Answered
gpuArray memory requirement estimation
In answer to your other questions: 1. Would it be useful to get a separate NVIDIA card, dedicated to parallel processing? ...

11 years ago | 1

Answered
Failed to complile MEX-Functions Containing CUDA Code
The problem appears to be that you're using a more recent version of the CUDA Toolkit than is supported by MATLAB. You are using...

11 years ago | 1

| accepted

Answered
gpuArray memory requirement estimation
Hi. It sounds like you're using the same card for both graphics and computation - this will put quite severe limitations on the ...

11 years ago | 1

| accepted

Answered
Is it possible to run Matlab with Parallel Computing toolbox on Amazon GPU ?
Looks like you might be using Windows Remote Desktop, is that right? Try using an alternative remote desktop tool like UltraVNC ...

12 years ago | 2

| accepted

Answered
CUDA_ERROR_UNKNOWN
The CUDA runtime is not fabulous about keeping track of the originating error in your kernel and it often falls back to CUDA_ERR...

12 years ago | 0

Answered
Why does the GPU load increase over time?
See the <http://www.mathworks.com/help/matlab/matlab_prog/techniques-for-improving-performance.html documentation on pre-allocat...

12 years ago | 1

| accepted

Answered
How to use GPU to calculate double integral with multivariable function
The function |integral2| is not supported for |gpuArray| data. You could implement your own version using supported functionalit...

12 years ago | 1

| accepted

Answered
nvcuda.dll is missing
Your AMD GPU is not supported by Parallel Computing Toolbox (PCT). The current version of PCT supports CUDA-enabled NVIDIA GPUs...

12 years ago | 1

Answered
Can I use Jacket(Matlab+GPU) OpenCL to use the ADM ATI GPU card for scientific calculation
MATLAB supports CUDA and not OpenCL at this point. Jacket was discontinued in 2012.

12 years ago | 1

Answered
Help with stereo matching and gpu, migrating code to reduce processing time
|gpuArray| is not suited to serial computation on matrices of small size. Convert your images to |gpuArrays| at the start, and t...

12 years ago | 2

Answered
Strange: Execution time per iteration increases, when GPU arrays are being used
As Jill points out, your timings are basically meaningless because you are using |tic| and |toc|. |c = a.*b| returns as soon as ...

12 years ago | 1

Answered
Matlab automatically work with multiple graphic cards?
To make use of both GPUs in parallel, open a parallel pool with two workers: parpool('local', 2); Use |gpuDevice| to sho...

12 years ago | 1

Answered
How do I optimize this code to run efficiently on the GPU?
Are your matrices always 4x2? This results in AT*A being 2x2, so you can just calculate your inverses manually: function Ai...

12 years ago | 0

Answered
Feature Tracking of a sequence of images
If you have Computer Vision System Toolbox and Image Processing Toolbox, there are examples on how implement feature trackers in...

12 years ago | 0

Answered
Can this GPU code snippet be redone without nested loops?
The GPU isn't going to work well with your nested loops. This looks like a classic case for |bsxfun|: matrix1 = permute(mat...

12 years ago | 3

| accepted

Answered
How to debug CUDA code in MEX function
Some instructions for Windows and Visual Studio can be found <http://www.mathworks.com/matlabcentral/answers/88541-failed-to-att...

12 years ago | 0