct ImageInfo {
     u32 maybe_unaligned_layer_stride = 0;
     u32 num_samples = 1;
     u32 tile_width_spacing = 0;
+    bool rescaleable = false;
+    bool downscaleable = false;
 };
 
 } // namespace VideoCommon
diff --git a/src/video_core/texture_cache/image_view_base.cpp b/src/video_core/texture_cache/image_view_base.cpp
index 450becbeb..c7b4fc231 100644
--- a/src/video_core/texture_cache/image_view_base.cpp
+++ b/src/video_core/texture_cache/image_view_base.cpp
@@ -37,14 +37,15 @@ ImageViewBase::ImageViewBase(const ImageViewInfo& info, const ImageInfo& image_i
 }
 
 ImageViewBase::ImageViewBase(const ImageInfo& info, const ImageViewInfo& view_info)
-    : format{info.format}, type{ImageViewType::Buffer}, size{
-                                                            .width = info.size.width,
-                                                            .height = 1,
-                                                            .depth = 1,
-                                                        } {
+    : image_id{NULL_IMAGE_ID}, format{info.format}, type{ImageViewType::Buffer},
+      size{
+          .width = info.size.width,
+          .height = 1,
+          .depth = 1,
+      } {
     ASSERT_MSG(view_info.type == ImageViewType::Buffer, "Expected texture buffer");
 }
 
-ImageViewBase::ImageViewBase(const NullImageParams&) {}
+ImageViewBase::ImageViewBase(const NullImageViewParams&) : image_id{NULL_IMAGE_ID} {}
 
 } // namespace VideoCommon
diff --git a/src/video_core/texture_cache/image_view_base.h b/src/video_core/texture_cache/image_view_base.h
index 903f715c5..9c24c5359 100644
--- a/src/video_core/texture_cache/image_view_base.h
+++ b/src/video_core/texture_cache/image_view_base.h
@@ -15,7 +15,7 @@ using VideoCore::Surface::PixelFormat;
 struct ImageViewInfo;
 struct ImageInfo;
 
-struct NullImageParams {};
+struct NullImageViewParams {};
 
 enum class ImageViewFlagBits : u16 {
     PreemtiveDownload = 1 << 0,
@@ -28,7 +28,7 @@ struct ImageViewBase {
     explicit ImageViewBase(const ImageViewInfo& info, const ImageInfo& image_info,
                            ImageId image_id);
     explicit ImageViewBase(const ImageInfo& info, const ImageViewInfo& view_info);
-    explicit ImageViewBase(const NullImageParams&);
+    explicit ImageViewBase(const NullImageViewParams&);
 
     [[nodiscard]] bool IsBuffer() const noexcept {
         return type == ImageViewType::Buffer;
diff --git a/src/video_core/texture_cache/texture_cache.h b/src/video_core/texture_cache/texture_cache.h
index f70c1f764..4d2874bf2 100644
--- a/src/video_core/texture_cache/texture_cache.h
+++ b/src/video_core/texture_cache/texture_cache.h
@@ -7,6 +7,7 @@
 #include <unordered_set>
 
 #include "common/alignment.h"
+#include "common/settings.h"
 #include "video_core/dirty_flags.h"
 #include "video_core/engines/kepler_compute.h"
 #include "video_core/texture_cache/image_view_base.h"
@@ -44,21 +45,22 @@ TextureCache<P>::TextureCache(Runtime& runtime_, VideoCore::RasterizerInterface&
 
     // Make sure the first index is reserved for the null resources
     // This way the null resource becomes a compile time constant
-    void(slot_image_views.insert(runtime, NullImageParams{}));
+    void(slot_images.insert(NullImageParams{}));
+    void(slot_image_views.insert(runtime, NullImageViewParams{}));
     void(slot_samplers.insert(runtime, sampler_descriptor));
 
     if constexpr (HAS_DEVICE_MEMORY_INFO) {
         const auto device_memory = runtime.GetDeviceLocalMemory();
-        const u64 possible_expected_memory = (device_memory * 3) / 10;
-        const u64 possible_critical_memory = (device_memory * 6) / 10;
-        expected_memory = std::max(possible_expected_memory, DEFAULT_EXPECTED_MEMORY);
-        critical_memory = std::max(possible_critical_memory, DEFAULT_CRITICAL_MEMORY);
+        const u64 possible_expected_memory = (device_memory * 4) / 10;
+        const u64 possible_critical_memory = (device_memory * 7) / 10;
+        expected_memory = std::max(possible_expected_memory, DEFAULT_EXPECTED_MEMORY - 256_MiB);
+        critical_memory = std::max(possible_critical_memory, DEFAULT_CRITICAL_MEMORY - 512_MiB);
         minimum_memory = 0;
     } else {
-        // on OGL we can be more conservatives as the driver takes care.
+        // On OpenGL we can be more conservatives as the driver takes care.
         expected_memory = DEFAULT_EXPECTED_MEMORY + 512_MiB;
         critical_memory = DEFAULT_CRITICAL_MEMORY + 1_GiB;
-        minimum_memory = expected_memory;
+        minimum_memory = 0;
     }
 }
 
@@ -67,7 +69,7 @@ void TextureCache<P>::RunGarbageCollector() {
     const bool high_priority_mode = total_used_memory >= expected_memory;
     const bool aggressive_mode = total_used_memory >= critical_memory;
     const u64 ticks_to_destroy = aggressive_mode ? 10ULL : high_priority_mode ? 25ULL : 100ULL;
-    size_t num_iterations = aggressive_mode ? 10000 : (high_priority_mode ? 100 : 5);
+    size_t num_iterations = aggressive_mode ? 300 : (high_priority_mode ? 50 : 10);
     const auto clean_up = [this, &num_iterations, high_priority_mode](ImageId image_id) {
         if (num_iterations == 0) {
             return true;
@@ -89,7 +91,7 @@ void TextureCache<P>::RunGarbageCollector() {
             UntrackImage(image, image_id);
         }
         UnregisterImage(image_id);
-        DeleteImage(image_id);
+        DeleteImage(image_id, image.scale_tick > frame_tick + 5);
         return false;
     };
     lru_cache.ForEachItemBelow(frame_tick - ticks_to_destroy, clean_up);
@@ -103,6 +105,7 @@ void TextureCache<P>::TickFrame() {
     sentenced_images.Tick();
     sentenced_framebuffers.Tick();
     sentenced_image_view.Tick();
+    runtime.TickFrame();
     ++frame_tick;
 }
 
