I have a TESLA C1060 GPU on my ASUS server running Windows 7 as non-display GPU. I have downloaded and install driver version 263.06 WHQL on my computer. Now I want to put Telsa in TCC mode but can’t find nvidia-smi.exe utility anywhere on my computer (no folder such as NVSMI under NVIDIA Corporation in Program Files)?
What am I doing wrong? Can anybody help.
I am currently in dire straits as I can’t put my TESLA C1060 into TCC mode on an ASUS server with ASPEED graphics card. I can’t find nvidia-smi.exe utility even after installing Tesla C1060 WHQL driver (file size 106 MB) for Windows 7. It claims to have TCC ability? Anybody who can help me please???
Hi, I´m working with a couple of nvidia tesla K8 under windows 7 64 bit. My version of NVSMI seems to be blocked, when I double click on nvidia-smi.exe it pops up but then vanishes…I read the cause could be that my drivers aren´t matched, but what do I exactly have to match? I have installed CUDA toolkits, both 6.5 and the most current one, the 8. What older driver do I have to revert to in case that´s the issue? Thanks
With a default Windows installation, nvidia-smi should be here:
C:\Program Files\NVIDIA Corporation\NVSMI
This should get refreshed every time you install a new driver. I have the latest WHQL drivers 377.55 installed and the time stamp of nvidia-smi.exe is 7/13/2017.
I could not find neither nvidia-smi.exe file nor C:\Program Files\NVIDIA Corporation\NVSMI folder in my lenovo thinkpad E470 laptop with windows 10 OS, NVIDIA Geforce 940MX GPU.
I have installed
Visual Studio Community 2015 with Update 3
CUDA 8.0
nvcc --version output is
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2016 NVIDIA Corporation
Built on Mon_Jan__9_17:32:33_CST_2017
Cuda compilation tools, release 8.0, V8.0.60
gcc --version output is
gcc.exe (GCC) 4.7.0 20111220 (experimental)
when I run deviceQuery.exe, I am getting output as below
CUDA Device Query (Runtime API) version (CUDART static linking)
cudaGetDeviceCount returned 38
→ no CUDA-capable device is detected
Result = FAIL
whenever I want to import keras, I get warning as below
“WARNING (theano.sandbox.cuda): The cuda backend is deprecated and will be removed in the next release (v0.10). Please switch to the gpuarray backend. You can get more information about how to switch at this URL: Converting to the new gpu back end(gpuarray) · Theano/Theano Wiki · GitHub
WARNING (theano.sandbox.cuda): CUDA is installed, but device gpu is not available (error: Unable to get the number of gpus available: no CUDA-capable device is detected)
Using Theano backend.”
Sometimes gpu works and sometimes does not work after some minutes(installing cuda again and running oceanfft.exe). I am getting frustrated for the cuda set up properly. Please help me to resolve the issue.
laptop GPUs are not necessarily available all the time due to Optimus design characteristics. You’ll need to investigate how to get your optimus laptop to work for CUDA. There are many questions on these forums that may help.
I installed the latest driver 388.71, cuda 8.0, and cudnn v6 on my laptop with Windows 10 and NVIDIA Geforce 940MX GPU. I can run TensorFlow codes with GPU. But I cannot find the nvidia-smi.exe and NVSMI folder.
Oh, thanks again though. To be honest I feel like this is out of my league at this point, I was just trying to use the AI Optix 5.0 denoiser with my machine, I use 3dsmax with iray, but it looks like this hasn’t been implemented yet (into iray for 3dsmax). I have no clue how to program at all, I need to learn a lot, this is all new to me. Thanks again, I appreciate your answers