Hello,

I was playing with the sample codes of JOCL and I also wrote some by myself and ported some from other tutorials to JOCL (like NBody problem).

However I think for very common mathemtical/numerical taks it would be better to use some well optimized code developed by some experts than try to write it by myself.

So I wonder if there is some possibility to use some 3rd party OpenCL libraris in JOCL for taks like

matrix operation ( multiplication, LU decomposition, inversion, solution of system of linear equation, eigenvalue ) and FFT.

for example libraries like this:

http://developer.amd.com/tools-and-sdks/heterogeneous-computing/amd-accelerated-parallel-processing-math-libraries/

The problem is that these libraries are in C, and I don’t want to do development in C just because of that. I’m currently working with jMonkey and LWJGL and I like it very much including the JOCL support. However, some well optimized prefabricated GPU accelerated mathematical library would be very welcome. a