@@ -122,15 +125,14 @@ void TextureCache<P>::MarkModification(ImageId id) noexcept {
 }
 
 template <class P>
-void TextureCache<P>::FillGraphicsImageViews(std::span<const u32> indices,
-                                             std::span<ImageViewId> image_view_ids) {
-    FillImageViews(graphics_image_table, graphics_image_view_ids, indices, image_view_ids);
+template <bool has_blacklists>
+void TextureCache<P>::FillGraphicsImageViews(std::span<ImageViewInOut> views) {
+    FillImageViews<has_blacklists>(graphics_image_table, graphics_image_view_ids, views);
 }
 
 template <class P>
-void TextureCache<P>::FillComputeImageViews(std::span<const u32> indices,
-                                            std::span<ImageViewId> image_view_ids) {
-    FillImageViews(compute_image_table, compute_image_view_ids, indices, image_view_ids);
+void TextureCache<P>::FillComputeImageViews(std::span<ImageViewInOut> views) {
+    FillImageViews<true>(compute_image_table, compute_image_view_ids, views);
 }
 
 template <class P>
@@ -190,6 +192,102 @@ void TextureCache<P>::SynchronizeComputeDescriptors() {
 }
 
 template <class P>
+bool TextureCache<P>::RescaleRenderTargets(bool is_clear) {
+    auto& flags = maxwell3d.dirty.flags;
+    u32 scale_rating = 0;
+    bool rescaled = false;
+    std::array<ImageId, NUM_RT> tmp_color_images{};
+    ImageId tmp_depth_image{};
+    do {
+        flags[Dirty::RenderTargets] = false;
+
+        has_deleted_images = false;
+        // Render target control is used on all render targets, so force look ups when this one is
+        // up
+        const bool force = flags[Dirty::RenderTargetControl];
+        flags[Dirty::RenderTargetControl] = false;
+
+        scale_rating = 0;
+        bool any_rescaled = false;
+        bool can_rescale = true;
+        const auto check_rescale = [&](ImageViewId view_id, ImageId& id_save) {
+            if (view_id != NULL_IMAGE_VIEW_ID && view_id != ImageViewId{}) {
+                const auto& view = slot_image_views[view_id];
+                const auto image_id = view.image_id;
+                id_save = image_id;
+                auto& image = slot_images[image_id];
+                can_rescale &= ImageCanRescale(image);
+                any_rescaled |= True(image.flags & ImageFlagBits::Rescaled) ||
+                                GetFormatType(image.info.format) != SurfaceType::ColorTexture;
+                scale_rating = std::max<u32>(scale_rating, image.scale_tick <= frame_tick
+                                                               ? image.scale_rating + 1U
+                                                               : image.scale_rating);
+            } else {
+                id_save = CORRUPT_ID;
+            }
+        };
+        for (size_t index = 0; index < NUM_RT; ++index) {
+            ImageViewId& color_buffer_id = render_targets.color_buffer_ids[index];
+            if (flags[Dirty::ColorBuffer0 + index] || force) {
+                flags[Dirty::ColorBuffer0 + index] = false;
+                BindRenderTarget(&color_buffer_id, FindColorBuffer(index, is_clear));
+            }
+            check_rescale(color_buffer_id, tmp_color_images[index]);
+        }
+        if (flags[Dirty::ZetaBuffer] || force) {
+            flags[Dirty::ZetaBuffer] = false;
+            BindRenderTarget(&render_targets.depth_buffer_id, FindDepthBuffer(is_clear));
+        }
+        check_rescale(render_targets.depth_buffer_id, tmp_depth_image);
+
+        if (can_rescale) {
+            rescaled = any_rescaled || scale_rating >= 2;
+            const auto scale_up = [this](ImageId image_id) {
+                if (image_id != CORRUPT_ID) {
+                    Image& image = slot_images[image_id];
+                    ScaleUp(image);
+                }
+            };
+            if (rescaled) {
+                for (size_t index = 0; index < NUM_RT; ++index) {
+                    scale_up(tmp_color_images[index]);
+                }
+                scale_up(tmp_depth_image);
+                scale_rating = 2;
+            }
+        } else {
+            rescaled = false;
+            const auto scale_down = [this](ImageId image_id) {
+                if (image_id != CORRUPT_ID) {
+                    Image& image = slot_images[image_id];
+                    ScaleDown(image);
+                }
+            };
+            for (size_t index = 0; index < NUM_RT; ++index) {
+                scale_down(tmp_color_images[index]);
+            }
+            scale_down(tmp_depth_image);
+            scale_rating = 1;
+        }
+    } while (has_deleted_images);
+    const auto set_rating = [this, scale_rating](ImageId image_id) {
+        if (image_id != CORRUPT_ID) {
+            Image& image = slot_images[image_id];
+            image.scale_rating = scale_rating;
+            if (image.scale_tick <= frame_tick) {
+                image.scale_tick = frame_tick + 1;
+            }
+        }
+    };
+    for (size_t index = 0; index < NUM_RT; ++index) {
+        set_rating(tmp_color_images[index]);
+    }
+    set_rating(tmp_depth_image);
+
+    return rescaled;
+}
+
+template <class P>
 void TextureCache<P>::UpdateRenderTargets(bool is_clear) {
     using namespace VideoCommon::Dirty;
     auto& flags = maxwell3d.dirty.flags;
@@ -202,24 +300,18 @@ void TextureCache<P>::UpdateRenderTargets(bool is_clear) {
         PrepareImageView(depth_buffer_id, true, is_clear && IsFullClear(depth_buffer_id));
         return;
     }
-    flags[Dirty::RenderTargets] = false;
 
-    // Render target control is used on all render targets, so force look ups when this one is up
-    const bool force = flags[Dirty::RenderTargetControl];
-    flags[Dirty::RenderTargetControl] = false;
+    const bool rescaled = RescaleRenderTargets(is_clear);
+    if (is_rescaling != rescaled) {
+        flags[Dirty::RescaleViewports] = true;
+        flags[Dirty::RescaleScissors] = true;
+        is_rescaling = rescaled;
+    }
 
     for (size_t index = 0; index < NUM_RT; ++index) {
         ImageViewId& color_buffer_id = render_targets.color_buffer_ids[index];
-        if (flags[Dirty::ColorBuffer0 + index] || force) {
-            flags[Dirty::ColorBuffer0 + index] = false;
-            BindRenderTarget(&color_buffer_id, FindColorBuffer(index, is_clear));
-        }
         PrepareImageView(color_buffer_id, true, is_clear && IsFullClear(color_buffer_id));
     }
-    if (flags[Dirty::ZetaBuffer] || force) {
-        flags[Dirty::ZetaBuffer] = false;
-        BindRenderTarget(&render_targets.depth_buffer_id, FindDepthBuffer(is_clear));
-    }
     const ImageViewId depth_buffer_id = render_targets.depth_buffer_id;
 
