Re: Size of objects
rweyrauch++at++camus.paradigmsim.com
Tue, 3 Dec 1996 16:12:03 -0600
On Dec 2, 1:53pm, Jude Anthony wrote:
> Subject: (Fwd) Size of objects
>
> [...]
>
> Part of our TTPRR includes measuring the size of objects on the
> screen. When we run perfly, the sizes come out correct. When we run
> any Vega application, the objects are too small. The error is not
> linear, but in general the percentage difference is worse the closer
> the object is. (Sizes range from 90.16% to 64.89% of their intended
> size.)
>
> Neither of the windows has a border. The Vega application is a
> minimal thing. I haven't added anything to it but my debugging
> printouts. The perfly code is "mostly" unmodified: I added changes
> for alpha blending in high-quality mode, flimmering fixes, and my
> debugging printouts.
>
> I've printed out the FOV and clip planes in both applications, and
> they match. I've printed the Vega viewing matrix, projection
> matrix, and WorldToScene matrix of the Vega app; all of them are
> trivial orthonormal matrices. I've checked the Vega skew, which is
> 0.0 all through. I'm running out of ideas.
>
> I've set up the Vega .adf file to put a wall at (0, 200, 0). The
> observer is default. When I call perfly, I set the observer at
> (0, -200, 0). The wall fills the screen horizontally in perfly, but
> has plenty of space around it in Vega. (It's easy to see the
> difference without getting out the ruler in this case.)
>
The default observer in Vega is attached to a the default motion model with is
placed at (0,-10,0) not at the origin. Sounds to me like you are viewing the
object from a distance of 200 meters in perfly and 210 meters in Vega. Try
changing the location of the motion model to the origin and rerun your test.
> Anybody know what's going on? Even a suggestion on where to look
> would be helpful.
>
> Thanks in advance,
> Jude Anthony
> jude++at++p3.enzian.com
>
>-- End of excerpt from Jude Anthony
--
Rick Weyrauch voice: (972) 960-2301
Paradigm Simulation Inc. fax: (972) 960-2303
14900 Landmark Blvd., Suite 400 rweyrauch++at++paradigmsim.com
Dallas TX 75240 www.paradigmsim.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
This archive was generated by hypermail 2.0b2
on Mon Aug 10 1998 - 17:54:05 PDT