On Tue, 9 Aug 2011 03:14:04 +0200, Vojtěch Král <kral.vojtech@gmail.com> wrote:
Hey everyone,... From recent discussions about troubles with OpenCL as well as comments in AUR, forums, etc.. I got the impression that the whole thing is pretty confusing for an outsider... and for an insider too for that matter. So I put up a wiki entry.
So far it's a wip article on my wiki profile here: https://wiki.archlinux.org/index.php/User:Kralyk/GPGPU
I'd like you to + have a look at it and tell me what you think of it ;-) + check if my wording is tolerable enough - English isn't my native language, feel free to correct + See if you know anything about the (missing) Nvidia/CUDA parts I myself am not really acquainted with CUDA and Nvidia-flavoured OpenCL so I'd like to ask if someone CUDA-savvy out there would be so kind as to help out and fill in those gaps.
Regards, ~kralyk
Hi, That's a very good idea, and your article looks really good :) I added some stuff about CUDA (I'm the maintainer of the cuda-{toolkit,sdk} packages on the AUR) and Nvidia's OpenCL (not familiar with it though). There is room for improvements, but it should be enough for a start. HTH. Should this page also mention some {CUDA,OpenCL}-accelerated apps or language bindings such as PyCUDA and PyOpenCL? Regards, -- Thomas/Schnouki