Also got some weird issues when enabling FSR3, not sure if something related to Proton or the game itself. Another thing happening to me is that the game is supposed to have DualSense native support but at least on my case, game wont detect gamepad if I dont enable Steam Input (And then loosing DualSense features)

Also got some weird issues when enabling FSR3, not sure if something related to Proton or the game itself. Another thing happening to me is that the game is supposed to have DualSense native support but at least on my case, game wont detect gamepad if I dont enable Steam Input (And then loosing DualSense features)

you can try using the proton build made by the creator of this issue: https://github.com/ValveSoftware/Proton/issues/5900

at least for me disabling steam input did get me playstation prompts so game recognizes dualsense, at least adaptive triggers should work, rumble is working.

Also got some weird issues when enabling FSR3, not sure if something related to Proton or the game itself. Another thing happening to me is that the game is supposed to have DualSense native support but at least on my case, game wont detect gamepad if I dont enable Steam Input (And then loosing DualSense features)

you can try using the proton build made by the creator of this issue: #5900

at least for me disabling steam input did get me playstation prompts so game recognizes dualsense, at least adaptive triggers should work, rumble is working.

May give it a try, but nowadays DualSense seems to be working even with Bluetooth (That proton says it needs USB connected), working nicely with all features with Bluetooth on games like Ghost of Tsushima or Ratchet & Clank

EDIT: Looks like using USB connected DualSense works perfectly with regular Proton Experimental, is a game specific thing the capability to work also wirelessly?

Also Im getting this message but pressing OK will launch the game correctly

I'm seeing graphical glitches in the game, sometimes they're pretty intense and sometimes less so. I tried out all the different graphical presets and they don't seem to have an effect on this issue. This video demonstrates the glitches:

