From: Andrew Dalke (dalke_at_acm.org)
Date: Sun Jul 16 2000 - 20:54:55 CDT

John Stone <johns_at_ks.uiuc.edu>:
>At some point the speed of the bus connecting
>the CPU and the framebuffer is going to be the limiting factor. Since
>glDrawPixels() doesn't really have a way of caching images on the
>framebuffer's local memory, it will eventually be slower than using
>glCallList() to render the gluSphere, simply because all of the geometry
>of the gluSphere will already be cached on the framebuffer. This isn't
>the case with the graphics boards we have this year, but its definitely
>coming.

I've was talking with Roger Sayle, the author of RasMol, about this. He's
a firm believer in software renderers, because of his experience. RasMol
for the past 10 years has been faster than GL or OpenGL, for what it does.
I've been trying to convince him that hardware will catch up to his
highly tuned code, which is very fast but supports neither perspective nor
multiple light sources. This year it seems to have done so. He has
a benchmark on the latest Sun graphics hardware which says using an OpenGL
call list for a sphere gives better performance than his code, and without
the aforementioned limitations. So John's "definitely coming" is here,
at least for a few cases, and Roger is learning OpenGL :)

                    Andrew
                    dalke_at_acm.org