I've just added a Mandelbrot example to the wiki:
http://wiki.tiker.net/PyCuda/Examples/Mandelbrot

It uses a pure GPU solution (ElementwiseKernel) and two CPU numpy
solutions, it was based on an older Numeric Mandelbrot example from a
couple of years back. Both the GPU and CPU solution use complex
numbers, I have it running on WinXP, CUDA 2.3 and pyCUDA 0.94 with
Visual Studio 9.

Could someone take a look and make sure the GPU solution is a 'good
answer' to the problem? I spent yesterday hacking at it, it does what
I expect but I'm still very much learning. Possibly instantiating 0+0j
complex numbers in the ElementWise kernel is a slow operation?

Regardless it scales very nicely, a 2000x2000 plot takes 2.1s with the
GPU and 51s with the fast numpy code.

Cheers,
Ian.

-- 
Ian Ozsvald (A.I. researcher, screencaster)
i...@ianozsvald.com

http://IanOzsvald.com
http://morconsulting.com/
http://TheScreencastingHandbook.com
http://ProCasts.co.uk/examples.html
http://twitter.com/ianozsvald

_______________________________________________
PyCUDA mailing list
pyc...@host304.hostmonster.com
http://host304.hostmonster.com/mailman/listinfo/pycuda_tiker.net

Reply via email to