> Is MATLAB actually 8 times faster than Sage at computing matrix > products and eigendecompositions than Sage on exactly the same > matrices?
Not exactly the same matrixes since to save time I'm using random matrixes. For both 2048x2048 and 8192x8192 matrix multiplication (64- bit float), I'm seeing 4x increase in speed between Matlab set to use 8 threads (top reports 800% CPU; any number of reasons can be proposed why this 8x isn't observed) and Sage/Numpy. > You'll have to figure out how to build ATLAS multithreaded. That's > not really a Sage question, but an ATLAS question. Try the ATLAS > docs, the ATLAS list, ATLAS web page, etc. That's what I wanted to know, that it was a non-trivial Sage spkg. I read that Sage builds ATLAS, hence the spkg for it. I'll start with that and study how to make it multithreaded (and LAPACK that it seems to rely on), and then how to make sure the Numpy spkg detects the multithreaded ATLAS. Thanks much. --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
