Stack Exchange Network

Stack Exchange network consists of 174 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

Questions tagged [cuda]

Computing architecture developed by Nvidia that allows programs to process algorithms in the graphics card GPU, instead of the computer's CPU.

1
vote
3answers
651 views

Failed to install cuda on centos 7

Following the instructions to install the cuda found here: https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html e.g. $ sudo yum groupinstall "Development tools" $ sudo yum install ...
1
vote
1answer
65 views

NVIDIA HEVC 4k 5.0 level support

I'm trying to encode a 4K (3840x2160) stream with level less or equal 5.0 with no success. I'm using ffmpeg latest build, nvidia drivers 410.66. All dockerized on nvidia/cuda:8.0-devel-ubuntu16.04 ...
0
votes
1answer
162 views

How to check CUDA Compute Capability?

I just playaround with open-source software called Meshroom. https://github.com/alicevision/meshroom One of it's processes called DepthMap gives an error: This program needs a CUDA-Enabled GPU (with ...
1
vote
1answer
51 views

Could using CUDA toolkit potentially lead to damaging logic board or power supply?

I'm not too familiar with hardware side of computers so excuse the question if it's not phrased in the best way. Story is I bought this computer for the sole purpose of training machine learning ...
0
votes
0answers
44 views

Files getting corrupted every time

I am trying to download/copy a file. I tried from web and also from pendrive. md5sum for the two files ( one in pendrive and one that I copied on desktop) are not matching at all. I copied using cp ...
1
vote
1answer
276 views

Encode video with ffmpeg using CUDA without NVENC

There's quite a bit of info about using hardware encoder built in Kepler-based and newer NVidia cards to encode video. As far as I understood all of them just pass the encoding to hardware encoder, ...
0
votes
1answer
374 views

Nvidia graphic cards not detected after reboot

I have an NVIDIA GeForce 1080 Ti working fine on Ubuntu 16.04. After a power outage a couple days ago, it stopped working: nvidia-smi not working and GPU not found in lspci/lsmod. After rebooting ...
1
vote
0answers
94 views

GPU Power consumption

I am running a computer with 4 Nvidia Titan X for training deep neural networks. I have problems with my power consumption. The consumption of Titan X is not supposed to be above 250W but as you can ...
0
votes
1answer
27 views

Windows 7 not turning monitor on after screensaver

this issue started when I installed Cuda graphics (on installation it asks something about setting it to be the default graphics driver). Weirdly, every time the screen shuts off (after the delay set ...
3
votes
0answers
788 views

configure not finding ffnvcodec dependency

When running ./configure --enable-cuda --enable-cuvid --enable-nvenc --enable-nonfree --enable-libnpp --extra-cflags=-I/usr/local/cuda/include --extra-ldflags=-L/usr/local/cuda/lib I am met with ...
1
vote
0answers
1k views

ffmpeg compile error with Nvidia Cuda 9.0 [duplicate]

I'm compiling ffmpeg on Centos7, Nvidia driver 384.81, CUDA 9.0. I'm using a Tesla P4. Here is my config line: ./configure --enable-cuda --enable-cuvid --enable-nvenc --enable-nonfree --enable-...
10
votes
3answers
7k views

ERROR: cuvid requested, but not all dependencies are satisfied: cuda/ffnvcodec

I am trying to compile FFMPEG with Nvidia Cuda support, on Debian 9.3. Parameters, what I am using: --enable-cuda --enable-cuvid --enable-nvenc --extra-cflags=-I/usr/local/cuda/include --extra-...
0
votes
0answers
92 views

Multiple RDP users deep learning on Windows 2012 with SINGLE Quadro P5000

My work has budget to provide a HP Workstation with one Quadro P5000 GPU. Our team has around 6 members and are planning to use this workstation to do deep learning, i.e. PyTorch with CUDA. We are ...
2
votes
2answers
2k views

CUDA 9.1 installation error: does not have a version.txt file! Ensure you are installing to an existing toolkit directory

I'm installing the run file version of CUDA 9.1 on a linux box which has CUDA 8.0 on it. The file I'm installing is: cuda_9.1.85.1_linux.run sudo cuda_9.1.85.1_linux.run The download options were: ...
0
votes
0answers
123 views

How to use pts with burn subtitles

I'm using ffmpeg to encode an udp live stream with dual audio and subtitles. Nowadays I'm burning subtitiles (dvb_subtitles), and I want to respect pts table but I don't know how. Here is my sample ...
1
vote
0answers
139 views

Ubuntu CUDA driver only working when using graphics card for display

I am running Ubuntu 14.04 and have a Nvidia Geforce GTX 970. I have successfully installed both the graphics card driver and the CUDA toolkit and have been using it with Ethminer to mine Ethereum. ...
0
votes
1answer
64 views

Disable CUDA watchdog timer on Ubuntu Server

I have an Ubuntu 16.04 server running, with no X Windows. I want to disable the CUDA watchdog timer, so that I can run lengthy kernels. All answers on the matter refer to modifying Xconfig. I don't ...
0
votes
0answers
276 views

Tesla V100 GPU too slow

I am using ubuntu 16.04 with a tesla V100 GPU. I have installed CUDA 9.1 and cuDNN 7. The driver version is 387.26. The GPU is correctly recognized; it shows up with nvidia-smi. I am using it with ...
1
vote
1answer
786 views

Is GeForce GTX 1050 Cuda-Enabled?

