Hi! On the one hand, I need multiplication of square matrices over finite non-prime fields. From the things readily available in Sage (+ my MeatAxe wrapper), MeatAxe is best, the native Sage implementation sucks, and the trick described in the Boothby-Bradshaw paper (using LinBox for the prime field under the hood) is almost as good as MeatAxe even in naive implementation.
On the other hand, I need multiplication of square matrices with vectors, again over non-prime fields. Here, Meataxe is clearly best. Over GF(125), multiplication of a 2000x2000 matrix with a 2000x1 matrix takes 24.6 ms, native Sage needs 327 ms, and the Boothby- Bradshaw trick needs 109 ms. The problem is that LinBox appears to be rather slow in multiplying a 2000x2000 matrix with a 2000x1 matrix over GF(5) -- slower than native Sage, actually. But I suppose that there is a special LinBox function for multiplication with a vector. How is that function called? Is it wrapped? Things are getting confusing. I need a break. Best regards, Simon -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