     PrepareImageView(depth_buffer_id, true, is_clear && IsFullClear(depth_buffer_id));
@@ -227,9 +319,15 @@ void TextureCache<P>::UpdateRenderTargets(bool is_clear) {
     for (size_t index = 0; index < NUM_RT; ++index) {
         render_targets.draw_buffers[index] = static_cast<u8>(maxwell3d.regs.rt_control.Map(index));
     }
+    u32 up_scale = 1;
+    u32 down_shift = 0;
+    if (is_rescaling) {
+        up_scale = Settings::values.resolution_info.up_scale;
+        down_shift = Settings::values.resolution_info.down_shift;
+    }
     render_targets.size = Extent2D{
-        maxwell3d.regs.render_area.width,
-        maxwell3d.regs.render_area.height,
+        (maxwell3d.regs.render_area.width * up_scale) >> down_shift,
+        (maxwell3d.regs.render_area.height * up_scale) >> down_shift,
     };
 
     flags[Dirty::DepthBiasGlobal] = true;
@@ -241,17 +339,28 @@ typename P::Framebuffer* TextureCache<P>::GetFramebuffer() {
 }
 
 template <class P>
+template <bool has_blacklists>
 void TextureCache<P>::FillImageViews(DescriptorTable<TICEntry>& table,
                                      std::span<ImageViewId> cached_image_view_ids,
-                                     std::span<const u32> indices,
-                                     std::span<ImageViewId> image_view_ids) {
-    ASSERT(indices.size() <= image_view_ids.size());
+                                     std::span<ImageViewInOut> views) {
+    bool has_blacklisted;
     do {
         has_deleted_images = false;
-        std::ranges::transform(indices, image_view_ids.begin(), [&](u32 index) {
-            return VisitImageView(table, cached_image_view_ids, index);
-        });
-    } while (has_deleted_images);
+        if constexpr (has_blacklists) {
+            has_blacklisted = false;
+        }
+        for (ImageViewInOut& view : views) {
+            view.id = VisitImageView(table, cached_image_view_ids, view.index);
+            if constexpr (has_blacklists) {
+                if (view.blacklist && view.id != NULL_IMAGE_VIEW_ID) {
+                    const ImageViewBase& image_view{slot_image_views[view.id]};
+                    auto& image = slot_images[image_view.image_id];
+                    has_blacklisted |= ScaleDown(image);
+                    image.scale_rating = 0;
+                }
+            }
+        }
+    } while (has_deleted_images || (has_blacklists && has_blacklisted));
 }
 
 template <class P>
@@ -369,8 +478,43 @@ void TextureCache<P>::BlitImage(const Tegra::Engines::Fermi2D::Surface& dst,
     PrepareImage(src_id, false, false);
     PrepareImage(dst_id, true, false);
 
-    ImageBase& dst_image = slot_images[dst_id];
-    const ImageBase& src_image = slot_images[src_id];
+    Image& dst_image = slot_images[dst_id];
+    Image& src_image = slot_images[src_id];
+    bool is_src_rescaled = True(src_image.flags & ImageFlagBits::Rescaled);
+    bool is_dst_rescaled = True(dst_image.flags & ImageFlagBits::Rescaled);
+
+    const bool is_resolve = src_image.info.num_samples != 1 && dst_image.info.num_samples == 1;
+    if (is_src_rescaled != is_dst_rescaled) {
+        if (ImageCanRescale(src_image)) {
+            ScaleUp(src_image);
+            is_src_rescaled = True(src_image.flags & ImageFlagBits::Rescaled);
+            if (is_resolve) {
+                dst_image.info.rescaleable = true;
+                for (const auto& alias : dst_image.aliased_images) {
+                    Image& other_image = slot_images[alias.id];
+                    other_image.info.rescaleable = true;
+                }
+            }
+        }
+        if (ImageCanRescale(dst_image)) {
+            ScaleUp(dst_image);
+            is_dst_rescaled = True(dst_image.flags & ImageFlagBits::Rescaled);
+        }
+    }
+    if (is_resolve && (is_src_rescaled != is_dst_rescaled)) {
+        // A resolve requires both images to be the same dimensions. Resize down if needed.
+        ScaleDown(src_image);
+        ScaleDown(dst_image);
+        is_src_rescaled = True(src_image.flags & ImageFlagBits::Rescaled);
+        is_dst_rescaled = True(dst_image.flags & ImageFlagBits::Rescaled);
+    }
+    const auto& resolution = Settings::values.resolution_info;
+    const auto scale_region = [&](Region2D& region) {
+        region.start.x = resolution.ScaleUp(region.start.x);
+        region.start.y = resolution.ScaleUp(region.start.y);
+        region.end.x = resolution.ScaleUp(region.end.x);
+        region.end.y = resolution.ScaleUp(region.end.y);
+    };
 
     // TODO: Deduplicate
     const std::optional src_base = src_image.TryFindBase(src.Address());
@@ -378,20 +522,26 @@ void TextureCache<P>::BlitImage(const Tegra::Engines::Fermi2D::Surface& dst,
     const ImageViewInfo src_view_info(ImageViewType::e2D, images.src_format, src_range);
     const auto [src_framebuffer_id, src_view_id] = RenderTargetFromImage(src_id, src_view_info);
     const auto [src_samples_x, src_samples_y] = SamplesLog2(src_image.info.num_samples);
-    const Region2D src_region{
+    Region2D src_region{
         Offset2D{.x = copy.src_x0 >> src_samples_x, .y = copy.src_y0 >> src_samples_y},
         Offset2D{.x = copy.src_x1 >> src_samples_x, .y = copy.src_y1 >> src_samples_y},
     };
+    if (is_src_rescaled) {
+        scale_region(src_region);
+    }
 
     const std::optional dst_base = dst_image.TryFindBase(dst.Address());
     const SubresourceRange dst_range{.base = dst_base.value(), .extent = {1, 1}};
     const ImageViewInfo dst_view_info(ImageViewType::e2D, images.dst_format, dst_range);
     const auto [dst_framebuffer_id, dst_view_id] = RenderTargetFromImage(dst_id, dst_view_info);
     const auto [dst_samples_x, dst_samples_y] = SamplesLog2(dst_image.info.num_samples);
-    const Region2D dst_region{
+    Region2D dst_region{
         Offset2D{.x = copy.dst_x0 >> dst_samples_x, .y = copy.dst_y0 >> dst_samples_y},
         Offset2D{.x = copy.dst_x1 >> dst_samples_x, .y = copy.dst_y1 >> dst_samples_y},
     };
+    if (is_dst_rescaled) {
+        scale_region(dst_region);
+    }
 
     // Always call this after src_framebuffer_id was queried, as the address might be invalidated.
     Framebuffer* const dst_framebuffer = &slot_framebuffers[dst_framebuffer_id];
@@ -487,6 +637,20 @@ void TextureCache<P>::PopAsyncFlushes() {
 }
 
 template <class P>
+bool TextureCache<P>::IsRescaling() const noexcept {
+    return is_rescaling;
+}
+
+template <class P>
+bool TextureCache<P>::IsRescaling(const ImageViewBase& image_view) const noexcept {
+    if (image_view.type == ImageViewType::Buffer) {
+        return false;
+    }
+    const ImageBase& image = slot_images[image_view.image_id];
+    return True(image.flags & ImageFlagBits::Rescaled);
+}
+
+template <class P>
 bool TextureCache<P>::IsRegionGpuModified(VAddr addr, size_t size) {
     bool is_modified = false;
     ForEachImageInRegion(addr, size, [&is_modified](ImageId, ImageBase& image) {
@@ -624,6 +788,105 @@ ImageId TextureCache<P>::FindImage(const ImageInfo& info, GPUVAddr gpu_addr,
 }
 
