This is an old release (2026-10-04), superseded since. The current DLLs are on the latest release page, which always points to the newest one. The text below describes this old release only.
Superseded by dll-2026-10-04.2, which also changes how the postpass writes its images. This release stays available.
A ready-made copy of AMD's FSR 4.1.1 upscaler DLL (amd_fidelityfx_upscaler_dx12.dll, version 4.1.1.2740) with the faster shaders from this repository built in: the phased postpass and the model pass 11 rewrite, for every output size and preset. Replace the DLL and FSR 4 uses them, with no launch option, add-on or override folder.
New in this release (2026-10-04)
Model pass 11 now stores each row of its output together instead of one word at a time. The pass reads 234 MB from memory per frame at 4K instead of 1,018 MB with AMD's shader, and FSR 4 as a whole reads about 30% less than with AMD's shaders. On a Radeon RX 7800 XT this does not change the frame rate (108 FPS and 3.12 ms of upscaler time in Shadow of the Tomb Raider's benchmark, with or without it); it may help GPUs with less memory bandwidth, where it has not been measured. The postpass is unchanged from the previous release (dll-2026-10-03).
The output is byte-for-byte identical to AMD's DLL.
Install
- In the game folder, find
amd_fidelityfx_upscaler_dx12.dll(often next to OptiScaler) and check that it is version 4.1.1.2740. Its SHA-256 isd0dcccc74a43c44ba435b7a369b456e0970d8a4464e4bd683119b374f2c9fb46. This DLL replaces that version only. - Rename the original to
amd_fidelityfx_upscaler_dx12.dll.origand keep it. - Put the DLL from this release in its place.
- Compare OptiScaler's upscaler time with the original and the patched DLL. In some games that number is only reliable with the frame rate uncapped.
To undo it, put the original back.
Experimental build for integrated GPUs (added 2026-10-04)
On RDNA3 integrated GPUs the normal DLL is reported slower than AMD's, and a tester traced that to the pass 11 rewrite. These two builds carry a different pass 11, made for integrated GPUs: it keeps AMD's loops, so the pass stays small (6.5 KB of code as compiled for a Radeon 780M, against 6.1 KB for AMD's and 31 KB for the normal rewrite), and only stores each row of its output together, which cuts the memory the pass reads to about a quarter. The postpass rewrite is included as well (reported to make no difference on an integrated GPU).
test-igpu.zip: the integrated-GPU buildtest-igpu-any-gpu.zip: the same with AMD's GPU check lifted. AMD's DLL does not offer this version of FSR 4 on integrated GPUs, so on Windows this is probably the one you need.
Both give output byte-for-byte identical to AMD's DLL in the seven test configurations under Proton. Neither has been run on an integrated GPU: whether they are faster than AMD's shaders there is exactly what needs testing. Please compare OptiScaler's upscaler time (frame rate uncapped, same spot) with AMD's original DLL and with this build, and report both numbers with your GPU, Windows or Linux and driver version, game, output resolution and FSR mode. They were built from what is now commit 18e3ddd on the main branch, with PASS11=rows.
Test builds for integrated GPUs
On RDNA3 integrated GPUs (Radeon 780M and similar), testers report that FSR 4 gets slower with these rewrites. A tester who ran them separately (with the 2026-10-03 release) found that the pass 11 rewrite makes the upscaler much slower there and that the postpass rewrite behaves like AMD's shader. So do not use this DLL on an integrated GPU; the two test builds each contain only one of the rewrites, for anyone who wants to check their own GPU:
test-pass11-only.zip: only the model pass 11 rewritetest-postpass-only.zip: only the postpass rewrite
Each zip holds amd_fidelityfx_upscaler_dx12.dll (install it like the full DLL) and AMD's notice. Both give output byte-for-byte identical to AMD's DLL under Proton.
If you have an integrated GPU, please compare OptiScaler's upscaler time (frame rate uncapped, same spot) with AMD's original DLL, the pass 11-only build and the postpass-only build, and report the three numbers with your GPU, Windows or Linux and driver version, game, output resolution and FSR mode. See Integrated GPUs.
Experimental: RDNA2 and other GPUs FSR 4 refuses
AMD's DLL only offers the INT8 version of FSR 4 on desktop RDNA3 (chip family 0x91); it refuses RDNA2 (RX 6000), RDNA3 integrated GPUs (Radeon 780M, 890M) and other vendors. These builds make that check always answer yes:
test-any-gpu-amd-shaders.zip: check lifted, AMD's shaders unchanged (does FSR 4 run at all?)test-any-gpu.zip: check lifted, both rewritestest-any-gpu-pass11-only.zip: check lifted, pass 11 rewrite only
The INT8 shaders are plain Direct3D 12 (Shader Model 6.6) and need no RDNA3-only feature, and under Proton on an RX 7800 XT all three give output byte-identical to AMD's DLL. But they have not been tried on any GPU the check refuses. Whether FSR 4 then runs correctly, and how fast, depends on that GPU and its driver. Do not use them on RDNA4, where both versions of FSR 4 would then report support. Please report your GPU, OS and driver, game, whether FSR 4 starts and looks right, and the upscaler time. See RDNA2 and other GPUs.
What was measured
- Correctness: AMD's DLL and this one, run under Proton in a small FSR 4 test program at seven output sizes and ratios (4K, 1440p and 1080p output), give byte-identical images; this was repeated for this release. See
dll/README.md. - Speed: on a Radeon RX 7800 XT under Proton, the postpass takes 0.67 ms at 4K with these shaders, against 2.1 ms for AMD's. In Shadow of the Tomb Raider's benchmark at 4K, the same shaders (as the Linux launch option) brought the upscaler time from 4.16 ms to 3.12 ms and the frame rate from 97 to 108 FPS. See
results/phased-postpass.
Things to know
- Windows: testers with Radeon RX 7000 graphics cards report large improvements with the first version of these rewrites, nearly as large as on Linux, and the phased version (what this DLL contains) has been run on Windows with an RX 7800 XT with about the same result. Measure before and after on your setup.
- Integrated GPUs (Radeon 780M and similar): a tester found the pass 11 rewrite makes FSR 4 much slower there and the postpass rewrite no faster, so don't use this DLL there. See Integrated GPUs.
- On Linux, the launch option is slightly faster (postpass 0.60 ms against 0.67 ms at 4K), because vkd3d-proton translates the DLL's DXIL into slightly slower code. See the quick start.
- AMD's digital signature has been removed, because it cannot survive the change. Neither the DLL nor AMD's FidelityFX loader checks it.
- Game and OptiScaler updates may put the original DLL back.
- Online games: this is modified code; use your own judgement in games with anti-cheat.
How it was made
python3 dll/patch_upscaler_dll.py amd_fidelityfx_upscaler_dx12.dll out.dll at commit ef57cf2. Running it on AMD's original produces exactly this file. The test builds are made the same way with --only pass11, --only postpass, --any-gpu, --any-gpu --only none and --any-gpu --only pass11. Checksums of all files, including the DLLs inside the zips, are in SHA256SUMS.
Licence
This is AMD's DLL with modified shaders. AMD's FidelityFX SDK licence lists the DLL under MIT terms; the shader modifications come from scripts adapted from bbport and are distributed under the GNU GPL v2 or later. NOTICE.md (attached) carries AMD's notice and the details. Not affiliated with or endorsed by AMD.