I have just started to render my images in high resolution (3000 pixels) and i have noticed VrayRT(production) is constantly crashing after a while when rendering the full image.
I use noise Max. Noise under Adaptive sampling.
If i render a small region, the render gets completed but not when i render the full image.
Here is the crash report. Any ideeas what could cause that ?
Problem signature:
Problem Event Name: APPCRASH
Application Name: vray.exe
Application Version: 3.0.7.0
Application Timestamp: 537fac4d
Fault Module Name: KERNELBASE.dll
Fault Module Version: 6.1.7601.18409
Fault Module Timestamp: 5315a05a
Exception Code: 80000003
Exception Offset: 0000000000033ca2
OS Version: 6.1.7601.2.1.0.256.1
Locale ID: 1033
Additional Information 1: aeec
Additional Information 2: aeec410f6760400ebf1c2e90108fbcb6
Additional Information 3: d5d5
Additional Information 4: d5d5133d99cb0f2bfa9d497d01c3b339
I’m using:
3DS Max 2015 sp1
VrayRT 3.00.07
Windows Ultimate sp1
Thanks for the help Vlado.
The scene is 71mb large so you will have to download it from our FTP.
I have send you an email with all the required details.
I just got the crash again and i can confirm it happens only when I use Noise Max.
Otherwise, without using Noise Max, it can render indefinitely and completely stable.
The error i’ve got is: Driver API error: 700
So this could very well be related to my drivers I suppose but I can confirm that all applications were clean installed(including the drivers) on a recently installed Windows.
Here’s the printscreen:
I couldn’t really find anything on google to my knowledge even though there are various issues regarding this error.
On CUDA developer website, they say: CUDA_ERROR_LAUNCH_FAILED = 700
An exception occurred on the device while executing a kernel.
Common causes include dereferencing an invalid device pointer and accessing out of bounds shared memory.
The context cannot be used, so it must be destroyed (and a new one should be created).
All existing device memory allocations from this context are invalid and must be reconstructed if the program is to continue using CUDA. https://www.clear.rice.edu/comp422/resources/cuda/html/cuda-driver-api/index.html