Re: pfDoubleDCS picking/intersects

New Message Reply Date view Thread view Subject view Author view

From: Yair Kurzion (yair++at++polygon.engr.sgi.com)
Date: 12/20/2000 18:34:11


Hello Patrick !

> I just started working with Performer 2.4, and have been awaiting the
> pfDoubleSCS/DCS/FCS nodes. I have installed them in my program and they
> work great solving my jumpy movement problems. Unfortunately intersecting
> doesn't seem to be implemented for them. Can this be fixed? If it cant
> be i will have to live with the jitter or hack custom intersection code,
> its not clear to me at this point which option is worse.

perf2.4 contains erroneous intersection code for pfDouble?CS (Thanks for finding
this). It will be fixed in two phases:

1. In an upcoming maintenance release (2.4.x) we'll add intersection code that
   uses only a floating point version of a pfDouble?CS matrix. As a result, for
   objects far from the origin, the intersection hits may be jumpy. This part
   of the fix makes pfDouble?CS intersection identical to pf?CS intersection.

2. In the 2.5 release, we'll add full support for pfDouble?CS intersection:
   When the intersection traversal encounters a pfDouble?CS node it upgrades
   its segment information to double precision. Result hits are available as
   double precision points.

The reason for the two phase fix: All 2.4.x releases have to be binary
compatible. The full fix requires some new API/classes/members and breaks
binary compatibility.

If you are interested, we can send you an alpha version of the full fix as it
becomes available.

-yair

-- 
\_________  \_____  \__    \__  \_____         Yair Kurzion
\_________  \_____   \__   \__  \_____         yair++at++sgi.com
       \__     \__   \____\__      \__   http://reality.sgi.com/yair
       \__          \__  \__                Work: (650) 933-6502
       \__          \__   \__               Home: (408) 226-9771
       \__          \__    \__             


New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Wed Dec 20 2000 - 18:34:16 PST

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