matmul intrinsic procedure

[Janne Blomqvist]

On Thu, Jun 30, 2005 at 02:11:50PM -0600, Jan Stanis³aw Owoc wrote:
> Unfortunately, many of my programs deal with large matrices (I do 
> finite-element analysis a lot), including multiplying them.

In addition to what others said about matmul, for large matrices you
might want to take a look at *gemm from BLAS.  For large matrices an
optimized BLAS is pretty hard to beat.

Or since you're doing finite elements, you might be interested in a
sparse matrix package?

-- 
Janne Blomqvist