https://github.com/user-attachments/assets/7b8a2b00-dac0-4cb1-a16e-7de79924771e

Proton log

Steam runtime diagnostics

System:
  Kernel: 6.10.5-200.fc40.x86_64 arch: x86_64 bits: 64 compiler: gcc
    v: 2.41-37.fc40
  Desktop: GNOME v: 46.4 Distro: Fedora Linux 40 (Workstation Edition)
Machine:
  Type: Desktop Mobo: ASRock model: X570 Taichi serial: <superuser required>
    UEFI: American Megatrends v: P5.60 date: 01/18/2024
Battery:
  Device-1: hidpp_battery_0 model: Logitech MX Keys Wireless Keyboard
    charge: 100% (should be ignored) status: discharging
CPU:
  Info: 8-core model: AMD Ryzen 7 5800X3D bits: 64 type: MT MCP arch: Zen 3+
    rev: 2 cache: L1: 512 KiB L2: 4 MiB L3: 96 MiB
  Speed (MHz): avg: 2199 high: 2200 min/max: 2200/4549 boost: enabled cores:
    1: 2200 2: 2200 3: 2200 4: 2200 5: 2200 6: 2200 7: 2200 8: 2200 9: 2200
    10: 2200 11: 2200 12: 2193 13: 2200 14: 2198 15: 2200 16: 2200
    bogomips: 108793
  Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3 svm
Graphics:
  Device-1: AMD Navi 31 [Radeon RX 7900 XT/7900 XTX/7900M]
    vendor: Sapphire PULSE driver: amdgpu v: kernel arch: RDNA-3 bus-ID: 0f:00.0
  Device-2: Realtek Streamplify CAM driver: snd-usb-audio,uvcvideo type: USB
    bus-ID: 5-3:4
  Display: wayland server: X.Org v: 24.1.2 with: Xwayland v: 24.1.2
    compositor: gnome-shell driver: dri: radeonsi gpu: amdgpu resolution:
    1: 2560x1440~240Hz 2: 2560x1440~75Hz
  API: OpenGL v: 4.6 vendor: amd mesa v: 24.1.5 glx-v: 1.4
    direct-render: yes renderer: AMD Radeon RX 7900 XT (radeonsi navi31 LLVM
    18.1.6 DRM 3.57 6.10.5-200.fc40.x86_64)
  API: EGL Message: EGL data requires eglinfo. Check --recommends.
Audio:
  Device-1: AMD Navi 31 HDMI/DP Audio driver: snd_hda_intel v: kernel
    bus-ID: 0f:00.1
  Device-2: AMD Starship/Matisse HD Audio vendor: ASRock
    driver: snd_hda_intel v: kernel bus-ID: 11:00.4
  Device-3: HP HyperX Cloud Alpha Wireless
    driver: cdc_acm,hid-generic,snd-usb-audio,usbhid type: USB bus-ID: 1-2.2:5
  Device-4: Focusrite-Novation Scarlett Solo 4th Gen
    driver: snd-usb-audio,usb-storage type: USB bus-ID: 3-5:4
  Device-5: Realtek Streamplify CAM driver: snd-usb-audio,uvcvideo type: USB
    bus-ID: 5-3:4
  API: ALSA v: k6.10.5-200.fc40.x86_64 status: kernel-api
  Server-1: JACK v: 1.9.22 status: off
  Server-2: PipeWire v: 1.0.7 status: active
Network:
  Device-1: Intel Wi-Fi 6 AX200 driver: iwlwifi v: kernel bus-ID: 05:00.0
  IF: wlp5s0 state: down mac: <filter>
  Device-2: Intel I211 Gigabit Network vendor: ASRock driver: igb v: kernel
    port: f000 bus-ID: 07:00.0
  IF: enp7s0 state: up speed: 1000 Mbps duplex: full mac: <filter>
Bluetooth:
  Device-1: Intel AX200 Bluetooth driver: btusb v: 0.8 type: USB bus-ID: 3-2:2
  Report: btmgmt ID: hci0 rfk-id: 0 state: up address: <filter> bt-v: 5.2
    lmp-v: 11
Drives:
  Local Storage: total: 3.64 TiB used: 1.98 TiB (54.3%)
  ID-1: /dev/nvme0n1 vendor: Samsung model: SSD 990 PRO 4TB size: 3.64 TiB
    temp: 49.9 C
  ID-2: /dev/sda model: Scarlett Welcome Disk size: 192 KiB type: USB
Partition:
  ID-1: / size: 3.64 TiB used: 1.98 TiB (54.4%) fs: btrfs dev: /dev/dm-0
    mapped: luks-765cb1ef-1195-42ea-a526-06f1e57c6cbc
  ID-2: /boot size: 973.4 MiB used: 409.8 MiB (42.1%) fs: ext4
    dev: /dev/nvme0n1p2
  ID-3: /boot/efi size: 598.8 MiB used: 19 MiB (3.2%) fs: vfat
    dev: /dev/nvme0n1p1
  ID-4: /home size: 3.64 TiB used: 1.98 TiB (54.4%) fs: btrfs dev: /dev/dm-0
    mapped: luks-765cb1ef-1195-42ea-a526-06f1e57c6cbc
Swap:
  ID-1: swap-1 type: zram size: 8 GiB used: 459.8 MiB (5.6%) dev: /dev/zram0
Sensors:
  System Temperatures: cpu: 40.5 C mobo: 43.0 C gpu: amdgpu temp: 61.0 C
  Fan Speeds (rpm): fan-1: 603 fan-2: 0 fan-3: 829 fan-4: 553 fan-5: 558
    fan-6: 2601 fan-7: 549 gpu: amdgpu fan: 639
Info:
  Memory: total: 32 GiB available: 31.26 GiB used: 6.9 GiB (22.1%)
  Processes: 483 Uptime: 17h 31m Init: systemd target: graphical (5)
  Packages: 30 Compilers: gcc: 14.2.1 Shell: Zsh v: 5.9 inxi: 3.3.34

I'm seeing graphical glitches in the game, sometimes they're pretty intense and sometimes less so. I tried out all the different graphical presets and they don't seem to have an effect on this issue. This video demonstrates the glitches:
2024-08-19_23-32-58.mp4

Proton log

Steam runtime diagnostics
System details

Me too, reported those on the mesa Gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/issues/11738

The graphical flickering is fixed for me by using VKD3D_CONFIG=no_invariant_position. I found that RADV_DEBUG=invariantgeom also worked but significantly impacted performance.

