595 release feedback & discussion

Below bugs have been filed internally and team is currently working on local reproduction.

6018056 Terrain textures appear blocky and grid-stitched in Captain of Industry after driver update 595.45.04

6017975 Incursion Red River has black screen issue with driver 595.58.03

6017831 Forever Winter locks up immediately upon shooting any drone enemy with driver 595.58.03

6017526 Crimson Desert still soft locks upon completing shader compilation part of loading (field of cubes) with driver 595.58.03

Hi @echoa

Please help to share nvidia bug report from repro state and also confirm if there is any last passing driver.

For me it passes shader compilation just fine.

Its when you press 'play’ and the loading screen appears it will hardlock/freeze the game, this may take a couple of minutes but it never gets past the loading screen.
This happens only on 50 series cards.

ER with the DLSS mod seems to work for me again on the new 595.58.03., although HDR is no longer functional.

Related to my previous post :

With 595.58.03, Elden Ring have the same issue and the same workaround.

With 595.58.03, Super Jigsaw Puzzle: Generations now works well.

Regards

I’m not able to test drivers or do the bug report ATM but to reproduce you just need to go to any of the first 3 zones in the game

Ashen Mesa

Tunnel A

Scorched Enclave

And either shoot the flying drones yourself or anyone else can shoot the drones be it other enemies or players. The game will immediately freeze and xid109 in systemd. This is with an RTX 5070 mobile aka 5060ti 8gb.

Specific system is an Alienware Aurora 16x running Arch Linux with kernel 6.19.9 and the latest KDE plasma on Wayland.

This can be avoided with PROTON_HIDE_NVIDIA_GPU=1 so its specific to UE5 + Nvidia in this title. Idk what UE5 does differently for nvidia GPUs.

Visual glitch from the Last of Us 1 & 2 are still present with latest drivers. Uncharted 4 doesn’t exhibit these issues and shares the same engine.

nvidia-bug-report.log.gz (338.3 KB)

steam-2531310.log (972.3 KB)

Last post

I am still experiencing persistent game freezes in The Witcher 3: Wild Hunt on 595.58.03 release. This issue started appearing with driver version 590.44.01.

The problem is triggered by enabling Full Ray Tracing (in my case Quality mode). It occurs during gameplay at random intervals, typically within a few to several minutes.

Several other users are also reporting the same issue, for example: @sorn.denis , @zebcom

The problem does not occur on any drivers from the 580 series.

AutoSave_11e177_7ea10c00_4c6dd0a.tar.gz (2.5 MB)
launch.log (56.2 KB)
nvidia-bug-report.log.gz (779.3 KB)

here is the output of nvidia-bug-report.sh

nvidia-bug-report.log.gz (535.0 KB)

now i cant reasonably go back many driver versions without changing kernel (i cant even get 575 to build for 6.19 anymore) and even then i can only go back so far without my laptop having issues. Do we want to roll back kernel, driver, and firmware versions all together? the lowest i might be able to go is 6.16 but this seems like changing way to many variables at once to really be a good test

EDIT: Luckily i didnt have to go far with this one, 590.48.01 doesnt have the issue

I am experience that the driver crash and all screens turn black, when gaming on a displayport monitor. I got a dual monitor setup with both screen connected to displayport.

After the crash I had to power down the computer (turn off power supply), wait 5 seconds, and turn on power supply and press power button to get any output on any screen next boot.

I was playing xcom2, civ 5, Witcher 3 (no ray tracing), Cyberpunk (no ray tracing), Expedition 33. Got crashes in all games, albeit very seldom on xcom2.

The problem disappeared when switched cables so I was gaming on HDMI, secondary still connected on displayport. Even though sometimes games crashes as described by @wierzbiniak95 above.

Graphic card MSI GeForce RTX 3070 VENTUS 2X OC LHR RTX 8GB.
OS Bazzite release 43

Kernel 6.17.7-ba28.fc43.x86_64.

