On Fri, 27 May 2011 09:45:42 -0400, Neal Becker <[email protected]> wrote:
> I am thinking of trying pyopencl rather than starting with pycuda.  It seems 
> pycuda has more activity (judging from the mail lists).  Which do you 
> recommend?

I just noticed that the download numbers are the exact
opposite--PyOpenCL gets many more downloads than PyCUDA. 
(check for yourself:
- http://pypi.python.org/pypi/pycuda
- http://pypi.python.org/pypi/pyopencl
)

Personally, I obviously would like to see OpenCL succeed. 

It has the right ingredients as a standard--mainly run-time code
generation and reasonable support of heterogeneous computing. On top of
that, being in a multi-vendor marketplace is a good thing--also for
Nvidia, although they might not immediately see it that way.

If I was starting something new, I would likely go with OpenCL, unless I
desperately needed one of the proprietary CUDA libraries.

The two Py* packages are roughly at feature parity, and will stay that
way for the foreseeable future. As in, if you're using PyCUDA, there's
no reason for concern--it'll continue to be supported. For instance, I
just added support for the new things in CUDA 4, and the 2011.1 version
of both packages will come out soon.

Andreas

Attachment: pgptPuVMy4wXb.pgp
Description: PGP signature

_______________________________________________
PyCUDA mailing list
[email protected]
http://lists.tiker.net/listinfo/pycuda

Reply via email to