Important info

I have a Radeon 6800 XT and multiple drivers installed (RADV 24.0.9 + AMDVLK + lavapipe 12 + lavapipe 15)

I was getting a:
"Graphics card driver is incompatible with Final fantasy XVI"

Then shutdown. After a quick debug I noticed the game was picking the wrong Vulkan device (I think it was picking AMDVLK):

6444.707:0130:0134:err:vkd3d-proton:vkd3d_init_device_caps: Push descriptors are not supported by this implementation. This is required for correct operation.

That was suspicious because RADV definitely supports VK_KHR_push_descriptor but AMDVLK doesn't (at least the old version I left installed).

I was able to force to pick RADV (the right device) by forcing it with environment variables:

VK_ICD_FILENAMES=/usr/share/vulkan/icd.d/radeon_icd.x86_64.json steam steam://rungameid/2738000

Update: This also works (by going in Steam to Game -> Options -> General -> Launch Options):

VK_ICD_FILENAMES=/usr/share/vulkan/icd.d/radeon_icd.x86_64.json %command%

The graphical flickering is fixed for me by using VKD3D_CONFIG=no_invariant_position. I found that RADV_DEBUG=invariantgeom also worked but significantly impacted performance.

Hmmm, tried both and still getting some graphical glitches on a 7800XT :S

VKD3D_CONFIG=no_invariant_position works for me

i'm using mesa-git and i don't have those graphical glitches on my end, running on rdna2, it may be a rdna3 issue.

I tend to use mesa frome my distro so currently 24.1.6, and was getting the graphical glitching. Switching to a 3rd party repo to test mesa git (24.3) and the graphical glitches did go away. So it definitely seems like maybe a mesa issue that is fixed on newer version. However that's just two points of data an a 7900XTX

Replying to https://github.com/ValveSoftware/Proton/issues/8001#issuecomment-2297201270

game-specific, in this case. (same behavior on Windows OS) It only works on either DualShock 4, DualSense and DualSense Edge on USB only. SteamInput need to be disabled for this one. But you can try out the "DualSense WIP" branch of Proton Experimental if it still doesn't work.

sidenote: FF16 does not do Auto-Glyphs, instead; it'll be entirely based on your first-time bootup sequence. So; there's a chance that switching to SteamInput will still show PlayStation prompts instead of Xbox. You can either press "Restore Default" or change Button Prompt Type manually.

VKD3D_CONFIG=no_invariant_position works for me

Only in the city, but not after :

For me the demo crashes my PC 100% of the time right after the intro where the guy is sitting near a campfire. This is what my displays look like afterwards

GPU: RX 5700
Kernel: 6.10.6-arch1-1
Mesa: 24.1.6-1

I made a relevant Mesa bug report
https://gitlab.freedesktop.org/mesa/mesa/-/issues/11760

