[MPlayer-users] Re: Re: CPU optimization
arpi at thot.banki.hu
Tue Oct 15 21:54:12 CEST 2002
> It would be very interesting to have some idea of the speed increase
> due to the optimization for the cpu and for the cpu features, otherwise
> what you are saying does not give much info to anybody except to the
> developers. For what I know [i.e. zero, as you pointed out] CPU features
> may count for a 30% speed increase and CPU optimization for a 5%, in
> which case I'd have no interest in optimizing for the cpu.
you maybe don't have, if your cpu is faste nough...
i know ppl and here are some too, who gives everything for 2% more cpu
power. except money :)
> this is definitely true, but IMHO the cause of this is also some deficiency
> in mplayer's DOCS: the only trace of the --enable-runtime-cpudetection
> apparently lies in the ./configure help and its comment [i.e. "Enable
> runtime CPU detection"] is not exactly enlightening, at least for me.
it's not for end-users, it's for package makers and they know it well.
for ppl compiling mplayer form source it shouldn't be the choice, it's VERY
RECOMMENDED to optimize mplayer for their own cpu/system. This is mplayer
and even the whole opensource world about.
A'rpi / Astral & ESP-team
Developer of MPlayer, the Movie Player for Linux - http://www.MPlayerHQ.hu
More information about the MPlayer-users