AnyPS5 / Test reports
AnyPS5 compatibility report: games, builds and evidence
Five PS5 game case studies with reviewed Windows and Linux reports, exact builds, patches and original evidence. Reported gameplay does not guarantee full-game completion.
This is a curated source report, not an independent benchmark. These five examples are not a representative sample of the PS5 library. Unknown fields stay unknown, and results apply only to the reported edition, platform, build and patches.
Gran Turismo 7
PPSA01317 · PS5 · PPSA01317
For GT7, identify the mode and exact stop before calling a result playable. A Music Rally report is not evidence for the campaign, online services or a full playthrough. Start with the Windows build and local changes in the selected source, and keep other PPSA editions separate.
Windows
Reviewed Oct 14, 2026
Community report
Gameplay · Locally modified
The local branch loads the first Music Rally song with HUD and Auto Drive and reaches TRY AGAIN at its end, at about 2 FPS. The minimap is corrupted; green/cyan rendering may recur, one of eight loads lost the GPU device, and skipping the intro caused a crash. Offline networking remains enabled. Manual driving, other races and completion are unverified. The follow-up does not restate its content version or hardware; this is not stock main.
- AnyPS5 build
- Same local integration as #4697; race-test revision and full patch set not pinned
- Game version
- Not specified
- CPU
- Not specified
- GPU
- Not specified
- Driver
- Not specified
- Test date
- Not specified
Community report
Gameplay · Locally modified
With profiling enabled, a local write-watch comparison on the same Music Rally song changes about 0.6 FPS to 0.6–1.2. Memory-table rebuilds still cause swings. This is not stock main, and the head, hardware and content version are not restated. Do not replace the earlier roughly 2 FPS observation with this different measured run.
- AnyPS5 build
- local integration · write-watch changes · exact head not stated
- Game version
- Not specified
- CPU
- Not specified
- GPU
- Not specified
- Driver
- Not specified
- Test date
- Not specified
Linux
No reviewed report
Hollow Knight: Silksong
PPSA12544 · Not specified
Compare Silksong’s Windows and Linux evidence separately. Copy the source commit and patch state, not only its reported FPS. A later success does not explain every earlier failure, and an FPS observation without matched hardware is not a PC requirement or benchmark.
Windows
Reviewed Oct 14, 2026
Community report
Gameplay · No local patches reported
RAM: 32 GB. Toolchain: WinLibs GCC 15.2.0 r7 · Release. Plain main boots into a new game with DualSense, correct text, rendering and audio, without runtime overrides. The window counter stays above 50 and mostly 60 FPS in focused gameplay. Background FIFO measurements fell to 3.9; mailbox held 59.95 only in a separate menu test. Completion is unverified.
- AnyPS5 build
- main 18c438b10
- Game version
- 01.000.010
- CPU
- Intel Core i9-10850K
- GPU
- NVIDIA RTX 3080 · 10 GB
- Driver
- 610.74
- Test date
- Not specified
Linux
Reviewed Oct 14, 2026
Community report
Gameplay · No local patches reported
A later unpatched main run reaches gameplay at about 16–24 FPS. Entering Moss Grotto drops to about 1.5 FPS at an unsupported depth/stencil access. One earlier launch stalled during loading; this is limited gameplay evidence, not a stable or full-game result.
- AnyPS5 build
- main a11c8f7f
- Game version
- Not specified
- CPU
- Intel i7-8750H
- GPU
- NVIDIA GTX 1050 Ti Max-Q 4 GB
- Driver
- 580.178.04
- Test date
- Not specified
Community report
Menu · Locally modified
A separate report measures 59.93–60 FPS in the interactive main menu. Although the engine is stock main, the game’s own assets were modified for entitlement and save-data initialization. This is a patched menu result, not unmodified gameplay.
- AnyPS5 build
- main a11c8f7f7 + guest-asset modifications
- Game version
- 1.0.30021
- CPU
- Intel i5-13420H
- GPU
- NVIDIA RTX 4050 Laptop
- Driver
- 610.57.04
- Test date
- Not specified
Dreaming Sarah
PPSA02929 · Not specified
Read the historical Windows release result alongside the newer Linux reports. They describe different builds and environments. An issue being closed or a fix being merged is not a post-fix game retest; use the linked source to establish what was actually observed.