Driver 595.45.04

Display LG Ultragear 2560x1440 @140Hz
Display S24C450 1920x1060 @60Hz

I’m seeing more people reporting Space Marine 2 issues on 595, crashing or unable to load the game past the intro splash logos

590 is last working version

nvidia-bug-report.log.gz (709.6 KB)

Thank you for taking the time! i hope after you find the bug we could get a quickfix driver release

Keep having “Xid 109 CTX SWITCH TIMEOUT” when I’m about to enter a saved game in The Last Of Us 2, even with stable driver 595.58.03 and Proton-GE Latest with my RTX 5060ti 16gb and running Bazzite OS. Here is the bug report.

nvidia-bug-report.log.gz (421.5 KB)

List of issues that still affect the 595 driver:

  • Random black screens when playing intensive games, sometimes resulting in a partial GPU driver crash. It is sometimes possible to switch to a TTY, but attempting to switch back to a graphical session will crash the driver for good.
  • Desktop memory usage is excessively high(over 2GB sometimes).
  • A large amount of GPU Memory is unaccounted for in nvidia-smi / NVML.
  • Applications crash when VRAM is full(maybe fixed?).
  • Shared RAM under Linux seems broken. It’s almost like there is an incredibly small amount(256MB) and once that’s gone things start crashing.
  • DLSS3 / DLSS4 / Smooth Motion is broken. It causes massive input lag and micro stuttering.
  • Gnome-shell / GDM can freeze on login from a fresh boot. Switching to a TTY and relogging in works around this.
  • Ray Tracing is broken on a lot of games, resulting in an XID 109 or some other error.
  • Some games show DLSS upscaling options but no frame generation(Portal 2: RTX). Maybe a Proton issue?
  • Dragging the HDR slider under Gnome can cause a driver crash(maybe fixed?).
  • DX12 games have a massive performance penalty.
  • VRR causes backlight strobing/flickering.
  • Shader cache doesn’t save properly, forcing it to be computing again.
  • Video decoding from a GTX 960 using Sunshine/Moonlight causes blocky artifacts.
  • HDR is broken.
  • Performance limiters are broken on Blackwell.
  • PRAGMATA and other games have graphical glitches(maybe fixed?).
  • Using a secondary dedicated GPU often fails with [drm:__nv_drm_gem_nvkms_prime_dup [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000c00] Failed to import NVKMS memory to GEM object . This error can happen regardless of whether you use dxvk.deviceFilter or in-game display adapter settings. Other times games start but run at 5 FPS.
  • Display sometimes fails to wakeup from sleep unless switched to TTY.

Aren’t those supposed to be fixed with the new update:

  • Applications crash when VRAM is full(maybe fixed?).

  • Shared RAM under Linux seems broken. It’s almost like there is an incredibly small amount(256MB) and once that’s gone things start crashing.

Are you sure they are still issues? Users here seems to say it’s fixed: Non-existent shared VRAM on NVIDIA Linux drivers - #153 by echoa

Correct, and HDR has started working as well. That user posts the same list of items in multiple threads, but does not provide any data to support their claims, does not provide bug report logs, does not include specific OS/driver/hardware versions etc.

Almost everyone else uploads bug report logs and provides specific, helpful test reports, so it’s disappointing that such behavior persists.

edit Just for another random example, the change log mentions that a bug was fixed that could cause flickering in VRR on some HDMI displays. If there’s still any bug with VRR and flickering, the exact monitor models, OS/library versions, driver version etc. would be needed along with the nvidia bug report archive upload.

Anything newer than 580 is out of support for maxwell/pascal too no? Wouldn’t the GTX970 need to be used with the 580 legacy drivers?

Right. The 590+ drivers is only for Turing (GTX 1600 series) and newer.

Regards

Keep forgetting to add that the wakeup from sleep bug still isn’t fixed. Added.

I can confirm this and have already opened a separate issue about it. I first noticed it with the 590 drivers and also came across a GitHub issue in proton where people are having the same artefacting.