Community Profile

photo

Ben Tordoff

MathWorks

Last seen: 2 meses ago Active since 2011

Statistics

All
  • GitHub Submissions Level 3
  • Personal Best Downloads Level 4
  • Editor's Pick
  • First Review
  • 5-Star Galaxy Level 5
  • First Submission
  • 3 Month Streak
  • Knowledgeable Level 2
  • First Answer
  • Solver

View badges

Content Feed

View by

Answered
Could I use Convolutional Neural Network in Neural Network toolbox with GPU card of capabilty less than 3.0
Hi Serghei, I'm afraid the answer is no. Neural Network Toolbox uses NVIDIA's cuDNN library for running Convolutional Neural Net...

más de 5 años ago | 2

| accepted

Answered
Why can't MATLAB parse function declarations in CUDA C files containing the "restrict" keyword?
Although __restrict is supported by Visual C++, the NVIDIA CUDA programming guide specifies the use of __restrict_...

más de 7 años ago | 0

| accepted

Answered
Matlab GPU processing: Nvidia Quadro vs Geforce
Hi Alex, I don't have access to either of these cards to test, but here's some information that might help you: * The "Com...

más de 7 años ago | 1

Answered
Why do I need 'wait' commands to assess GPU performance in the Parallel Computing Toolbox 6.0 (R2012a)?
If you have R2013b or above, an even better way to measure the time taken for a function is to use GPUTIMEIT (for GPU code that ...

más de 7 años ago | 0

Answered
GPU FFT2 Discrepancies ( Different from CPU based FFT2 )
See the answer(s) to this question: <http://www.mathworks.com/matlabcentral/answers/1032-why-do-some-calculations-like-the-fft...

casi 8 años ago | 1

| accepted

Answered
CUDA_ERROR_UNKNOWN when using floats instead of double precision
Thanks for sending the code. I’ve done some initial investigation and it looks like you have an illegal memory access somewhe...

casi 8 años ago | 1

| accepted

Answered
Getting NaN when evaluating PTX kernel file
First thing to do to track this down is to clearly specify your inputs and outputs. Inputs should be const. Based on your descri...

alrededor de 8 años ago | 1

| accepted

Answered
How to fix error : unrecognized token in Cuda code with nvcc compiler
There are two problems with what you have attempted, one just a typo and the solution to the other will depend on what you are t...

alrededor de 8 años ago | 0

| accepted

Answered
getframe issue in for-loop.
Hi Mike, you could try inserting a 'drawnow' just before the 'getframe'. This tells all MATLAB graphics and events to flush. ...

más de 8 años ago | 0

| accepted

Answered
Memory leak while using gpuArray in parallel computing toolbox 2013a
Hi Laith, I believe this should be fixed by the following workaround: <http://www.mathworks.com/support/bugreports/954239> ...

más de 8 años ago | 0

Answered
Matlab 2013a GPU memory leak
Hi Michael, could you read the following bug-report and try the workaround it contains (being careful about the backing-up step!...

más de 8 años ago | 0

| accepted

Answered
How to measure GPU memory bandwidth ?
Hi Anterrieu, you might like to have a look at the following article: <http://blogs.mathworks.com/loren/2012/12/14/measuri...

más de 8 años ago | 2

| accepted

Answered
CUDA: faster indexing methods than logical for large arrays?
You can sometimes get a speed-up by replacing the logical indexing with some element-wise maths. For example: trian_x = M.*...

más de 8 años ago | 0

| accepted

Answered
CUDA Kernels and long vectors
If you want to go down the "x" dimension first, you probably want int const globalBlockIdx = blockIdx.y * gridDim.x + block...

más de 8 años ago | 0

Answered
Cuda PTX kernel function boolean argument not supported?
Thanks for providing the compile line. It seems that recent versions of NVCC switch from mapping bools as "s8" to "u8" according...

alrededor de 9 años ago | 3

| accepted

Answered
error when using gpu
Unfortunately, an expression like: gpuArray(1):gpuArray(10) won't work, and (as Walter says) it looks like one of your "...

más de 9 años ago | 2

| accepted

Answered
gpuArray and memory management
Hi Gunnar, this is more a work-around than an answer, but try inserting a "wait(gpu)" after freeing the memory. For example: ...

más de 9 años ago | 0

Answered
GPU implementation of Min function
All syntaxes for MIN and MAX are supported in Parallel Computing Toolbox starting in release R2011b.

más de 9 años ago | 1

Answered
CUDA toolkit version | using constant memory
Hi, I will tackle the second question first as I am pretty certain of the problem. __constants__ must be defined at the global s...

más de 9 años ago | 0

| accepted

Answered
GPU time slower than CPU time, what went wrong with my GPU implementation?
Hi Ruby, I've just uploaded a benchmarking tool to the File Exchange which runs a whole load of these type of timings to put ...

más de 9 años ago | 1

| accepted

Answered
GPU and CPU computing results different
if you look at the magnitude of the difference compared to the magnitude of the values, it is pretty small: >> max(max(gathe...

alrededor de 10 años ago | 4

Answered
GPU implementation of Min function
Hi Sarah, if your MATLAB license is up to date you might like to have a look at the pre-release of R2011b which has just become ...

más de 10 años ago | 1

Answered
Indexing in GPUArrays
Hi Jesse, if your MATLAB license is up to date you might like to have a look at the pre-release of R2011a which is now available...

más de 10 años ago | 1

Answered
2D Graphs behind each other in 3D. Something like bar3
Hi Mirco, the way that PLOTYY works (and most of the other multi Y-axis MATLAB plots I've seen) is to have multiple axes obje...

más de 10 años ago | 0