I have GTX 470 - I’m not interested in gaming at all - only CUDA. However I really need CUDA on RDP, is it doable via this driver with modified inf files ?
many thanks
Mirko
Nope.
Nope.
Thanks, for info.
I wonder why don’t you enable it for G100 based cards ? It seems to me it is only matter of driver checks ?
thanks
Mirko
Thanks, for info.
I wonder why don’t you enable it for G100 based cards ? It seems to me it is only matter of driver checks ?
thanks
Mirko
Hi!
I have an ASUS Server with ASPEED for display and NVIDIA TESLA C1060 as a general purpose processor. I want to use MALTAB for GPU using MATLAB R2010b that supports NVIDIA GPU processing. Will TCC work in this configuration? Operating system is Windows Server 2008. Thanks
TCC should work fine in that case. Install a normal driver (with version >= 260), the C1060 will be in TCC automatically, and things should just work.
As TM said it will work without any problem.
There are very few issues with nVidia development toolkit & drivers. I have encountered only IUnknown & size_t issue that were giving me a bit of a headache when compiling code in VS 2010 on 2008R2. However it was a breeze to resolve.
M.
Thanks for your reply!
I have installed TESLA driver version 263.06 (that installs fine and no yellow exclamation mark in Device manager). However, when I try to access GPU through MATLAB R2010b using ‘gpuDevice’, I get the following error:
??? Error using ==> feval
An error occured during: setting up the CUDA device. The CUDA error was: (Unknown cod: 10100).
Error in ==> currentDeviceIndex at 7
idx = feval(‘_gpu_selectedDeviceIdx’);
Error in ==> deviceProperties at 8
idx = parallel.internal.gpu.currentDeviceIndex;
Error in ==> GPUDevice.GPUDevice>GPUDevice.current at 34
[props, E] = parallel.internal.gpu.deviceProperties();
Error in ==> gpuDevice at 18
dev = parallel.gpu.GPUDevice.current();
One of the forum members suggested to use nvidia-smi.exe utility to set the TCC mode. Unfortunately, I could not locate it on my computer. It is supposed to be located in (I am now using Windows 7) C:\Program Files\NVIDIA Corporation\NVSMI but there is no such folder in NVIDIA Corporation on my computer.
Any suggestions will be highly appreciated. Thanks
Dear moderator!
I have Tesla C1060 and cannot find nvidia-smi.exe utility to put it in TCC mode in C:\Program Files\NVIDIA Corporation in my Windows 7 as the folder NVSMI does not exist there. The drive I downloaded is 263.06 WHQL (file size 106 MB and its release notes specifically mention use of nvidia-smi utility for switching it to TCC mode). Please help me!!!
Hi CUDAReloaded,
We recently put up a blog post describing how to get Jacket working over Remote Desktop. It might help you solve this.
At the same time, you might check out how Jacket compares to PCT, here: http://www.accelereyes.com/products/compare
Good luck!
Best,
John
TCC is not supported on 32-bit OSes.
Thank you for the reply. I installed Windows 7 64-bit and TCC driver for Tesla C1060 (for Windows 7 64-bit) and now things are working fine.
Hi!
I have a new server running Windows Server 2008 R2 with 2 Tesla Cards C2070.
I would also like to use the CUDA 4.0 release candidate (which already proved fine on linux for our software).
And I need remote desktop access to the server.
Is the current TCC driver ok for this (270.61)?
Or do I need the dev driver from the 4.0 release candidate (and would this enable remote desktop?).
Or do I have to revert to version 3.2 of CUDA for the time being?
best regards Rolf
It would save a lot of time if NVidia mentioned this on the Release Notes of the driver in the TCC section. Since the same document is used for both 32 bit and 64 bit OSes.
Hi,
Can I use CUDA C in intel DPDK environment for using GPU with Intel DPDK.
regards
mati ur rahman
Hi,
Can I use CUDA C in intel DPDK environment for using GPU with Intel DPDK.
regards
mati ur rahman
Perhaps you know whether or not TCC mode is support under
windows server 2008 r2 64 bit OS
?