[geeks] Itanium 32 bit performace.... hahahaha

[jdboyd at jdboyd.zill.net]

On Tue, Dec 09, 2003 at 04:15:57PM -0800, Francisco Javier Mesa-Martinez wrote:
> On Tue, 9 Dec 2003 jdboyd at jdboyd.zill.net wrote:
> 
> >
> > * I determined that it was transform bound by noting that it ran no
> >   faster in a 320x240 window than it did in a 800x600 window.
> 
> Then that is not transform based, but rather raster based.

If it runs the same reguardless of resolution, I thought that meant that
the hold up was something other than raster speed.  After all, there are
less pixels at 320x240 than there are in an 800x600 window.  Maybe I'm
having a dense moment.
 
> > ** I used code from an OpenGL -> postscript demo to cache where every
> >    vertex was really located after the hierarchy of objects was
> >    transformed.  SGI seems happy with things like this, but some source
> >    suggested that it may be sub-optimal for AGP based machines, which I
> >    somewhat disregarded on the theory that the SGI way was the one true
> >    way.
> 
> True dat.

Besides, there is a certain simplicity to the SGI way that is quite
nice.  If only people in the PC/Mac world would start using PCI-X for
graphics instead of AGP (my recall is that all AGPs, including 10x do
read backs at slightly less than PCI speeds, and in my book that just
sucks).