Mercurial > mplayer.hg
view TOOLS/perlbench.pl @ 10578:b9d289fd8a57
10000l, the old code was slow as hell, copying stuff extra times and
actually broken -- blanking the whole screen at each 'page flip' with
-dr enabled. benchmarks:
before:
56% cpu for decode
56% cpu for vo with no -dr
25% cpu for vo with -dr
after:
56% cpu for decode
25% cpu for vo without -dr
0% cpu for vo with -dr
if vo_fbdev is going to do pageflip, it needs to do it for REAL, using
vertical scroll registers (like g2), not copying a temp buffer (which
will shear anyway and is super-slow).
author | rfelker |
---|---|
date | Tue, 12 Aug 2003 08:24:24 +0000 |
parents | 5cbbf47cae1e |
children |
line wrap: on
line source
#!/usr/bin/perl -w use strict; use constant CMD => "./fastmem2-k7"; sub dobench { my $i; my ($runs, $sleep, $command) = @_; for($i = 0; $i < $runs; $i++) { sleep $sleep; system $command; } } print "Single run of sse bench with 1sec sleep:\n"; &dobench(1,1,CMD); print "Sleeping 10seconds before starting next bench!\n"; sleep 10; print "10 runs of sse bench with 0sec sleep:\n"; &dobench(10,0,CMD); print "Sleeping 10seconds before starting next bench!\n"; sleep 10; print "10 runs of sse bench with 1sec sleep:\n"; &dobench(10,1,CMD); print "Sleeping 10seconds before starting next bench!\n"; sleep 10; print "10 runs of sse bench with 2sec sleep:\n"; &dobench(10,2,CMD); print "Sleeping 10seconds before starting next bench!\n"; sleep 10; print "10 runs of sse bench with 3sec sleep:\n"; &dobench(10,3,CMD); print "Bench finished!\n";