 template <class P>
+bool TextureCache<P>::ImageCanRescale(ImageBase& image) {
+    if (!image.info.rescaleable) {
+        return false;
+    }
+    if (Settings::values.resolution_info.downscale && !image.info.downscaleable) {
+        return false;
+    }
+    if (True(image.flags & (ImageFlagBits::Rescaled | ImageFlagBits::CheckingRescalable))) {
+        return true;
+    }
+    if (True(image.flags & ImageFlagBits::IsRescalable)) {
+        return true;
+    }
+    image.flags |= ImageFlagBits::CheckingRescalable;
+    for (const auto& alias : image.aliased_images) {
+        Image& other_image = slot_images[alias.id];
+        if (!ImageCanRescale(other_image)) {
+            image.flags &= ~ImageFlagBits::CheckingRescalable;
+            return false;
+        }
+    }
+    image.flags &= ~ImageFlagBits::CheckingRescalable;
+    image.flags |= ImageFlagBits::IsRescalable;
+    return true;
+}
+
+template <class P>
+void TextureCache<P>::InvalidateScale(Image& image) {
+    if (image.scale_tick <= frame_tick) {
+        image.scale_tick = frame_tick + 1;
+    }
+    const std::span<const ImageViewId> image_view_ids = image.image_view_ids;
+    auto& dirty = maxwell3d.dirty.flags;
+    dirty[Dirty::RenderTargets] = true;
+    dirty[Dirty::ZetaBuffer] = true;
+    for (size_t rt = 0; rt < NUM_RT; ++rt) {
+        dirty[Dirty::ColorBuffer0 + rt] = true;
+    }
+    for (const ImageViewId image_view_id : image_view_ids) {
+        std::ranges::replace(render_targets.color_buffer_ids, image_view_id, ImageViewId{});
+        if (render_targets.depth_buffer_id == image_view_id) {
+            render_targets.depth_buffer_id = ImageViewId{};
+        }
+    }
+    RemoveImageViewReferences(image_view_ids);
+    RemoveFramebuffers(image_view_ids);
+    for (const ImageViewId image_view_id : image_view_ids) {
+        sentenced_image_view.Push(std::move(slot_image_views[image_view_id]));
+        slot_image_views.erase(image_view_id);
+    }
+    image.image_view_ids.clear();
+    image.image_view_infos.clear();
+    if constexpr (ENABLE_VALIDATION) {
+        std::ranges::fill(graphics_image_view_ids, CORRUPT_ID);
+        std::ranges::fill(compute_image_view_ids, CORRUPT_ID);
+    }
+    graphics_image_table.Invalidate();
+    compute_image_table.Invalidate();
+    has_deleted_images = true;
+}
+
+template <class P>
+u64 TextureCache<P>::GetScaledImageSizeBytes(ImageBase& image) {
+    const u64 scale_up = static_cast<u64>(Settings::values.resolution_info.up_scale *
+                                          Settings::values.resolution_info.up_scale);
+    const u64 down_shift = static_cast<u64>(Settings::values.resolution_info.down_shift +
+                                            Settings::values.resolution_info.down_shift);
+    const u64 image_size_bytes =
+        static_cast<u64>(std::max(image.guest_size_bytes, image.unswizzled_size_bytes));
+    const u64 tentative_size = (image_size_bytes * scale_up) >> down_shift;
+    const u64 fitted_size = Common::AlignUp(tentative_size, 1024);
+    return fitted_size;
+}
+
+template <class P>
+bool TextureCache<P>::ScaleUp(Image& image) {
+    const bool has_copy = image.HasScaled();
+    const bool rescaled = image.ScaleUp();
+    if (!rescaled) {
+        return false;
+    }
+    if (!has_copy) {
+        total_used_memory += GetScaledImageSizeBytes(image);
+    }
+    InvalidateScale(image);
+    return true;
+}
+
+template <class P>
+bool TextureCache<P>::ScaleDown(Image& image) {
+    const bool rescaled = image.ScaleDown();
+    if (!rescaled) {
+        return false;
+    }
+    InvalidateScale(image);
+    return true;
+}
+
+template <class P>
 ImageId TextureCache<P>::InsertImage(const ImageInfo& info, GPUVAddr gpu_addr,
                                      RelaxedOptions options) {
     std::optional<VAddr> cpu_addr = gpu_memory.GpuToCpuAddress(gpu_addr);
@@ -660,12 +923,18 @@ ImageId TextureCache<P>::JoinImages(const ImageInfo& info, GPUVAddr gpu_addr, VA
     std::vector<ImageId> right_aliased_ids;
     std::unordered_set<ImageId> ignore_textures;
     std::vector<ImageId> bad_overlap_ids;
+    std::vector<ImageId> all_siblings;
+    const bool this_is_linear = info.type == ImageType::Linear;
     const auto region_check = [&](ImageId overlap_id, ImageBase& overlap) {
         if (True(overlap.flags & ImageFlagBits::Remapped)) {
             ignore_textures.insert(overlap_id);
             return;
         }
-        if (info.type == ImageType::Linear) {
+        const bool overlap_is_linear = overlap.info.type == ImageType::Linear;
+        if (this_is_linear != overlap_is_linear) {
+            return;
+        }
+        if (this_is_linear && overlap_is_linear) {
             if (info.pitch == overlap.info.pitch && gpu_addr == overlap.gpu_addr) {
                 // Alias linear images with the same pitch
                 left_aliased_ids.push_back(overlap_id);
@@ -681,6 +950,7 @@ ImageId TextureCache<P>::JoinImages(const ImageInfo& info, GPUVAddr gpu_addr, VA
             cpu_addr = solution->cpu_addr;
             new_info.resources = solution->resources;
             overlap_ids.push_back(overlap_id);
+            all_siblings.push_back(overlap_id);
             return;
         }
         static constexpr auto options = RelaxedOptions::Size | RelaxedOptions::Format;
@@ -688,10 +958,12 @@ ImageId TextureCache<P>::JoinImages(const ImageInfo& info, GPUVAddr gpu_addr, VA
         if (IsSubresource(new_info, overlap, gpu_addr, options, broken_views, native_bgr)) {
             left_aliased_ids.push_back(overlap_id);
             overlap.flags |= ImageFlagBits::Alias;
+            all_siblings.push_back(overlap_id);
         } else if (IsSubresource(overlap.info, new_image_base, overlap.gpu_addr, options,
                                  broken_views, native_bgr)) {
             right_aliased_ids.push_back(overlap_id);
             overlap.flags |= ImageFlagBits::Alias;
+            all_siblings.push_back(overlap_id);
         } else {
             bad_overlap_ids.push_back(overlap_id);
             overlap.flags |= ImageFlagBits::BadOverlap;
@@ -709,6 +981,32 @@ ImageId TextureCache<P>::JoinImages(const ImageInfo& info, GPUVAddr gpu_addr, VA
         }
     };
     ForEachSparseImageInRegion(gpu_addr, size_bytes, region_check_gpu);
+
+    bool can_rescale = info.rescaleable;
+    bool any_rescaled = false;
+    for (const ImageId sibling_id : all_siblings) {
+        if (!can_rescale) {
+            break;
+        }
+        Image& sibling = slot_images[sibling_id];
+        can_rescale &= ImageCanRescale(sibling);
+        any_rescaled |= True(sibling.flags & ImageFlagBits::Rescaled);
+    }
+
+    can_rescale &= any_rescaled;
+
+    if (can_rescale) {
+        for (const ImageId sibling_id : all_siblings) {
+            Image& sibling = slot_images[sibling_id];
+            ScaleUp(sibling);
+        }
+    } else {
+        for (const ImageId sibling_id : all_siblings) {
+            Image& sibling = slot_images[sibling_id];
+            ScaleDown(sibling);
+        }
+    }
+
     const ImageId new_image_id = slot_images.insert(runtime, new_info, gpu_addr, cpu_addr);
     Image& new_image = slot_images[new_image_id];
 
