Re: Fixed point cairo.. or no cairo?

[Aivars Kalvans]

It seams that huge number of cycles is due to accessing RAM when matrix is used for the first time. It gets better after first use:

Athlon XP
branching:              22 cycles
branching:              21 cycles
branching:              23 cycles
inline multiply:        601 cycles
inline multiply:        17 cycles
inline multiply:        17 cycles
function call:          92 cycles
function call:          55 cycles
function call:          56 cycles

Pentium 4
branching:              88 cycles
branching:              92 cycles
branching:              92 cycles
inline multiply:        4360 cycles
inline multiply:        152 cycles
inline multiply:        156 cycles
function call:          164 cycles
function call:          160 cycles
function call:          160 cycles

 

--
Aivars

__________
Advertisement:

Es iesu uz mobilo festivalu „re:loud”!
14 grupas, 2 skatuves, DJ telts, atrakcijas
1.septembri Mezaparka estrade www.reloud.lv

_______________________________________________
Performance-list mailing list
Performance-list@gnome.org
http://mail.gnome.org/mailman/listinfo/performance-list