Hi Joe, I noticed that types weren't propagating as well as they could be in the SIMD matrix code. Turns out making 'scale-matrix4' inline fixed it. This reduced the runtime of the 3d matrix benchmark from 350ms to 80ms on my Mac Pro.
Then I made (byte-array) open-coded for larger allocation sizes. This improved performance further, now the benchmarks runs in 60ms. Slava ------------------------------------------------------------------------------ Come build with us! The BlackBerry® Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9-12, 2009. Register now! http://p.sf.net/sfu/devconf _______________________________________________ Factor-talk mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/factor-talk