@@ -731,14 +1029,23 @@ ImageId TextureCache<P>::JoinImages(const ImageInfo& info, GPUVAddr gpu_addr, VA
     // TODO: Only upload what we need
     RefreshContents(new_image, new_image_id);
 
+    if (can_rescale) {
+        ScaleUp(new_image);
+    } else {
+        ScaleDown(new_image);
+    }
+
     for (const ImageId overlap_id : overlap_ids) {
         Image& overlap = slot_images[overlap_id];
         if (overlap.info.num_samples != new_image.info.num_samples) {
             LOG_WARNING(HW_GPU, "Copying between images with different samples is not implemented");
         } else {
+            const auto& resolution = Settings::values.resolution_info;
             const SubresourceBase base = new_image.TryFindBase(overlap.gpu_addr).value();
-            const auto copies = MakeShrinkImageCopies(new_info, overlap.info, base);
-            runtime.CopyImage(new_image, overlap, copies);
+            const u32 up_scale = can_rescale ? resolution.up_scale : 1;
+            const u32 down_shift = can_rescale ? resolution.down_shift : 0;
+            auto copies = MakeShrinkImageCopies(new_info, overlap.info, base, up_scale, down_shift);
+            runtime.CopyImage(new_image, overlap, std::move(copies));
         }
         if (True(overlap.flags & ImageFlagBits::Tracked)) {
             UntrackImage(overlap, overlap_id);
@@ -1083,13 +1390,6 @@ void TextureCache<P>::UnregisterImage(ImageId image_id) {
                "Trying to unregister an already registered image");
     image.flags &= ~ImageFlagBits::Registered;
     image.flags &= ~ImageFlagBits::BadOverlap;
-    u64 tentative_size = std::max(image.guest_size_bytes, image.unswizzled_size_bytes);
-    if ((IsPixelFormatASTC(image.info.format) &&
-         True(image.flags & ImageFlagBits::AcceleratedUpload)) ||
-        True(image.flags & ImageFlagBits::Converted)) {
-        tentative_size = EstimatedDecompressedSize(tentative_size, image.info.format);
-    }
-    total_used_memory -= Common::AlignUp(tentative_size, 1024);
     lru_cache.Free(image.lru_index);
     const auto& clear_page_table =
         [this, image_id](
@@ -1213,8 +1513,18 @@ void TextureCache<P>::UntrackImage(ImageBase& image, ImageId image_id) {
 }
 
 template <class P>
-void TextureCache<P>::DeleteImage(ImageId image_id) {
+void TextureCache<P>::DeleteImage(ImageId image_id, bool immediate_delete) {
     ImageBase& image = slot_images[image_id];
+    if (image.HasScaled()) {
+        total_used_memory -= GetScaledImageSizeBytes(image);
+    }
+    u64 tentative_size = std::max(image.guest_size_bytes, image.unswizzled_size_bytes);
+    if ((IsPixelFormatASTC(image.info.format) &&
+         True(image.flags & ImageFlagBits::AcceleratedUpload)) ||
+        True(image.flags & ImageFlagBits::Converted)) {
+        tentative_size = EstimatedDecompressedSize(tentative_size, image.info.format);
+    }
+    total_used_memory -= Common::AlignUp(tentative_size, 1024);
     const GPUVAddr gpu_addr = image.gpu_addr;
     const auto alloc_it = image_allocs_table.find(gpu_addr);
     if (alloc_it == image_allocs_table.end()) {
@@ -1269,10 +1579,14 @@ void TextureCache<P>::DeleteImage(ImageId image_id) {
                    num_removed_overlaps);
     }
     for (const ImageViewId image_view_id : image_view_ids) {
-        sentenced_image_view.Push(std::move(slot_image_views[image_view_id]));
+        if (!immediate_delete) {
+            sentenced_image_view.Push(std::move(slot_image_views[image_view_id]));
+        }
         slot_image_views.erase(image_view_id);
     }
-    sentenced_images.Push(std::move(slot_images[image_id]));
+    if (!immediate_delete) {
+        sentenced_images.Push(std::move(slot_images[image_id]));
+    }
     slot_images.erase(image_id);
 
     alloc_images.erase(alloc_image_it);
@@ -1306,6 +1620,9 @@ void TextureCache<P>::RemoveFramebuffers(std::span<const ImageViewId> removed_vi
     auto it = framebuffers.begin();
     while (it != framebuffers.end()) {
         if (it->first.Contains(removed_views)) {
+            auto framebuffer_id = it->second;
+            ASSERT(framebuffer_id);
+            sentenced_framebuffers.Push(std::move(slot_framebuffers[framebuffer_id]));
             it = framebuffers.erase(it);
         } else {
             ++it;
@@ -1322,26 +1639,60 @@ void TextureCache<P>::MarkModification(ImageBase& image) noexcept {
 template <class P>
 void TextureCache<P>::SynchronizeAliases(ImageId image_id) {
     boost::container::small_vector<const AliasedImage*, 1> aliased_images;
-    ImageBase& image = slot_images[image_id];
+    Image& image = slot_images[image_id];
+    bool any_rescaled = True(image.flags & ImageFlagBits::Rescaled);
     u64 most_recent_tick = image.modification_tick;
     for (const AliasedImage& aliased : image.aliased_images) {
         ImageBase& aliased_image = slot_images[aliased.id];
         if (image.modification_tick < aliased_image.modification_tick) {
             most_recent_tick = std::max(most_recent_tick, aliased_image.modification_tick);
             aliased_images.push_back(&aliased);
+            any_rescaled |= True(aliased_image.flags & ImageFlagBits::Rescaled);
         }
     }
     if (aliased_images.empty()) {
         return;
     }
+    const bool can_rescale = ImageCanRescale(image);
+    if (any_rescaled) {
+        if (can_rescale) {
+            ScaleUp(image);
+        } else {
+            ScaleDown(image);
+        }
+    }
     image.modification_tick = most_recent_tick;
     std::ranges::sort(aliased_images, [this](const AliasedImage* lhs, const AliasedImage* rhs) {
         const ImageBase& lhs_image = slot_images[lhs->id];
         const ImageBase& rhs_image = slot_images[rhs->id];
         return lhs_image.modification_tick < rhs_image.modification_tick;
     });
+    const auto& resolution = Settings::values.resolution_info;
     for (const AliasedImage* const aliased : aliased_images) {
-        CopyImage(image_id, aliased->id, aliased->copies);
+        if (!resolution.active | !any_rescaled) {
+            CopyImage(image_id, aliased->id, aliased->copies);
+            continue;
+        }
+        Image& aliased_image = slot_images[aliased->id];
+        if (!can_rescale) {
+            ScaleDown(aliased_image);
+            CopyImage(image_id, aliased->id, aliased->copies);
+            continue;
+        }
+        ScaleUp(aliased_image);
+
+        const bool both_2d{image.info.type == ImageType::e2D &&
+                           aliased_image.info.type == ImageType::e2D};
+        auto copies = aliased->copies;
+        for (auto copy : copies) {
+            copy.extent.width = std::max<u32>(
+                (copy.extent.width * resolution.up_scale) >> resolution.down_shift, 1);
+            if (both_2d) {
+                copy.extent.height = std::max<u32>(
+                    (copy.extent.height * resolution.up_scale) >> resolution.down_shift, 1);
+            }
+        }
+        CopyImage(image_id, aliased->id, copies);
     }
 }
 
