Rhino & NVIDIA Graphics Driver Kernel Error causing crashes

Rhino
, , , ,
1

On multiple Dell Pro Max 16 Premium MA16250 laptops with Intel i9 and NVIDIA RTX PRO 3000 Blackwell Generation Laptop GPUs, 64GB of RAM we have been having the following issue in Rhino. On my own laptop, this issue got so severe today that I am now in the process of setting up a new laptop.

Essentially, after sometime working in Rhino in any model, doing anything whether it is rotating around in it or modelling, in any display mode, the following Kernel Error will appear:

image
image1920×1920 1.17 MB

After clicking “Yes”, Rhino closes, and the computer becomes unresponsive. Other applications that are using the Graphics card, such as Photoshop, Illustrator, etc., then come up with some error like this:

image
image413×192 11.5 KB

The only way to keep going is to force restart the machine… and then the cycle begins again.

It got so bad today, that just opening Rhino - not even opening a model - would cause this Kernel Crash.

I am sure some of this is some issue in my specific laptop, but the issue exists on many laptops. On other laptops however we have seemingly solved the issue for the most part by doing an uninstall and reinstall of Rhino, the NVIDIA graphics card, and making sure all the drivers are up to date. On my laptop this did not fix it.

Checking the reliability monitor it is always these errors:

image
image880×655 17.9 KB

image
image1429×602 19.1 KB

Running the laptop with other apps does not trigger this. It is only with Rhino.

We thought at first it may be an issue with the Dell Dock and Dell Monitor connection, and so I was trying first by just connecting to the monitor via an HDMI cable rather than the dock, but that did not remove the issue. I then tried no monitor at all, and just using the laptop, and still the issue appears.

I thought it may be due to the complexity of the meshing in some of my model, so I tried a much simpler model - just some closed brep boxes, but still that causes the issue to occur.

I thought it may be due to displaying meshes from Ladybug GH simulation results, but that did not seem to be the trigger.

Any help would be much appreciated since this is really frustrating.

2

Run systeminfo and post it here, that will help the developers.

Worth trying is repairing Rhino:
https://discourse.mcneel.com/t/repair-rhino-vs-uninstall-reinstall/210235/2

Check what GPU drivers you are on, some times newer are better, but Windows have changed how they handle performance for GPU’s a bit, taken over a bit of the control the drivers had before, so this can be the reason. Thus newest drivers might help.

I helped a user figuring out what caused issues, and went the “is it the dock, the cables, the drivers road” but repairing Rhino was what fixed it.

Good luck, and hopefully some more skilled comes along :wink:

3

Hi @Holo

Since the problem has been persistent over the last few months with several laptops we have had the luxury of trying multiple NVIDIA drivers, and always on the latest.

Have repaired / reinstalled Rhino multiple times, it sometimes alleviates the issue for a bit… but not sure if that is from Rhino being repaired or something else.

Because it happens with or without the dock, cables, I do not think it is the dock or the cables.

System Info:

Rhino 8 SR23 2025-9-8 (Rhino 8, 8.23.25251.13001, Git hash:master @ e75a9d62e23cc6f52d4ef68eb7eda9f8da9b6cd9)
License type: Commercial, build 2025-09-08
License details: Cloud Zoo

Windows 11 (10.0.26100 SR0.0) or greater (Physical RAM: 63GB)
.NET 8.0.26

Computer platform: LAPTOP - Plugged in [100% battery remaining]

Hybrid graphics configuration.
Primary display: Intel(R) Arc™ Pro 140T GPU (32GB) (Intel) Memory: 2GB, Driver date: 9-19-2025 (M-D-Y).

Integrated graphics device with 4 adapter port(s)

  • Windows Main Display is laptop’s integrated screen or built-in port
    Primary OpenGL: NVIDIA RTX PRO 3000 Blackwell Generation Laptop GPU (NVidia) Memory: 12GB, Driver date: 12-18-2025 (M-D-Y). OpenGL Ver: 4.6.0 NVIDIA 591.64

Integrated accelerated graphics device with 4 adapter port(s)

  • Video pass-through to primary display device

OpenGL Settings
Safe mode: Off
Use accelerated hardware modes: On
GPU Tessellation is: On
Redraw scene when viewports are exposed: On
Graphics level being used: OpenGL 4.6 (primary GPU’s maximum)

Anti-alias mode: 4x
Mip Map Filtering: Linear
Anisotropic Filtering Mode: High

Vendor Name: NVIDIA Corporation
Render version: 4.6
Shading Language: 4.60 NVIDIA
Driver Date: 12-18-2025
Driver Version: 32.0.15.9164
Maximum Texture size: 32768 x 32768
Z-Buffer depth: 24 bits
Maximum Viewport size: 32768 x 32768
Total Video Memory: 12227 MB

4

Intel ARC (ARM) processors are not supported as far as I know.

5

Rhino is set to use the NVIDIA Graphics card though, not the Intel ARC Processor

6

ARC isn’t ARM, it’s their GPU.

7

The Intel Arc 140T GPU is the far more powerful replacement for the old UHD stuff. It shouldn’t be bothering anything. In fact, Rhino 8 was planned to be compatible with them for rendering.

Anyway, at at your own risk… you could try to use the DDU to completely uninstall your Nvidia drivers; and then start afresh.

https://www.guru3d.com/download/display-driver-uninstaller-download/

Nvidia driver stability has been questionable since Blackwell release. I haven’t updated mine in ages, owing to some reports of RTX 4090 bricking. But I hope they are passed this now.

8

Hi Remy,

This is a well-known error that is experienced in Blender, Autodesk, games et al, as well as Rhino. It’s something to do with overcommitting resources. There are several things to try that could help and a quick google of “nvidia rtx kernel error code 3 subcode 7” should reveal them.

HTH
Jeremy

10

@jeremy5 thanks, I have tried most of these things, and for most of the computers we had this issue on it solved it, but for mine and some others it did not.

Is there a way to force the computer to not overcommit resources?

@David53 we have also tried this multiple times, and again, it worked for some computers and not others.

11

Remy,

Sorry you don’t have a solution yet. I’d suggest looking at what is different about the machines that are fixed from those that are not and where there is a difference, is the value on the faulty machines common to all of them.

Start with the obvious like model, generation, spec, firmware version, os version, drivers versions, power bricks, power management settings, NVIDIA software loaded, games performance settings in the os and video s/w. (You’ll probably want to spreadsheet all this!)

If you don’t find any differences, I would recommend involving the laptop manufacturer given this is not application specific.