Windows
Reviewed Oct 14, 2026
Community report
Gameplay · No local patches reported
RAM: 128 GB. Toolchain: MinGW-w64 GCC 15.2.0 · Release. The author reports steady 60 FPS in tested gameplay on clean main with matching libraries. An earlier build on this PC gave about 15 FPS; no responsible change was isolated. Driver, content version, test duration, audio and completion are not stated.
- AnyPS5 build
- main 18c438b10
- Game version
- Not specified
- CPU
- AMD Ryzen 9 7950X
- GPU
- NVIDIA RTX 4070 Ti SUPER
- Driver
- Not specified
- Test date
- Not specified
Linux
Reviewed Oct 14, 2026
Community report
Gameplay · Patch status unknown
The reporter now confirms Linux success and supplies a forest scene screenshot. The visible 59.93 counter is a single screenshot, not a timed benchmark. The follow-up does not pin its new commit, patch set, game version or hardware; audio, controls and completion remain unverified. Keep the older release and c60c123f failures separate.
- AnyPS5 build
- Not specified
- Game version
- Not specified
- CPU
- Not specified
- GPU
- Not specified
- Driver
- Not specified
- Test date
- Not specified
ASTRO BOT
PPSA21564 · Not specified
Match Astro Bot’s PPSA edition before comparing reports. Distinguish a scene that advances internally from one that renders and responds to input. Software Vulkan, local integration changes and hardware GPU runs describe different environments and cannot share a single performance verdict.
Windows
Reviewed Oct 14, 2026
Community report
Gameplay · Locally modified
A later Windows report in the discussion describes gameplay at roughly 17–19 FPS with this fork and fix. The image freezes for long periods while audio continues. The earlier blocked baseline in the original discussion comes from a separate run.
- AnyPS5 build
- astrobot b58df66c + #637
- Game version
- 01.007.000
- CPU
- AMD Ryzen 7 5700X
- GPU
- AMD Radeon RX 9060 XT
- Driver
- AMD 26.9.2
- Test date
- Not specified
Linux
Reviewed Oct 14, 2026
Community report
Gameplay · Locally modified
Three quiet Linux runs per configuration observe title, menu, first-level space and Space Station scenes on integration bdb0ba35. The stack reports 26.4 FPS at title and 24.8 in menus; level frame times are 33.1 ms and 317 ms, not whole-game FPS. Clouds and sky remain missing, and not all astrobot optimizations are ported. Plain main still needs depth/stencil fixes before title startup. Controls, audio, saves and completion are not retested.
- AnyPS5 build
- astro-main/integration-1013a bdb0ba35 · main 682ef0c1 + PR stack
- Game version
- Not specified
- CPU
- Not specified
- GPU
- NVIDIA RTX 4090
- Driver
- Not specified
- Test date
- Not specified
Community report
Gameplay · Locally modified
Toolchain: GCC 16.2.1 · Release. Combined integration 54cb611d61c renders Crash Site arrival and verifies a scripted jump and landing in a planned 600-second test. It uses direct level selection, a 16 GiB import budget and temporary 8 GiB udmabuf limit. Audio, ordinary New Game progression and matched FPS gain are unverified. Newer c058cf416b3 profiling has no published result; do not attach this run to it.
- AnyPS5 build
- JJenkx combined integration 54cb611d61c · pipeline ON
- Game version
- 01.018.000
- CPU
- AMD Ryzen 9 7950X
- GPU
- AMD Radeon RX 6650 XT · 8 GiB
- Driver
- Mesa 26.2.4 RADV
- Test date
- Oct 13, 2026
DOOM · PS5 homebrew port
PPSA99666 · Not specified
This entry is a PS5 homebrew port of DOOM. Keep the software-rendered and hardware-GPU reports separate, including their command-line options and observation periods. A short homebrew run does not establish compatibility or performance for commercial PS5 games.
Windows
Reviewed Oct 13, 2026
Community report
Gameplay · No local patches reported
Unmodified main b42ef24e0 reaches E1M1 with menus, movement, firing, pickups, Xbox controller input and audible audio on this AMD device. The homebrew SELF/fSELF files were unwrapped and guest libc converted. First runs stuttered; FPS, later maps and save/load were not measured. This is DOOM homebrew v1.1.1, not a commercial PS5 DOOM edition.