@@ -1377,9 +1728,25 @@ void TextureCache<P>::PrepareImageView(ImageViewId image_view_id, bool is_modifi
 }
 
 template <class P>
-void TextureCache<P>::CopyImage(ImageId dst_id, ImageId src_id, std::span<const ImageCopy> copies) {
+void TextureCache<P>::CopyImage(ImageId dst_id, ImageId src_id, std::vector<ImageCopy> copies) {
     Image& dst = slot_images[dst_id];
     Image& src = slot_images[src_id];
+    const bool is_rescaled = True(src.flags & ImageFlagBits::Rescaled);
+    if (is_rescaled) {
+        ASSERT(True(dst.flags & ImageFlagBits::Rescaled));
+        const bool both_2d{src.info.type == ImageType::e2D && dst.info.type == ImageType::e2D};
+        const auto& resolution = Settings::values.resolution_info;
+        for (auto& copy : copies) {
+            copy.src_offset.x = resolution.ScaleUp(copy.src_offset.x);
+            copy.dst_offset.x = resolution.ScaleUp(copy.dst_offset.x);
+            copy.extent.width = resolution.ScaleUp(copy.extent.width);
+            if (both_2d) {
+                copy.src_offset.y = resolution.ScaleUp(copy.src_offset.y);
+                copy.dst_offset.y = resolution.ScaleUp(copy.dst_offset.y);
+                copy.extent.height = resolution.ScaleUp(copy.extent.height);
+            }
+        }
+    }
     const auto dst_format_type = GetFormatType(dst.info.format);
     const auto src_format_type = GetFormatType(src.info.format);
     if (src_format_type == dst_format_type) {
@@ -1424,7 +1791,7 @@ void TextureCache<P>::CopyImage(ImageId dst_id, ImageId src_id, std::span<const
         };
         UNIMPLEMENTED_IF(copy.extent != expected_size);
 
-        runtime.ConvertImage(dst_framebuffer, dst_view, src_view);
+        runtime.ConvertImage(dst_framebuffer, dst_view, src_view, is_rescaled);
     }
 }
 
@@ -1433,8 +1800,8 @@ void TextureCache<P>::BindRenderTarget(ImageViewId* old_id, ImageViewId new_id)
     if (*old_id == new_id) {
         return;
     }
-    if (*old_id) {
-        const ImageViewBase& old_view = slot_image_views[*old_id];
+    if (new_id) {
+        const ImageViewBase& old_view = slot_image_views[new_id];
         if (True(old_view.flags & ImageViewFlagBits::PreemtiveDownload)) {
             uncommitted_downloads.push_back(old_view.image_id);
         }
@@ -1447,10 +1814,18 @@ std::pair<FramebufferId, ImageViewId> TextureCache<P>::RenderTargetFromImage(
     ImageId image_id, const ImageViewInfo& view_info) {
     const ImageViewId view_id = FindOrEmplaceImageView(image_id, view_info);
     const ImageBase& image = slot_images[image_id];
+    const bool is_rescaled = True(image.flags & ImageFlagBits::Rescaled);
     const bool is_color = GetFormatType(image.info.format) == SurfaceType::ColorTexture;
     const ImageViewId color_view_id = is_color ? view_id : ImageViewId{};
     const ImageViewId depth_view_id = is_color ? ImageViewId{} : view_id;
-    const Extent3D extent = MipSize(image.info.size, view_info.range.base.level);
+    Extent3D extent = MipSize(image.info.size, view_info.range.base.level);
+    if (is_rescaled) {
+        const auto& resolution = Settings::values.resolution_info;
+        extent.width = resolution.ScaleUp(extent.width);
+        if (image.info.type == ImageType::e2D) {
+            extent.height = resolution.ScaleUp(extent.height);
+        }
+    }
     const u32 num_samples = image.info.num_samples;
     const auto [samples_x, samples_y] = SamplesLog2(num_samples);
     const FramebufferId framebuffer_id = GetFramebufferId(RenderTargets{
diff --git a/src/video_core/texture_cache/texture_cache_base.h b/src/video_core/texture_cache/texture_cache_base.h
index 2d1893c1c..643ad811c 100644
--- a/src/video_core/texture_cache/texture_cache_base.h
+++ b/src/video_core/texture_cache/texture_cache_base.h
@@ -21,6 +21,7 @@
 #include "video_core/texture_cache/descriptor_table.h"
 #include "video_core/texture_cache/image_base.h"
 #include "video_core/texture_cache/image_info.h"
+#include "video_core/texture_cache/image_view_base.h"
 #include "video_core/texture_cache/image_view_info.h"
 #include "video_core/texture_cache/render_targets.h"
 #include "video_core/texture_cache/slot_vector.h"
@@ -39,6 +40,12 @@ using VideoCore::Surface::PixelFormatFromDepthFormat;
 using VideoCore::Surface::PixelFormatFromRenderTargetFormat;
 using namespace Common::Literals;
 
+struct ImageViewInOut {
+    u32 index{};
+    bool blacklist{};
+    ImageViewId id{};
+};
+
 template <class P>
 class TextureCache {
     /// Address shift for caching images into a hash table
@@ -53,11 +60,6 @@ class TextureCache {
     /// True when the API can provide info about the memory of the device.
     static constexpr bool HAS_DEVICE_MEMORY_INFO = P::HAS_DEVICE_MEMORY_INFO;
 
-    /// Image view ID for null descriptors
-    static constexpr ImageViewId NULL_IMAGE_VIEW_ID{0};
-    /// Sampler ID for bugged sampler ids
-    static constexpr SamplerId NULL_SAMPLER_ID{0};
-
     static constexpr u64 DEFAULT_EXPECTED_MEMORY = 1_GiB;
     static constexpr u64 DEFAULT_CRITICAL_MEMORY = 2_GiB;
 
