Error as above, and render stops.
its an old (gpu) scene which rendered just fine on gpu before, in max 2018, and vray next..
ive reopened it to do an update, in max 2021/vray 5
it opened, and rendered ok, but missing some textures as i didnt have multitexture installed on 2021 (i used it for all the brick materials in original scene).
i downloaded multitexture and installed it for 2021..
now i get this error and it wont render.
whats up there?
Is it possible to send over the scene in question via our contact form so we can test? Make sure to mention this thread in the email, thanks!
hi, i could not log on to the contact form (timeout error) so i sent a stripped scene via email to support@chaosgroup.com
as an update ive found that switching from bucket sampler to progressive fixes the problem (kind of, my machine shut itself down during the render, but i suspect thats my crusty old titan X gpus rather than this issue itself)
I have the same issue with my GPU rendering. After reinstalling drivers few times to 461.72 on RTX 2080 Ti was stable for a while, but now render crashes after 3-4 frames of animation.
In other exterior scenes feels that there is the same problem after a few tryes of rendering.
Guys, what is happening? This month I have so many issues with VRay…
BTW. Max 2021, Vray 5, newest update..
I’m getting the exact same error with a brand new GPU RTX A4000.
Started randomly today, have tried new drivers and upgrading to latest version, also banging my head against the wall as I have a deadline tomorrow, please help!!!
@trainofthought Try this build here,
This error was fixed a while back and I haven’t been seeing it
Thanks for the help. This error went away for a while, but now seems to be back - especially when using Vray GPU RTX on the pair of Ampere cards I have. Have tried the latest build and recommended drivers, but this issue persist and seems to be something to do with Optix building kernels for the device. I do wish Vray GPU was more stable, particularly with multiple GPUs of varying configurations.
--
i7 7700K, 64 GB, RTX A4000, RTX 3080Ti, Windows 10, Maya 2020.4
Hello I am getting this error “[MemoryManagerGpu::releaseUnmappedMemoryFromInstanc e] 700: Could not release device buffer”
I am running the drivers recommended on the Vray GPU page and an RTX 4090, usually after a few frames of an animation as someone else stated above it throws this error and stops. Running on CUDA.
EDIT: Upgraded to latest studio drivers and is way more stable (also realised I was on the game-ready driver version of the recommended, not sure if that makes a difference)
Hello. We are getting this error with V-Ray 7 in a production scene. We are not using the recommended driver, because recently there was an announcement by Nvidia of potential risk of hacking due to security holes in the older drivers.
CVE‑2024‑0126
Severity: High
Description: NVIDIA GPU Display Driver for Windows and Linux contains a vulnerability which could allow a privileged attacker to escalate permissions. A successful exploit of this vulnerability might lead to code execution, denial of service, escalation of privileges, information disclosure, and data tampering.
Is there something we can do to avoid this apart from installing the recommended driver? We switched to the RTX engine and so far there are no “700” errors.
We are taking a look at this now
We are also in the process of updating the recommended drivers, hopefully by the end of the week if everything goes well
Best,
Muhammed
Thanks, @Muhammed_Hamed !
Hi, @Muhammed_Hamed . Is there any progress on this issue? I’m still getting this error.
Hello,
We are also experiencing a LOT of stability issue with our Vray7 test.
Full CUDA, Vray 7 7.00.05 32856, dual 3090 with nvlink, win10, max 2025
Stability issue is when using of basic tools during IPR. Sometime it crash because we switch a material, or add a light, or even moving a basic geo … which is not at all the case in Vray6 (mostly used in RTX)
Either it will get a red CUDA error or simply crash and shutdown max.
Nicolas
There is a current Nvidia issue with these drivers causing memory leaks. Thread discussion here. Try reverting to 552 or 551 and see if there’s a difference.
The antivirus false-positives issues were already addressed here. If the vdenoise.exe is still quarantined, try allowing the tool and see if the error persists.
Thanks for the fast answers, I was already following those threads in parallel.
For the memory leak, I though the memory leak was only in RTX mode, not in CUDA, but I might have wrong there. Since there is a strong security issue with the 552 & the 551, I might wait something something that combines safety and stability.
I also knew for the false positive, we tried to reinstall all our Vray with no new alert, but I might have check the quarantine log on every computer.
Best,
Nicolas
Pardon me, didn’t notice you were using CUDA.
Otherwise, according to the article on this type of error, it might be a good idea to test-run a benchmark to see if it is indeed a V-Ray issue. Maybe the NVLink has something to do with it. Does rendering the scene in question yield issues on another workstation? If yes, do send the scene so we can take a look.
I’ll try to switch for one of our work station with single 3090 see how it goes.
If I can’t pin point it, i’ll send you the scene (which is really very basic)
