pfSwitch or pfCycleBuffer?

New Message Reply Date view Thread view Subject view Author view

Dan Oberlin (dan++at++mak.com)
Fri, 19 Dec 1997 14:59:02 -0500 (EST)

Thanks to all the people who responded to my question about the
implementation of a dynamically updating meter in Performer. Andy Shein
suggested using pfCycleBuffers to produce the dynamically updating
geometry. My co-worker suggested the possibility of using a pfSwitch as
a kind of double buffer. In this approach, you would change the geometry
in the node that is switched off and then reset the switch to point to
the node with the new geometry.

>From the pfGeode manpage:

...

However, the application may handle its own multibuffering of pfGeodes
through mutual exclusion with locks or through the use of parallel data
structures and pfSwitch nodes to achieve any kind of dynamic geometry.

...

I would like to know what the benefits are of the pfCycleBuffer approach vs.
the pfSwitch approach. Are there cases when one method is better to use
than the other? It would seem that the pfSwitch approach would be more
efficient in cases where the geometry doesn't change very often. Any
comments on these topics would be greatly appreciated.

Thanks,
Dan Oberlin

=======================================================================
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 - 17:56:27 PDT

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