@@ -99,11 +101,11 @@ public:
     void MarkModification(ImageId id) noexcept;
 
     /// Fill image_view_ids with the graphics images in indices
-    void FillGraphicsImageViews(std::span<const u32> indices,
-                                std::span<ImageViewId> image_view_ids);
+    template <bool has_blacklists>
+    void FillGraphicsImageViews(std::span<ImageViewInOut> views);
 
     /// Fill image_view_ids with the compute images in indices
-    void FillComputeImageViews(std::span<const u32> indices, std::span<ImageViewId> image_view_ids);
+    void FillComputeImageViews(std::span<ImageViewInOut> views);
 
     /// Get the sampler from the graphics descriptor table in the specified index
     Sampler* GetGraphicsSampler(u32 index);
@@ -117,6 +119,11 @@ public:
     /// Refresh the state for compute image view and sampler descriptors
     void SynchronizeComputeDescriptors();
 
+    /// Updates the Render Targets if they can be rescaled
+    /// @param is_clear True when the render targets are being used for clears
+    /// @retval True if the Render Targets have been rescaled.
+    bool RescaleRenderTargets(bool is_clear);
+
     /// Update bound render targets and upload memory if necessary
     /// @param is_clear True when the render targets are being used for clears
     void UpdateRenderTargets(bool is_clear);
@@ -160,6 +167,10 @@ public:
     /// Return true when a CPU region is modified from the GPU
     [[nodiscard]] bool IsRegionGpuModified(VAddr addr, size_t size);
 
+    [[nodiscard]] bool IsRescaling() const noexcept;
+
+    [[nodiscard]] bool IsRescaling(const ImageViewBase& image_view) const noexcept;
+
     std::mutex mutex;
 
 private:
@@ -198,9 +209,10 @@ private:
     void RunGarbageCollector();
 
     /// Fills image_view_ids in the image views in indices
+    template <bool has_blacklists>
     void FillImageViews(DescriptorTable<TICEntry>& table,
-                        std::span<ImageViewId> cached_image_view_ids, std::span<const u32> indices,
-                        std::span<ImageViewId> image_view_ids);
+                        std::span<ImageViewId> cached_image_view_ids,
+                        std::span<ImageViewInOut> views);
 
     /// Find or create an image view in the guest descriptor table
     ImageViewId VisitImageView(DescriptorTable<TICEntry>& table,
@@ -285,7 +297,7 @@ private:
     void UntrackImage(ImageBase& image, ImageId image_id);
 
     /// Delete image from the cache
-    void DeleteImage(ImageId image);
+    void DeleteImage(ImageId image, bool immediate_delete = false);
 
     /// Remove image views references from the cache
     void RemoveImageViewReferences(std::span<const ImageViewId> removed_views);
@@ -306,7 +318,7 @@ private:
     void PrepareImageView(ImageViewId image_view_id, bool is_modification, bool invalidate);
 
     /// Execute copies from one image to the other, even if they are incompatible
-    void CopyImage(ImageId dst_id, ImageId src_id, std::span<const ImageCopy> copies);
+    void CopyImage(ImageId dst_id, ImageId src_id, std::vector<ImageCopy> copies);
 
     /// Bind an image view as render target, downloading resources preemtively if needed
     void BindRenderTarget(ImageViewId* old_id, ImageViewId new_id);
@@ -318,6 +330,12 @@ private:
     /// Returns true if the current clear parameters clear the whole image of a given image view
     [[nodiscard]] bool IsFullClear(ImageViewId id);
 
+    bool ImageCanRescale(ImageBase& image);
+    void InvalidateScale(Image& image);
+    bool ScaleUp(Image& image);
+    bool ScaleDown(Image& image);
+    u64 GetScaledImageSizeBytes(ImageBase& image);
+
     Runtime& runtime;
     VideoCore::RasterizerInterface& rasterizer;
     Tegra::Engines::Maxwell3D& maxwell3d;
@@ -349,6 +367,7 @@ private:
     VAddr virtual_invalid_space{};
 
     bool has_deleted_images = false;
+    bool is_rescaling = false;
     u64 total_used_memory = 0;
     u64 minimum_memory;
     u64 expected_memory;
diff --git a/src/video_core/texture_cache/types.h b/src/video_core/texture_cache/types.h
index 47a11cb2f..5c274abdf 100644
--- a/src/video_core/texture_cache/types.h
+++ b/src/video_core/texture_cache/types.h
@@ -22,6 +22,13 @@ using ImageAllocId = SlotId;
 using SamplerId = SlotId;
 using FramebufferId = SlotId;
 
+/// Fake image ID for null image views
+constexpr ImageId NULL_IMAGE_ID{0};
+/// Image view ID for null descriptors
+constexpr ImageViewId NULL_IMAGE_VIEW_ID{0};
+/// Sampler ID for bugged sampler ids
+constexpr SamplerId NULL_SAMPLER_ID{0};
+
 enum class ImageType : u32 {
     e1D,
     e2D,
diff --git a/src/video_core/texture_cache/util.cpp b/src/video_core/texture_cache/util.cpp
index 59cf2f561..ddc9fb13a 100644
--- a/src/video_core/texture_cache/util.cpp
+++ b/src/video_core/texture_cache/util.cpp
@@ -723,7 +723,7 @@ ImageViewType RenderTargetImageViewType(const ImageInfo& info) noexcept {
 }
 
 std::vector<ImageCopy> MakeShrinkImageCopies(const ImageInfo& dst, const ImageInfo& src,
-                                             SubresourceBase base) {
+                                             SubresourceBase base, u32 up_scale, u32 down_shift) {
     ASSERT(dst.resources.levels >= src.resources.levels);
     ASSERT(dst.num_samples == src.num_samples);
 
@@ -732,7 +732,7 @@ std::vector<ImageCopy> MakeShrinkImageCopies(const ImageInfo& dst, const ImageIn
         ASSERT(src.type == ImageType::e3D);
         ASSERT(src.resources.levels == 1);
     }
-
+    const bool both_2d{src.type == ImageType::e2D && dst.type == ImageType::e2D};
     std::vector<ImageCopy> copies;
     copies.reserve(src.resources.levels);
     for (s32 level = 0; level < src.resources.levels; ++level) {
@@ -762,6 +762,10 @@ std::vector<ImageCopy> MakeShrinkImageCopies(const ImageInfo& dst, const ImageIn
         if (is_dst_3d) {
             copy.extent.depth = src.size.depth;
         }
+        copy.extent.width = std::max<u32>((copy.extent.width * up_scale) >> down_shift, 1);
+        if (both_2d) {
+            copy.extent.height = std::max<u32>((copy.extent.height * up_scale) >> down_shift, 1);
+        }
     }
     return copies;
 }
@@ -1153,10 +1157,10 @@ void DeduceBlitImages(ImageInfo& dst_info, ImageInfo& src_info, const ImageBase*
     if (dst && GetFormatType(dst->info.format) != SurfaceType::ColorTexture) {
         dst_info.format = dst->info.format;
     }
-    if (!dst && src && GetFormatType(src->info.format) != SurfaceType::ColorTexture) {
+    if (src && GetFormatType(src->info.format) != SurfaceType::ColorTexture) {
         dst_info.format = src->info.format;
     }
-    if (!src && dst && GetFormatType(dst->info.format) != SurfaceType::ColorTexture) {
+    if (dst && GetFormatType(dst->info.format) != SurfaceType::ColorTexture) {
         src_info.format = dst->info.format;
     }
 }
diff --git a/src/video_core/texture_cache/util.h b/src/video_core/texture_cache/util.h
index 766502908..7af52de2e 100644
--- a/src/video_core/texture_cache/util.h
+++ b/src/video_core/texture_cache/util.h
@@ -55,7 +55,8 @@ struct OverlapResult {
 
