Re: gl display lists

New Message Reply Date view Thread view Subject view Author view

Angus Dorbie (dorbie++at++multipass.engr.sgi.com)
Mon, 27 Oct 1997 10:54:46 -0800

On Oct 27, 11:19am, Jean-Claude Bachmann wrote:
> Subject: Re: gl display lists
>
> Angus Dorbie wrote:
>
> > You may not see a performance win. Display lists are
> > probably of most benefit on ONYX(1) iR or on single CPU
> > systems where you run an APPCULL_o_DRAW processing model.
> >
> >
>
> Could a knowing soul give an explanation for that. I thought display lists
> reside on the graphics hardware, and wasexpecting some relief for the draw
> process since it would only be necessary to activate a gl display list
instead
> of
> transferring all the data to the hardware.
>

The HW can transform geometry at a limited rate, but there is a limit
to the speed you can feed the graphics pipe data. By display listing
on iR you can move data to reside on the graphics board and avoid
the bandwidth bottleneck. If the bandwidth to graphics is significantly
less than the geometry performance of the graphics system then display
listing can be a big win, otherwise it can be less of a win. Hence
ONYX it's a win but ONYX2 it's not as big a deal.

Cheers,Angus.
=======================================================================
List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
            Submissions: info-performer++at++sgi.com
        Admin. requests: info-performer-request++at++sgi.com

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:56:08 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.