On Saturday, 2. März 2002 23:03:06, you wrote:
> On Sun, Mar 03, 2002 at 11:44:44PM +0100, Laurent Pinchart wrote:
> > I'll have a look at that.
> >
> > Which CVS branch should I start with ?
>
> On the glide side you want to use glide3x, but I don't think they ever
> created any branches there. There are glide3x for V3 and glide3x for
> V{4,5} branches which are different. Reconciling those will be a bit of
> a hassle, but the architectures are really close so the differences
> aren't too large.
Could we start with fixing the last outstanding Glide3 bug?
It is in the 3DNow! asm code. I had it going before Joseph Kain (former 3tdfx
employee) fixed the "pixel smearing" bug on my older V3 3000 AGP.
His "bug fix" brought HUGE performance degradation in
/*
* tenmillion.c
*
* OpenGL performance test for triangles in a display list.
* [EMAIL PROTECTED], March 1996
[-]
Mesa/demos> ./tenmillion smooth
running on SunWave1 (Mesa DRI 20020221 Voodoo4 x86/MMX/3DNow!) Sun Mar 3
18:41:29 2002
visual: 0x23
50 pixel triangles, 90 triangles/strip, 1 strips
vertex data: v3f
smooth shading, zbuffer off, light off, texture off, immediate off
test time: 1.503 secs.
geometry rate: 929370 triangles/sec
fill rate: 46.468 Mpixels/sec
draw_strips_count: 1
I got ~160 Mpixels/sec with my "old" Athlon 500 MHz and V3 3000 !!!
Nearly equal results measured with Accelerated-X OpenGL 2.x beta test package
on my 1 GHz Athlon II...
More benchmarks/numbers available if somebody want and I have the "right" code
handy but I am from the 68k asm side...;-)
gemessen
Regards,
Dieter
_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel
