Re: Double buffers used in ASD sample

New Message Reply Date view Thread view Subject view Author view

Gan Wang (gan++at++cavalier.cambridge.com)
Fri, 23 Aug 1996 14:25:24 -0400

On Aug 23, 10:54am, Mario Veraart wrote:
> I think it depends on what kind of a machine you try to run it.
> If you run it on a 1 CPU machine, like an Impact, you probably don't get
> a conflict between APP and DRAW because they will operate sequential,
> not intermixed. But I think if you run in APP_CULL_DRAW mode on a 4 CPU
machine
> you must have a problem or the changing of the mesh by the App is done at
> a different part in the frame as when DRAW needs the geosets. If it's done
> between pfSync and pfFrame(), not very wise, than there wouldn't be a problem
> because the DRAW is suspended until the pfFrame().

A little more background: I run the "asdfly" sample on an Onyx/iR with 4
processors and the program is configured in PFMP_APP_CULL_DRAW.

On Aug 23, 9:37am, Scott McMillan wrote:
> Subject: Re: Double buffers used in ASD sample
> I would assume this works because DRAW actually works on display
> lists that are generated by the CULL and not the actual gsets (this
> is just a guess though).

Correct me if I am wrong, I don't believe that pfDispList copies gsets. I
think it only has libpr pfCommands and pointers to the only copy of gsets in
the world.

I am not convinced and still puzzled...

Gan

-- 


Gan Wang


Cambridge Research Associates            Office:   703-790-0505 ext.7210
1430 Spring Hill Road, Suite 200         Fax:      703-790-0370
McLean, VA 22102                         Internet: gan++at++cambridge.com              
=======================================================================
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 - 17:53:24 PDT

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