[Mplayer-users] fastmemcpy benchmark

[Arpi]

Hi,

> Am Sonntag, 22. April 2001 12:17 schrieben Sie:
> > Highly interesting....
> >
> > Asus P5A, K6-2/500, G200/16MB. Linux 2.4.3.
> >
> > # ./fastmem.sh
> > k6 : v2-v1=915325227 = 1826477us  (54.750fps)   82.1MB/s
> > mmx: v2-v1=790536059 = 1577468us  (63.393fps)   95.1MB/s
> > mga-k6 : v2-v1=3012802007 = 6011851us  (16.634fps)   25.0MB/s
> > mga-mmx: v2-v1=3798322327 = 7579307us  (13.194fps)   19.8MB/s
> 
> Hey, where did these come from? They are not in the CVS I get! 
They are.

cvs update
make

cd drivers
insmod mga_vid.o
cd ..

cd TOOLS
./c
./fastmem.sh

> Only the fastmem.sh script is, but not the programs it tries to call!
Run ./c to get them compiled!

> > I had an idea about memory bandwidth of this system sucking big time. But
> > *this* bad...??
> >
> > How come the mga-* test is so much worse than the others? Does this
> > indicate a bug? (Using the module in latest cvs.)
> 
> I don't think so. test mga-  version tests graphics memory, which is 
> slower than main memory (no caching?) 
video memory is always slower than system ram. it goes through your slow
pci/agp (33/66mhz) bus, while system ram is on your 100/200 mhz cpu bus.


A'rpi / Astral & ESP-team

--
mailto:arpi at thot.banki.hu
http://esp-team.scene.hu

_______________________________________________
Mplayer-users mailing list
Mplayer-users at lists.sourceforge.net
http://lists.sourceforge.net/lists/listinfo/mplayer-users