[ublas] Performance with SSE2

From: Preben Holm (preben_at_[hidden])
Date: 2007-03-12 11:16:01

Hi

Is there any significant performance gain in using SSE2 optimization for
the code! I mean by using: "-mfpmath=sse -msse2"!
I just did a simple example with random multiplications in a test-program,
but actually only little performance was gained here.

Has there been made any other flags by boost which could optimize
performance by using SSE2 vector extensions!

I haven't read much about the vector-extensions, but as far as I've read,
it should be "worth the effort" in some applications - I wonder if it
would be that together with ublas?

Best regards,
Preben Holm