Can I simulate this example (diffraction pattern) on my CPU not my GPU?

Question

Muhamed Sewidan 2020-9-1

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/587297-can-i-simulate-this-example-diffraction-pattern-on-my-cpu-not-my-gpu

评论： Muhamed Sewidan 2020-9-1

I tried to apply this example but it doesn't work, because I don't have an external Nividea GPU, I have an intel internal GPU.

Thus, can I apply this example with my CPU.

https://www.mathworks.com/help/parallel-computing/examples/using-fft2-on-the-gpu-to-simulate-diffraction-patterns.html;jsessionid=90bcd5410dbfbdb095d779eed228?fbclid=IwAR2iNNpYvOQ0q1esCj0mE83WKjPhjCNMFLDY1XF9klRnzDnwUIB4vsoVneI

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Raymond Norris 2020-9-1

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/587297-can-i-simulate-this-example-diffraction-pattern-on-my-cpu-not-my-gpu#answer_487898

在 MATLAB Online 中打开

Hi Muhamed,

In the best situation, by simply casting your data to a GPU, the code following the data creation will run, so long as all of the functions are overloaded for GPU arrays. For example

% CPU version
a = rand(10);
b = fft(a);
% GPU version
a = gpuArray.rand(10);     % Create GPU array instead of a standard numeric array
b = fft(a);                % Call your code as you normally would

In the example you've highlited, the code should run fine by just making the following change

[gx, gy] = meshgrid( gpuArray.colon( -1, 1/N2, (N2-1)/N2 ) );

to

[gx, gy] = meshgrid( colon( -1, 1/N2, (N2-1)/N2 ) );

or what would be more common coded as

[gx, gy] = meshgrid(-1:1/N2:(N2-1)/N2);