From: Angus Dorbie (dorbie++at++sgi.com)
Date: 01/25/2000 12:10:51
Harry Overs wrote:
>
> But as previously described, it isnt the graphics hardware that we are
> really waiting on, its the SMP version of Performer that will prevent full
> performance and antialiasing in the PC graphics hardware ( Voodoo 5 ? ).
>
What makes you arrive at this conclusion?
When I mentioned the SMP Performer it was in relation to multihead
support only. Please don't take that as an indication of agreement with
the view that everything is gated on Performer SMP.
SMP will get you more performance on MP systems but it largely depends
how busy your app & cull are and whether or not you are fill limited.
Irrespective of this support for features like antialiasing will have to
rely on application level OpenGL calls (probably in the open pipe window
and channel draw callbacks), unless we get specific support in Performer
or the drivers expose the functionality using the multisample SGIX
extension, in which case it would just work even with the currently
released Performer. This would be unrelated to SMP Performer unless you
had to implement this using multiple cards with independent OpenGL
contexts and it's my understanding that (for example) a T-buffer does
not work that way. So that would have nothing to do with SMP support.
Driver features support, performance and ease of installation are
currently the main obstacles in this whole process.
Cheers,Angus.
--
"It takes a vision to beat a vision."
-Thomas Sowell
This archive was generated by hypermail 2b29 : Tue Jan 25 2000 - 12:11:12 PST