I want to buy a laptop with graphic GeForce GTX 1050. In nvidia site 1050 desktop version is cuda-enabled, but it's not in notebook section. Does that mean it's not cuda-enabled? anyone has this ...
-5
votes
1answer
443 views

GT 730 specs comparison: CUDA Cores; 64/128bit; DDR3/GDDR5, MHz, Bandwidth, 1GB/2GB

I wonder why nVidia offers these type of configurations of it's graphics cards, in this case > GT730. I know that bandwidth = Memory clock x Memory interface width but what about memory config or ...
0
votes
1answer
1k views

libnppig.so.8.0 Missing FFmpeg

I've bought GeForce GTX 1050TI for testing FFmpeg with Nvidia encoder. Installed the requirements using CUDA Utils run file with these parameters:  (I'm using Debian; I can't change to supported ...
2
votes
0answers
87 views

Can one install a cuda-capable gpu in a Dell R900?

Does anybody have any experience putting a cuda-capable gpu in a R900? It doesn't seem to be officially supported, so I wanted to get some advice before proceeding.
2
votes
1answer
889 views

Is there a way of restarting/resetting a graphics card without disabling the device or restarting the computer? [duplicate]

(Found one similar question - but this causes Windows 10 to hang). I've been doing some machine learning and CUDA testing recently - Whenever I max my graphics card out for a while, I see a few ...
1
vote
0answers
265 views

Nvidia device missing after rebooting in Linux

I have a PC equipped with three NVIDIA graphics cards, and the OS is Cent OS 7. I have installed NVIDIA driver as well as CUDA toolkit. Normally, in /dev/, there are several nvidia devices: /dev/...
0
votes
1answer
4k views

Adobe Premiere Pro doesn't recognize the GPU

I'm using as a GPU the Nvidia GeForce GTX 1070, and I installed the NVIDIA CUDA software from the official NVIDIA website. When I open a project I always select (CUDA) for the rendering environment, ...
2
votes
4answers
15k views

Can't login to Ubuntu after installing CUDA

I want to install CUDA 8.0 in my Ubuntu 16.04. First I stopped the X server with: sudo /etc/init.d/lightdm stop Then I installed CUDA 8.0 from the official Nvidia website. (The graphics driver is ...
2
votes
0answers
57 views

After Nvidia software crash computer is 3x faster?

Around 30 minutes ago my nvidia software/cuda crashed and it opened my default browser to the support page for nvidia. It asked me to restart Adobe Premiere as the "CUDA renderer" was no longer ...
0
votes
0answers
277 views

CUDA compilation inOpenCV in Arch Linux

I'm trying to compile programs in OpenCV 3.2.0 with CUDA, I have a Graphic Card Nvidia Gforce 950M (Maxwell) and I use Arch Linux First of all, I had installed on Arch Linux X64 and the lastest ...
1
vote
0answers
57 views

Watching videos during CUDA computation on NVIDIA graphic card

I currently spent much of my PC time to compute Neural Nets using CUDA and my NVIDIA GTX 980 graphic cards. While CUDA works I have some kind of FPS drop and little freezes on my monitor. It actually ...
0
votes
1answer
421 views

CUDA fails after resume

CUDA (in this case Blender) works fine after clean boot, but upon resume from suspend errors with this from the CUDA client: CUDA cuInit: Unknown error and this in dmesg: NVRM: GPU at PCI:0000:01:...
0
votes
1answer
724 views

Which NVIDIA driver is suitable for GeForce GTX 660M?