I recorded some logs
dmesg reports this at the time of crash

[  595.179529] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=252009, emitted seq=252011
[  595.180078] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process ffxvi_demo.exe pid 5107 thread vkd3d_queue pid 5194
[  595.180627] amdgpu 0000:0c:00.0: amdgpu: GPU reset begin!
[  595.542574] amdgpu 0000:0c:00.0: amdgpu: BACO reset
[  597.709585] amdgpu 0000:0c:00.0: amdgpu: GPU reset succeeded, trying to resume
[  597.710304] [drm] PCIE GART of 512M enabled (table at 0x0000008000F00000).
[  597.710378] [drm] VRAM is lost due to GPU reset!
[  597.710380] amdgpu 0000:0c:00.0: amdgpu: PSP is resuming...
[  597.756981] amdgpu 0000:0c:00.0: amdgpu: reserve 0x900000 from 0x81fd000000 for PSP TMR
[  597.799965] amdgpu 0000:0c:00.0: amdgpu: RAS: optional ras ta ucode is not available
[  597.805894] amdgpu 0000:0c:00.0: amdgpu: RAP: optional rap ta ucode is not available
[  597.805898] amdgpu 0000:0c:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[  597.805902] amdgpu 0000:0c:00.0: amdgpu: SMU is resuming...
[  597.805941] amdgpu 0000:0c:00.0: amdgpu: use vbios provided pptable
[  597.805944] amdgpu 0000:0c:00.0: amdgpu: smc_dpm_info table revision(format.content): 4.5
[  597.808914] amdgpu 0000:0c:00.0: amdgpu: SMU is resumed successfully!
[  598.160452] [drm] kiq ring mec 2 pipe 1 q 0
[  598.161734] [drm] VCN decode and encode initialized successfully(under DPG Mode).
[  598.162064] [drm] JPEG decode initialized successfully.
[  598.162078] amdgpu 0000:0c:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[  598.162081] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[  598.162083] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[  598.162085] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 5 on hub 0
[  598.162087] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 6 on hub 0
[  598.162089] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 7 on hub 0
[  598.162091] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 8 on hub 0
[  598.162093] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 9 on hub 0
[  598.162095] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 10 on hub 0
[  598.162097] amdgpu 0000:0c:00.0: amdgpu: ring kiq_0.2.1.0 uses VM inv eng 11 on hub 0
[  598.162099] amdgpu 0000:0c:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[  598.162101] amdgpu 0000:0c:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0
[  598.162103] amdgpu 0000:0c:00.0: amdgpu: ring vcn_dec uses VM inv eng 0 on hub 8
[  598.162105] amdgpu 0000:0c:00.0: amdgpu: ring vcn_enc0 uses VM inv eng 1 on hub 8
[  598.162107] amdgpu 0000:0c:00.0: amdgpu: ring vcn_enc1 uses VM inv eng 4 on hub 8
[  598.162109] amdgpu 0000:0c:00.0: amdgpu: ring jpeg_dec uses VM inv eng 5 on hub 8
[  598.164214] amdgpu 0000:0c:00.0: amdgpu: recover vram bo from shadow start
[  598.187980] amdgpu 0000:0c:00.0: amdgpu: recover vram bo from shadow done
[  598.188039] amdgpu 0000:0c:00.0: amdgpu: GPU reset(2) succeeded!
[  598.189074] [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!
[  603.666281] sched: RT throttling activated

And then it just spams

[  703.402431] get_swap_device: Bad swap offset entry 3ffffffffffff
[  703.402442] get_swap_device: Bad swap offset entry 3ffffffffffff
[  703.402456] get_swap_device: Bad swap offset entry 3ffffffffffff
[  703.402468] get_swap_device: Bad swap offset entry 3ffffffffffff
[  703.402479] get_swap_device: Bad swap offset entry 3ffffffffffff
[  703.402492] get_swap_device: Bad swap offset entry 3ffffffffffff
[  703.402505] get_swap_device: Bad swap offset entry 3ffffffffffff
[  703.402518] get_swap_device: Bad swap offset entry 3ffffffffffff

Sure fire way to do away with the artifacts for now in my testing:

Hardware Setup

• 7900XTX
• Ryzen 7 5800x
• 64 GB RAM

Software

• Mesa Git >= 24.3.0_devel.193803.c5156257d93-1
• Linux 6.10.3
• Proton Experimental

Noticing a strange pattern effect on Joshua's clothes, and FSR 3 frame generation seems completely busted on Proton.

This may help apparently:

https://www.reddit.com/r/FFXVI/comments/1f0vxpa/pc_demo_improving_fps/

The Trick:
The game is constantly trying to open 5 .ddl files that don't exist. From what I've researched, they're only used to debug the game during development. So I put 5 empty files in the game folder using the names of these dlls (I created a text file and renamed it to the name of the dlls). Since this version doesn't seem to access the contents of the dlls, it shouldn't crash.

