home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.graphics
- Path: sparky!uunet!usc!sdd.hp.com!cs.utexas.edu!torn!cunews!nrcnet0!bnrgate!bmerh85!nbrwh34!kaustine
- From: kaustine@nbrwh34.bnr.ca (Mathew Kaustinen)
- Subject: Re: Hidden Surface Removal
- Message-ID: <1992Jul31.140057.13785@bmerh85.bnr.ca>
- Sender: news@bmerh85.bnr.ca (Usenet News)
- Organization: Northern Telecom, Bramalea, Canada
- References: <1992Jul30.142742.5874@tc.cornell.edu>
- Date: Fri, 31 Jul 92 14:00:57 GMT
- Lines: 36
-
- In article <1992Jul30.142742.5874@tc.cornell.edu> elan@tasha.cheme.cornell.edu (Elan Feingold) writes:
- >
- >to front, but this doesn't seem to work allways for me. Example:
- >
- >
- >
- > --\
- > / / \
- > / /\--
- > / /
- > --
- >
- >I don't know if this picture is enlightening. :) Basically I have a long
- >polygon and then a smaller one going behind it. When the viewpoint is slightly
- >above and to the right of the scene, the longer polygon is plotted first,
- >messing up the scene. Can anyone help as far as an algorithm is concerned?
-
- Just a thought (I lay awake half the night thinking about it!). Are
- the co-ordinates of your patches ordered in the same direction? ie.
- if you look "above" each patch do the points always rotate in the
- same direction, such as:
-
- 1a---2a 1b---2b
- | | | | Patches A and B have same "rotation"
- | A | | B |
- | | | |
- 4a---3a 4b---3b
- If not this may be your problem.
- If so, since the patches are non intersecting, I think that a algorithm
- that checks the z-distance for the same point # on each patch should
- work. I used this method on a Bezier surface program I wrote and it
- seemed to work for this application.
- --
- Mathew A. Kaustinen | ESN: (333) 2066
- Field Service Engineering | PHONE: 416-452-2066
- Northern Telecom, Bramalee | PAGER: 416-378-3246
-