Re: Multiple channels

New Message Reply Date view Thread view Subject view Author view

Ran Yakir (rany++at++bvr.co.il)
Wed, 14 Aug 1996 17:24:06 +0000


On Aug 13, 6:26am, Michael T. Jones wrote:

> :Is there a substantial overhead in having multiple channels,
> :occupying the same viewport, drawing multiple passes of the
> :database ?
>
> It causes more work, in that each channel gets traversed in the
> cull, whereas calling pfDraw() multiple times causes the same
> pfDispList built by the cull to be reused. Since most applications
> are not cull-limited (6-chan and 8-chan apps like MagicEdge are
> common exceptions) then you probably won't see any performance
> difference on an MP APP_CULL_DRAW system.
>
> By the way, if you want to do multipass rendering where different
> types of geometry are handled separately (diffuse, emissive, and
> so on) check out the pfGeoSet PassFilter and DrawBin functions.

The problem wit hcalling pfDraw() multiple times is that you can not
selectively draw other parts of the database in each pass (that is, in cases
where PassFilter does not supply the criteria you want).
pfDrawBin() could be the answer, although I can't make it work. Whenever I call
pfDrawBin() instead of pfDraw(), nothing gets drawn.

BTW, in single process mode - does multiple pfDraw()s cull the database
multiple
 time ?

Ran

-- 
 __                                  | Ran Yakir
 /_)  _  __   \  / _   / o __        | BVR Technologies Ltd.
/ )_ (_(_) )   \/ (_(_/<_(_)(        | 
              _/                     |   
-------------------------------------+--------------------------------
Phone :                              | E-mail : rany++at++bvr.co.il
  Work : 972-3-5715671               |
  Res. : 972-9-989974                |
Fax    : 972-3-5715668               |

======================================================================= 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 - 17:53:21 PDT

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