The files are
capture-x64.dll
d3d12-state-tracker-x64.dll
Nvda.Graphics.Interception.dll
renderdoc.dll
winPixGpuCapturer.dll

Further investigation may be needed by Wine devs.

Seconding the cutscene crash issue on my 5700 XT. At the time of writing, the latest vkd3d-proton artifacts prevent the GPU reset crash, but the game still crashes at the same spot, and sends me back to the login screen instead. RADV_DEBUG=nohiz seems to fix the crash and let me progress into the rest of the demo.

Got to first fight in Stillwind.

• Launches in Window mode every time, which is cropped and blurry, and you need to switch to Fullscreen then Borderless or Borderless will be super zoomed in.
• FSR Frame Generation does not seem to be working correctly. Does not increase FPS all that much and is stuttery.
• Seems to be very poorly optimized. Not much difference between Low and Ultra. I am GPU limited.
• No graphical issues that I have noticed or crashes so far. Previous versions of Proton Experimental from a week or so ago did have lots of issues.
• Going from 70+ FPS to a 30 FPS cut-scene is very jarring but that is a game design issue (Consolitis).

OS: Bazzite HTPC Edition
CPU: 5800X
GPU: 6900XT
Resolution: 4K@120hz HDR
Proton: Experimental Bleeding-Edge (version as of the date of this post)
Launch Options: None

Transfer demo files

• Run the game for the first time -> save for the first time
• Locate the Demo Proton prefix and the following dirtree:
  • <STEAM_LIBRARY_DIR>/compatdata/2738000/pfx/drive_c/users/steamuser/My Documents/My Games/FINAL FANTASY XVI DEMO/Steam/<RANDOM_ID>/
  • Copy the numbered "<000n>.png" AND the "auto.png" files from the demo in the previous directory. (FROM: <STEAM_LIBRARY_DIR>/compatdata/2738000/pfx/drive_c/users/steamuser/My Documents/My Games/FINAL FANTASY XVI DEMO/Steam/<RANDOM_ID>/<000n>.png+auto.png)
  • To the final game release equivalent:
    *( TO:) <STEAM_LIBRARY_DIR>/compatdata/2515020/pfx/drive_c/users/steamuser/My Documents/My Games/FINAL FANTASY XVI/Steam/<SAME_RANDOM_ID>/

Launch the final game release: You should now be able to continue from your demo saves.

I've been playing the full version since yesterday and it runs much better than the demo - also no visual artefacts here (Nvidia, driver version 550.107.02).

However, in the "Hideout" a few hours into the game, there is constant stuttering. Looking at my system log, I can see there are some issues with fossilize, when only playing for a minute or so, there are tens of thousands of lines like this in the log, one every couple of milliseconds with different handle ids:

024-09-18T19:25:03.106956+02:00 <...> steam[5768]: It has either not been recorded, or it failed to be recorded earlier (which is expected if application uses an extension that is not recognized by Fossilize). 2024-09-18T19:25:03.106977+02:00 <...> steam[5768]: Fossilize WARN: Pipeline library handle 0x00007a1892634260 is not registered.

Seems there is some issue with the shaders in the "Hideout" level, if I read this correctly.

It's sadly not only the hideout, the somewhat larger area after the one after the Hideout suffers from the same issues. Funnily, if you don't move the camera or your character, you can see that the framerate is mostly fine. As soon as you move anything, it takes a nose dive.

I'm not so sure you're right about fossilize, I'm experiencing the same thing and I don't have any of those logs. I am however getting a bunch of these when it happens:

Sep 19 02:12:13 daniel-manjaro kernel: umip: ffxvi.exe[14459] ip:14a0881a2 sp:1169f0: For now, expensive software emulation returns the result.
Sep 19 02:12:13 daniel-manjaro kernel: umip: ffxvi.exe[14459] ip:1505d6274 sp:1169f0: SGDT instruction cannot be used by applications.
Sep 19 02:12:13 daniel-manjaro kernel: umip: ffxvi.exe[14459] ip:1505d6274 sp:1169f0: For now, expensive software emulation returns the result.
Sep 19 02:12:13 daniel-manjaro kernel: umip: ffxvi.exe[14459] ip:14dcd9338 sp:11efe8: SGDT instruction cannot be used by applications.