 [[nodiscard]] std::vector<ImageCopy> MakeShrinkImageCopies(const ImageInfo& dst,
                                                            const ImageInfo& src,
-                                                           SubresourceBase base);
+                                                           SubresourceBase base, u32 up_scale = 1,
+                                                           u32 down_shift = 0);
 
 [[nodiscard]] bool IsValidEntry(const Tegra::MemoryManager& gpu_memory, const TICEntry& config);
 
diff --git a/src/video_core/textures/texture.cpp b/src/video_core/textures/texture.cpp
index a552543ed..06954963d 100644
--- a/src/video_core/textures/texture.cpp
+++ b/src/video_core/textures/texture.cpp
@@ -51,22 +51,6 @@ constexpr std::array<float, 256> SRGB_CONVERSION_LUT = {
     0.917104f, 0.929242f, 0.941493f, 0.953859f, 0.966338f, 1.000000f, 1.000000f, 1.000000f,
 };
 
-unsigned SettingsMinimumAnisotropy() noexcept {
-    switch (static_cast<Anisotropy>(Settings::values.max_anisotropy.GetValue())) {
-    default:
-    case Anisotropy::Default:
-        return 1U;
-    case Anisotropy::Filter2x:
-        return 2U;
-    case Anisotropy::Filter4x:
-        return 4U;
-    case Anisotropy::Filter8x:
-        return 8U;
-    case Anisotropy::Filter16x:
-        return 16U;
-    }
-}
-
 } // Anonymous namespace
 
 std::array<float, 4> TSCEntry::BorderColor() const noexcept {
@@ -78,7 +62,18 @@ std::array<float, 4> TSCEntry::BorderColor() const noexcept {
 }
 
 float TSCEntry::MaxAnisotropy() const noexcept {
-    return static_cast<float>(std::max(1U << max_anisotropy, SettingsMinimumAnisotropy()));
+    if (max_anisotropy == 0 && mipmap_filter != TextureMipmapFilter::Linear) {
+        return 1.0f;
+    }
+    const auto anisotropic_settings = Settings::values.max_anisotropy.GetValue();
+    u32 added_anisotropic{};
+    if (anisotropic_settings == 0) {
+        added_anisotropic = Settings::values.resolution_info.up_scale >>
+                            Settings::values.resolution_info.down_shift;
+    } else {
+        added_anisotropic = Settings::values.max_anisotropy.GetValue() - 1U;
+    }
+    return static_cast<float>(1U << (max_anisotropy + added_anisotropic));
 }
 
 } // namespace Tegra::Texture
diff --git a/src/video_core/video_core.cpp b/src/video_core/video_core.cpp
index cae543a51..e852c817e 100644
--- a/src/video_core/video_core.cpp
+++ b/src/video_core/video_core.cpp
@@ -37,6 +37,8 @@ std::unique_ptr<VideoCore::RendererBase> CreateRenderer(
 namespace VideoCore {
 
 std::unique_ptr<Tegra::GPU> CreateGPU(Core::Frontend::EmuWindow& emu_window, Core::System& system) {
+    Settings::UpdateRescalingInfo();
+
     const auto nvdec_value = Settings::values.nvdec_emulation.GetValue();
     const bool use_nvdec = nvdec_value != Settings::NvdecEmulation::Off;
     const bool use_async = Settings::values.use_asynchronous_gpu_emulation.GetValue();
@@ -53,11 +55,10 @@ std::unique_ptr<Tegra::GPU> CreateGPU(Core::Frontend::EmuWindow& emu_window, Cor
     }
 }
 
-u16 GetResolutionScaleFactor(const RendererBase& renderer) {
-    return static_cast<u16>(
-        Settings::values.resolution_factor.GetValue() != 0
-            ? Settings::values.resolution_factor.GetValue()
-            : renderer.GetRenderWindow().GetFramebufferLayout().GetScalingRatio());
+float GetResolutionScaleFactor(const RendererBase& renderer) {
+    return Settings::values.resolution_info.active
+               ? Settings::values.resolution_info.up_factor
+               : renderer.GetRenderWindow().GetFramebufferLayout().GetScalingRatio();
 }
 
 } // namespace VideoCore
diff --git a/src/video_core/video_core.h b/src/video_core/video_core.h
index f5c27125d..f86877e86 100644
--- a/src/video_core/video_core.h
+++ b/src/video_core/video_core.h
@@ -25,6 +25,6 @@ class RendererBase;
 /// Creates an emulated GPU instance using the given system context.
 std::unique_ptr<Tegra::GPU> CreateGPU(Core::Frontend::EmuWindow& emu_window, Core::System& system);
 
-u16 GetResolutionScaleFactor(const RendererBase& renderer);
+float GetResolutionScaleFactor(const RendererBase& renderer);
 
 } // namespace VideoCore
diff --git a/src/video_core/vulkan_common/vulkan_device.h b/src/video_core/vulkan_common/vulkan_device.h
index 2d5daf6cd..10653ac6b 100644
--- a/src/video_core/vulkan_common/vulkan_device.h
+++ b/src/video_core/vulkan_common/vulkan_device.h
@@ -40,6 +40,10 @@ public:
     VkFormat GetSupportedFormat(VkFormat wanted_format, VkFormatFeatureFlags wanted_usage,
                                 FormatType format_type) const;
 
+    /// Returns true if a format is supported.
+    bool IsFormatSupported(VkFormat wanted_format, VkFormatFeatureFlags wanted_usage,
+                           FormatType format_type) const;
+
     /// Reports a device loss.
     void ReportLoss() const;
 
@@ -370,10 +374,6 @@ private:
     /// Returns true if the device natively supports blitting depth stencil images.
     bool TestDepthStencilBlits() const;
 
-    /// Returns true if a format is supported.
-    bool IsFormatSupported(VkFormat wanted_format, VkFormatFeatureFlags wanted_usage,
-                           FormatType format_type) const;
-
     VkInstance instance;                                         ///< Vulkan instance.
     vk::DeviceDispatch dld;                                      ///< Device function pointers.
     vk::PhysicalDevice physical;                                 ///< Physical device.