[Mplayer-users] fastmemcpy benchmark

[Felix Buenemann]

On Sonntag, 22. April 2001 12:17 you wrote:
> 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
>
> I had an idea about memory bandwidth of this system sucking big time. But
> *this* bad...??
>
It's really a little slow, here are results to compare (first and second 
run), system is PIII 750 Coppermine at 930MHz with 512MB SD-RAM (CL3), Board 
is Abit BX6 2.0 with Intel 440BX chipset:

atmosfear:~/develop/mplayer/main/TOOLS # ./fastmem.sh
k7 : v2-v1=199387832 = 214239us  (466.768fps)  700.2MB/s
mmx: v2-v1=618539835 = 664605us  (150.465fps)  225.7MB/s
sse: v2-v1=203580149 = 218742us  (457.160fps)  685.7MB/s

atmosfear:~/develop/mplayer/main/TOOLS # ./fastmem.sh
k7 : v2-v1=190158733 = 204321us  (489.426fps)  734.1MB/s
mmx: v2-v1=616706678 = 662635us  (150.913fps)  226.4MB/s
sse: v2-v1=196876137 = 211538us  (472.728fps)  709.1MB/s

> How come the mga-* test is so much worse than the others? Does this
> indicate a bug? (Using the module in latest cvs.)
>

I guess it might have to do with no or improper agp support of you 
system/kernel.

> Thank you for excellent hacking.
>
> Dag B
>

-- 
Best Regards,
   Felix

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