- AnyPS5 build
- main b42ef24e0
- Game version
- 01.001.001 · homebrew v1.1.1
- CPU
- Not specified
- GPU
- AMD Radeon RX 9060 XT
- Driver
- 32.0.31041.1004
- Test date
- Not specified
Community report
Gameplay · No local patches reported
Unmodified main b42ef24e0 reaches E1M1 with menus, movement, firing, pickups, Xbox controller input and audible audio on this AMD device. The homebrew SELF/fSELF files were unwrapped and guest libc converted. First runs stuttered; FPS, later maps and save/load were not measured. This is DOOM homebrew v1.1.1, not a commercial PS5 DOOM edition.
- AnyPS5 build
- main b42ef24e0
- Game version
- 01.001.001 · homebrew v1.1.1
- CPU
- AMD Ryzen 9 7950X
- GPU
- AMD Ryzen 9 7950X integrated graphics
- Driver
- 32.0.21045.5002
- Test date
- Not specified
Community report
Startup · No local patches reported
Unmodified main 59a2167a9 keeps AGC compute presentation in four Windows/NVIDIA startup runs, each with an empty AnyPS5 shader cache. The NVIDIA driver cache was not cleared, so only the first run was driver-cold. Temporary-delay experiments are separate. Gameplay, audio and FPS were not retested, and Linux Intel CPU-scaler fallback remains unexplained. This is DOOM Shareware homebrew PPSA99666, not a retail DOOM edition.
- AnyPS5 build
- main 59a2167a9
- Game version
- 01.001.000 · DOOM Shareware
- CPU
- Not specified
- GPU
- NVIDIA RTX 4070 Laptop GPU
- Driver
- 32.0.16.1692
- Test date
- Not specified
Linux
Reviewed Oct 13, 2026
Community report
Gameplay · No local patches reported
Intel HD 520 main testing reaches E1M1 movement, firing and automap. Frame-dump timestamps average about 29 FPS in game and 28 in the launcher. A cold first frame takes about five minutes; warm launch takes 2.1 seconds, but both still use the CPU scaler. Audio was not listened to; later maps, saves and physical controllers remain unchecked. This does not contradict a different NVIDIA cold-start test.
- AnyPS5 build
- main 9aa5fd290
- Game version
- 01.001.001 · homebrew v1.1.1
- CPU
- Intel Core i5-6300U
- GPU
- Intel HD Graphics 520
- Driver
- Mesa 26.2.4 ANV
- Test date
- Not specified
Community report
Gameplay · No local patches reported
The archived Linux NVIDIA main report reaches E1M1 at 35 FPS after a 60 FPS launcher. Movement, firing, item pickup and automap were checked with file-driven controller input. Audio opened at 48000 Hz but was not listened to; later maps and save/load were not tested. Guest libc was converted; no extra PR or deprecated module exclusion was used.
- AnyPS5 build
- main a11c8f7f7
- Game version
- 01.001.001 · homebrew v1.1.1
- CPU
- Not specified
- GPU
- NVIDIA RTX 3080 Ti
- Driver
- 610.57.04
- Test date
- Not specified
Community report
Gameplay · No local patches reported
The archived Linux AMD integrated-GPU report completes E1M1 and E1M2 at 35 FPS with DualShock 4 input, heard audio and a save/load round trip. Warm-cache AGC presentation works; the first cache-empty run stays on compute presentation. Accessible udmabuf was part of this host setup. Keep these conditions with main a11c8f7f7; this is the homebrew port, not a retail DOOM game.
- AnyPS5 build
- main a11c8f7f7
- Game version
- 01.001.001 · homebrew v1.1.1
- CPU
- AMD Ryzen 5 PRO 5650U
- GPU
- AMD RENOIR integrated graphics
- Driver
- Mesa 26.2.4 RADV
- Test date
- Not specified
Original sources
When citing a result, link to its original report and retain the PPSA ID, operating system, build, patch status and test date. The review date records our source review, not a new game test.
The stage shown is the furthest point reported, not a guarantee of stable play. Locally modified results may rely on forks, patches or changes to game files.
How we review reports