Not sure if it's the issue or not, but maybe you could look in your journal and see if you have the same messages? For what it's worth, I tried disabling umip as well but without it, the game refuses to start for me.

The game runs great on my 3080 otherwise (Nvidia driver 560.35.03), so it's really unfortunate that this is happening. It kinda makes the game unplayable right now.

SGDT stuff sounds like Denuvo stuff.

@Dwarsen Yes, I can confirm that I get the same entries in my kernel log (SGDT stuff). I could even trigger it at will by entering or exiting the in-game menus. (By the way, I also can no longer find the fossilize error messages in my logs, so I guess it might just have been an intermittent problem related to the shader cache that fixed itself eventually.)

(EDIT: SGDT seems to be just Denuvo as was mentioned above. I just played Sonic Frontiers which also has Denuvo and got the same log entries but that game performs flawlessly.)

Concerning FF XVI, I also did another iteration of tests, testing all different upsamplers, low to high settings, also tested disabling resizable BAR via the vkd3d "no_upload_hvv" setting, but to no avail. Hideout still stutters no matter the actual load on the GPU/. There's just something really off here. People on Windows do report similar issues though, so perhaps it's more of a bug than a Proton issue.

Still, any workarounds would of course be appreciated. :)

I'm seeing some fossilize messages but definitely not thousands of them and some SGDT too. I found that while the hideout had some stutters, the game just fell apart in Lostwing. Something about the fires seem to trigger the worst of it.

Edit: Can reproduce it so easily at Lostwing, look at the ground or a wall, 100% GPU and 50% CPU usage and expected frame rate/timing. Look over the lake, 0-40% GPU usage and 30% CPU usage and huge frame timing jumps

This thread suggests removing the in game shader cache which does seem to improve the situation on my system.

https://steamcommunity.com/app/2515020/discussions/0/4845399560303939096/

For example:

rm -f '/home/steam/games/saves/2515020/pfx/drive_c/users/steamuser/AppData/Local/SquareEnix/FINAL FANTASY XVI'/*.psol

Tried deleting the game's own shader cache but no improvements for the stuttering in the Hideout (which is presumably caused by the same problem as the stuttering even later in the game). On ProtonDB, there was a recent comment that suggests the Denuvo implementation might be responsible similar to what happened in Dragon's Dogma 2, but that's pure speculation at this point.

Tried deleting the game's own shader cache but no improvements for the stuttering in the Hideout (which is presumably caused by the same problem as the stuttering even later in the game). On ProtonDB, there was a recent comment that suggests the Denuvo implementation might be responsible similar to what happened in Dragon's Dogma 2, but that's pure speculation at this point.

Given the pattern of the stutters (very reproducible in transition screens, when the game just zooms in within a scene that is already rendered and was running at stable framerate before the camera zoom happens, all with CPU cores and GPU not at full load), it would indicate something in the general logic of the game is affecting the rendering loop indeed.

The main issues seem to be in populated areas. The hideout has minor issues but its a small enough place that is just a minor annoyance. Lost Wing and Marthas Rest on the other hand are a disaster. Even in the areas surrounding Marthas Rest, if the camera is in the direction of it you will end up playing a slideshow.

Replying to https://github.com/ValveSoftware/Proton/issues/8001#issuecomment-2365340990

I thought so as well. That was until I have dual booted into Windows and realized game runs almost flawlessly there with the same exact settings.

3060, all min settings, 1080p - slideshow on Linux, full speed on Windows.

Replying to [#8001 (comment)](https://github.com/ValveSoftware/Proton/issues/8001#issuecomment-2365340990)

I thought so as well. That was until I have dual booted into Windows and realized game runs almost flawlessly there with the same exact settings.

3060, all min settings, 1080p - slideshow on Linux, full speed on Windows.

Then we may be looking at a Wine / VKD3D / Mesa RADV issue.

