Angus MacDonald Dorbie (dorbie++at++multipass.asd.sgi.com)
Wed, 21 May 1997 20:16:17 -0700
Must have gotten lost somewhere.
> On May 19, 2:24pm, Angus W.S. Henderson wrote:
> > Subject: Dynamic Resolution & Framebuffer Copies
> >
> > If I am using dynamic resolution, and also copying the framebuffer to
texture
> > memory for distortion correction. Will I copy the constant size image or
the
> > shrunk down version ?
DVR will not affect anything you do in the GL, you'll readback the
framebuffer sized image, DVR is video only.
If you consider the DVR & distortion correction situation it should be
obvious that DVR is of limited value in the distortion correction
implementation. Look:
CONVENTIONAL DVR
________________
DVR
(Draw image to variable resolution)--->(Video output at video resolution)
DISTORTION CORRECTION DVR
_________________________
DVR
(Draw Primary Channels)--->(Texture Readback)--->(Draw to Mesh)--->(Video)
You can trivially change the resolution of both the primary channels
and the texture readback(8 texel increments) to dramatically change
the load on the system if fill or readback limited (a texture matrix
would automatically scale mesh texture coordinates). Using DVR would
only allow you to make minor adjustments to the Depth Complexity==1
load on the rendering of the final mesh. Ofcourse it might make sense
to make all adjustments simultaneously.
I didn't know an N64 could do DVR :-)
> >
> > ANGus
> >-- End of excerpt from Angus W.S. Henderson
>
> If you won't help we'll steal your new angus as soon as they've paid for his
> green card....
Paradigm ain't big enough for the both of us.
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
This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:55:16 PDT