I am using a Lenovo Y580 running Ubuntu 14.04/Windows 10, which has a GeForce GTX 660M Graphics card. I'm trying to train neural networks using Caffe and it would be really helpful if I could use my ...
1
vote
0answers
469 views

Webcam recording with JPEG CUDA codec in FFmpeg

Put simply, I have to capture a DirectShow webcam raw video as single JPEG images. I want to use NVIDIA GPU power for that. As far as I know from the FFmpeg hardware acceleration and from NVIDIAs ...
0
votes
1answer
2k views

ubuntu : cuda package not found (ubuntu 14.04, cuda 8.0)

Installing cuda on ubuntu 14.04: went to : https://developer.nvidia.com/cuda-downloads selected the options: linux, x86_64, ubuntu, 14.04 and deb(local) get the instruction: sudo dpkg -i cuda-repo-...
2
votes
1answer
1k views

how can I install CUDA on linux on a custom path (not in /usr/local)

Is there any good guide about how to install CUDA on linux on a custom path (other than /usr/local) ? I have already installed one version and I would like to install a second one. Also, how can I ...
0
votes
0answers
1k views

Bumblebee on Centos 7.2

Trying to install bumblebee on Centos 7.2. My system is an Asus laptop with integrated Intel graphics as well as a Geforce 960M, with no BIOS option for disabling the intel graphics module. I ...
0
votes
1answer
161 views

CUDA Performance problems depending on a OS

I have 2 machines that are exactly the same hardware-wise. On them, I have 1 Debian, and 1 custom build with buildroot linux image. The problem I'm experiencing is that: For my application, I'm ...
14
votes
3answers
20k views

How can I know about compute capability and sm of my Graphics card?

I know I can get the compute capabilty by just visiting this official cuda page, or this wiki page. But I dont know how I am supposed to find the sm of my card. Is this short for shader model? or ...
1
vote
0answers
39 views

Want to try CUDA program on windows

I have a CUDA program written on linux, when I open the .cu files in Visual Studio 13 on Windows 10 I don't get the build option. How can I run it in Visual Studio 13 on Windows 10 ?
2
votes
0answers
196 views

CUDA 7.0 and 7.5 drivers do not work on Debian Wheezy KVM guest

Hit across a roadblock and any help would be much appreciated! I have successfully managed to passthrough a GTX970 card to a KVM guest running Debian Jessie(3.16 Kernel), and am able to run CUDA 7.5 ...
1
vote
1answer
2k views

Cuda programming on Android Phone? [closed]

I'm looking for the most compact device that features a Cuda compatible device (optimal would be size of a smartphone / tablet). In the end I would like to install the Caffe Neural Network library on ...
0
votes
1answer
1k views

CUDA program ignores CUDA_VISIBLE_DEVICES variable?

I'm using CUDA 7.5 on a machine with Ubuntu Server 14.04 and 3 GPUs: 2x Tesla K20C and 1x GeForce GT 630. I've tried to set the CUDA_VISIBLE_DEVICES to the according IDs so that the code only runs on ...
4
votes
1answer
588 views

no CUDA-capable device is detected inside LXC container

I followed the following instructions and read the official guide in detail in order to set up Cuda inside an LXC container. When I try to execute the sample ./deviceQuery script from the official ...
1
vote
1answer
5k views

Failed loading CUDA library when trying to use ffmpeg with nvenc

On Ubuntu 14.04 I am following the instruction guide from NVIDIA's site to get ffmpeg running using GPU encoding. I followed the instructions (as far as I can tell), installed the NVIDIA SDK local ....
1
vote
1answer
104 views

Cuda toolkit draining battery windows 10

I installed CUDA toolkit 7.5 from NVIDIA (for Parallel programming course). But from the very next day, I've observed drastic change in the power consumption. Earlier the battery used to last for ...
0
votes
1answer
477 views

CUDA 6.0 does not detect Visual Studio

I am trying to install CUDA version 6.0 because I have a NVIDIA Geforce 9600 GT graphics card. I got the message No supported version of Visual Studio was found. Some components of the CUDA ...
3
votes
2answers
5k views

Ubuntu Box with multiple NVIDIA graphic cards

I recently bought a box from System76 that has multiple GPU's: one Quadro M6000, and two Tesla K40's. When I do lspci | grep -i nvidia it says 05:00.0 VGA compatible controller: NVIDIA Corporation ...
3
votes
1answer
3k views

What is the performance difference between workstation and desktop GPU for CUDA calculations?

Comparing workstation and desktop GPU cards, it is not clear what is actually different between the two. I understand that a workstation card is better for professional use and a desktop card is for ...
1
vote
1answer
1k views

no CUDA-capable device is detected (using ubuntu 12.04.4 server)

Migrating this post from stack overflow where it was closed because I feel it is useful I recently installed the cuda toolkit 5.5 with driver 331.67 (I have a GeForce GTX 680). For some reason, I ...
-1
votes
1answer
1k views

How to enable the disabled Streaming Processors (SM)?

I am using NVIDIA Quadro K2000 GPU. I ran deviceQuery.exe, the results of which are below. It says that I have only 2 SM units. I am not sure if I really have only 2 SMs or some of my SMs are disabled ...
0
votes
1answer
3k views

nvcc can't be found error

im trying to setup a tool that requires nvcc but im unable to to do. This is what i got so far: Installed the CUDA toolkit Verified it by using nvcc -V output: nvcc: NVIDIA (R) Cuda compiler driver ...