KYTYPS5 / GAME COMPATIBILITY

ASTRO's PLAYROOM

PPSA01325 · 3 platform reports

Check the result for your operating system, then compare your build and game version with the test.

Windows

In game

Tested 6 Oct 2026
Read the Windows report

Linux

In game

Tested 4 Oct 2026
Read the Linux report

macOS

Doesn’t boot

Tested 28 Sept 2026
Read the macOS report

Imported from the project’s published compatibility database on . Report dates, builds and hardware remain those of the original community tests.

LINUX TEST

ASTRO's PLAYROOM on Linux

In game
Tested build
0.3.0
Test date
Tested title ID
PPSA01325
Game version
Not recorded
Test hardware
AMD Ryzen 7 5700X / AMD RX 7800XT / 16GB DDR4 RAM / 16GB VRAM

Community report notes

vkQueueSubmit failed: ErrorDeviceLost (-4), tick=1345396 debug_op=3 debug_submit=37611 args=4,1,0,0,0x0000000500803978 --- Build --- Official build KytyPS5-2026-10-04-719e025 --- Fatal Error --- Not implemented (result != vk::Result::eSuccess) in /home/runner/work/KytyPS5/KytyPS5/src/graphics/host_gpu/renderer/commandScheduler.cpp:388

Steps to reproduce

  1. Open KytyPS5
  2. Boot ASTRO'S PLAYROOM
  3. Finish the controller showcase scene
  4. Get to initial tutorial
  5. Crash!

Expected behavior

Continue normal gameplay

Last working build / first broken build

Remember it being broken but not crashing in an earlier commit, new latest version broke it

Extra notes

Found a fix, go to the save data, and set <IsCpuTowerActivated>false</IsCpuTowerActivated> (this never gets turned on since the game crashes) to true

KytyPS5 issue #1054 · Original community test.

MACOS TEST

ASTRO's PLAYROOM on macOS

Doesn’t boot
Tested build
22ff469
Test date
Tested title ID
PPSA01325
Game version
Not recorded
Test hardware
Apple Sillicon M3 / Apple Sillicon M3 / 8 GB Unified Ram

Community report notes

I launch it; it doesn't boot

Steps to reproduce

  1. Open KytyPS5
  2. Try to open the game
  3. crash

Expected behavior

No known issue because it doesn't even launch

KytyPS5 issue #892 · Original community test.

WINDOWS TEST

ASTRO's PLAYROOM on Windows

In game
Tested build
Official build KytyPS5-2026-10-06-fb7ff7e (ver 0.3.0)
Test date
Tested title ID
PPSA01325
Game version
Not recorded
Test hardware
AMD Ryzen 5 5600 (6C/12T) / AMD Radeon RX 6700 XT, Adrenalin 26.8.1 (driver 32.0.21045.5002), Vulkan / 32 gb ddr4 3600 4 planks

Community report notes

The game reaches controllable gameplay (CPU Plaza and levels) but performance is very low on RDNA2: 5-8 fps in levels, ~25 fps with the camera pointed away from level geometry.

Profile data (Windows GPU engine perf counters + thread sampling, measured while in level):

  • GPU engine utilization: 3D ~64-68% average; NO Compute engine activity; NO Copy engine activity (uploads appear serialized on the 3D queue).
  • Emulator CPU usage: ~1 core total of 12; one dominant thread ~53% of a single core, all other threads <=7%. Serialized prepare/submit loop waiting on the GPU - neither side saturated, the rest is bubbles.
  • Frame math: ~166 ms/frame, ~110 ms of it GPU 3D busy at a forced 1080p internal render target. The console renders this scene at 60 Hz on similar TFLOPs, so the effective recompiler/pipeline overhead is roughly 6-7x.
  • Shader counts keep growing slowly during gameplay (VS 79 / PS 118 / CS 31) - on-the-fly compilation present but not a stutter storm.
  • Title screen hits the --vblank-frequency 30 cap (30 fps) with CPU/GPU mostly idle.

Steps to reproduce

  1. Open KytyPS5 (build KytyPS5-2026-10-06-fb7ff7e).
  2. Boot ASTRO's PLAYROOM (PPSA01325, 01.904.000).
  3. Enter CPU Plaza and any level (e.g. Memory Heaven).
  4. Look at the level geometry with the camera: fps counter in the window title drops to 5-8.
  5. Point the camera at the sky/empty space: fps rises to ~25.

Expected behavior

On a PS5-class GPU (RX 6700 XT is similar in TFLOPs to the console) this scene renders at 60 Hz in 16.6 ms. The emulator currently spends ~110 ms of GPU 3D time per frame at a forced 1080p internal target, so I would expect roughly an order of magnitude less GPU time per frame once known inefficiencies are addressed.

Extra notes

A community memory patch was applied and verified in the guest log ("Successfully applied cheat: Select the common GI-unavailable state" + "Fixed render resolution 1920x1080"), so the internal render target is forced to 1080p and GI is disabled - the measurements above are WITH those active.

Full guest printf log attached (prof_fb7ff7e_20261006.zip, ~28 MB uncompressed). Patched JSON used for the partial cheat application can be shared if useful.

Questions:

  • Is ~110 ms 3D/frame at 1080p on RDNA2 expected at this stage, or are known inefficiencies already identified here?
  • Would a Tracy capture (--profile) or command buffer dumps from this level be useful? Happy to run specific env switches/patches and re-measure.

KytyPS5 issue #1113 · Original community test.

Compare another title.

Search all 333 games, or browse reports for your operating system.

Browse all games