A pre-release of GPULib was made available today. GPULib is a library of common scientific computational routines that use modern graphics processing units (GPUs) to improve performance. I saw speedups of 5-50 times faster than normal IDL code for example code I was writing. Vectorized code that is still not fast enough can be improved with GPULib.
Bindings for Java, Python, MATLAB, and IDL are available. The IDL bindings are fairly simple to use and have a compatibility mode that allows running the code even without a GPU (performance was approximately equivalent to normal IDL code for the examples I wrote).
The documentation lists all the routines available (IDLdoc 3.0b3 spotted in the wild!).
Full disclosure: I work for Tech-X Corporation and worked on the IDL bindings and examples for GPULib.
January 2nd, 2008 at 2:28 pm
Sorry, the recent database crash has destroyed some comments on this thread. I’ll try to put back my responses.
January 2nd, 2008 at 2:32 pm
> Is GPUlib available for Windows? I don’t see a DLL in the download.
Windows is definitely available. Only the source is included in the download. Contact me if you would like a Windows binary made for you.
See this page for a list of supported graphics cards.
March 11th, 2008 at 3:11 am
Is there any link for the windows binary, i feel very interesting on wrapping the binary dll for C#, so it could benifit more ppl.
March 11th, 2008 at 3:33 pm
You can use the build_win.bat script in the IDL directory (you may need to modify that script) to build on Windows (you will need MSVC version 7.1 or 8). If that doesn’t work for you, let me know and we can make a Windows build.
March 13th, 2008 at 12:14 pm
I forgot to mention it when it came out; there’s now a blog for GPULib.