I have been playing the full version of this game over the last week, should we continue to discuss Final Fantasy XVI full version in this issue thread or open a new one and leave this only to the Demo version? For now I will comment my experience around here.

My CPU is a AMD 3600X and my GPU is a AMD 6700 XT, I am running it on a Arch Linux with the latest kernel and mesa drivers. The game works, full stop. There is no crashes, no graphics defects or anything. However performance is a mixed bag.

With Dynamic Resolution and Frame Generation on I can run the game at 4k and almost 60 fps, the performance does dip on more populated areas but it remains about 30 fps and since it is outside the battles it is not really an big issue. If I try to run at lower resolution I can get higher FPS but the dips below 60 are still there which is at least a little strange.

But the big issue is that after about 26 minutes (I measured 3 times and it was fairly consistent) the frametimes become unstable, the stuttering is visible and it does not come back. I kept mangohud open for the whole session and the moment it triggers is very visible on the frametime plot (it goes from a fairly stable line to a noisy mess), FPS continue at the same level as before but it is very stuttery. I can just reset the game and the stuttering will disappear but if it triggers in the middle of a long boss battle there is nothing I can do without losing the save file.

I haven't tried to play the game at Windows so I don't know if the problems are Linux exclusive.

Adding to the previous comment, I ran a fourth test and again the stuttering started after 26 minutes and 50 seconds. Here is a proton log:

steam-2515020.log

I was running on Proton GE, but did this run on Proton Experimental. Default proton fails to compile shaders.

I never seen a problem like this before, it is very strange to me that a problem starts after a specific amount of time. Here is the line that appears everyone on the proton log:

20703.368:0188:018c:warn:vkd3d-proton:d3d12_device_QueryInterface: {a7aa6116-9c8d-4bba-9083-b4d816b71b78} not implemented, returning E_NOINTERFACE.

I don't know if it is the source of the problem, but I suspect that some minor error is building up on the background until it result in this stuttering thing and this is the warning that is happening most frequently. I considered a memory leak problem but VRAM and RAM seems stable.

It may be worth mentioning this game is criticized for being very inefficient, and many Windows users are reporting slowdowns (materializing as lower FPS, not as stutter) after playing for a while (how long appears to be inconsistent) causing major slowdowns that fix themselves by simply restarting the game.

OK I just looked at @Figuera 's log and a few things stand out:

1. a7aa6116-9c8d-4bba-9083-b4d816b71b78 is a debug D3D12 interface. In this same thread a tip mentions that the game appears to be actively polling for debug interfaces, likely as an anti-tamper mechanism, and that creating dummy empty files for them (such as renderdoc.dll) fixes the problem. While there seems to be a Wine issue at play (i.e. Wine taking too long to look for that DLL in all possible locations) you should try that workaround. Alternatively, it could be a Wine bug if QueryInterface takes too long to simply return E_NOINTERFACE and Wine should short-circuit to it.
2. Fossilize complains VK_STRUCTURE_TYPE_DEPTH_BIAS_REPRESENTATION_INFO_EXT is not recognized. This looks like either @Figuera 's Fossilize version is too old, or Fossilize needs to implement support for this structure. Otherwise some PSOs can't be cached and thus are going to cause stutter. This error appeared 907 times in his log, so it's definitely a problem.

Update: Support for VK_STRUCTURE_TYPE_DEPTH_BIAS_REPRESENTATION_INFO_EXT was added 3 weeks ago so a very recent version of Fossilize is needed.

@darksylinc

I am not very familiar with Fossilize, how do I check the version being used? and/or how would I try to update it?

Fossilize is included with Steam so it may be a matter of waiting or raise the issue in https://github.com/ValveSoftware/steam-for-linux

You may be able to build it from source https://github.com/ValveSoftware/Fossilize/ but I haven't tried that myself.

Creating dummy files (I followed this reddit thread) does not solve the problem.

I haven't had any luck with fossilize either, I can build the library but can't tell steam to use my version instead of its own.

any news on this?