OTOY Forums

Hi,
NVIDIA has confirmed that the crash introduced with the 591.x drivers has been fixed in the latest build (591.74). If you’re still experiencing crashes, please try updating to the latest Studio driver, version 591.74.
Regards
Paride

Anybody have already test it ?

Please note that black screens and monitors switching to power-saving mode are still present. It is necessary to reboot the computer 5–6 times, but this can be resolved by setting Windows to Maximum Performance power mode. When using Balanced mode, monitor dropouts will occur. I am currently testing the driver; so far, everything is working fine.

Additional observations:

If the system manages to boot successfully (usually after 4–5 restarts), it then works completely stable — for hours, days, or even a full week — without any crashes or GPU-related issues.
This strongly suggests that the problem occurs only during the early boot / driver initialization stage, not under load.
I suspect that during startup the NVIDIA driver intermittently loses the DisplayPort signal:
The computer boots and continues running,
but the screens remain black from the very beginning,
while the system itself is not frozen.
Important detail:
Entering the BIOS is always stable, with no signal loss or black screens.
The issue appears only after Windows starts loading and the NVIDIA driver initializes.
Because the system is fully stable once the driver initializes correctly, this further supports the conclusion that this is a driver-related issue, not a hardware failure.
In my opinion, this is a faulty driver behavior related to:
DisplayPort initialization
GPU power state transitions during boot
or early-stage driver resource allocation

The computer has rebooted after a critical error.
Error code: 0x00000116
Parameters:
(0xffffcf0fee12c010, 0xfffff80179d0f140, 0xffffffffc000009a, 0x0000000000000004)

The memory dump was saved to:
C:\WINDOWS\Minidump\010926-9625-01.dmp

Report ID:
50a79b67-05d4-44cf-b32c-f8748651d8e8

This driver-related bug has existed in NVIDIA drivers since August 2024.
Initially, it affected a smaller number of users and appeared to depend on specific system configurations, motherboard BIOS versions, and hardware combinations.
However, with the release of the RTX 5000 series, this issue has become widespread and systematic.
The entire RTX 5000 lineup is affected, and no fully stable driver currently exists that completely resolves this problem.
Additionally, alongside the RTX 5000 series, older NVIDIA GPUs have also begun to experience the same issue, including:
RTX 2000 series
RTX 3000 series
RTX 4000 series
This indicates that the problem is not hardware-specific, but rather a driver-level regression introduced in newer driver branches, affecting multiple GPU generations.

I have complete shutdown of my computer, no matter the driver, no matter the Performance Power mode, no matter the time passed on a scene (it could be 10 minutes or 2 hours), but it happens everytime when I render a sequence of images. Thought it was a octane plugin's version issue but today (for the first time) I had it with the 16.10.

Could it be windows 11 related ?