OpenCL with Python(interpreted) or C/C++(compiled) ?

Hi all,

When should I develop OpenCL with Python(-> pyopencl) knowing that it’s a interpreted language. And same question with C/C++ compiled language.

Does the fact that Python is a interpreted affect the performance on execution.

What matters is mostly performance on big project : image processing, 3D visualization, etc…

Thanks in advance.

PyOpenCL seems to be a wrapper around the OpenCL API. You will still call the “original” OpenCL library functions and your kernels will get executed just like in C/C++.
It therefore shouldn’t make any difference at all whether you use Python or C++ in terms of performance of the OpenCL code.

I must say that I haven’t work with PyOpenCL myself, but this seems to be my impression.

Thanks for replying.

I think that the interpretation time is negligible beside computation time of processing.