home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!news.cs.bham.ac.uk!christopher-robin!rjg
- From: rjg@christopher-robin.cs.bham.ac.uk (Robin J. Green)
- Newsgroups: comp.text.tex
- Subject: At last! Heaven found...
- Message-ID: <Btwn69.707@cs.bham.ac.uk>
- Date: 1 Sep 92 15:03:44 GMT
- References: <1992Aug31.215407.22911@PA.dec.com> <1992Sep1.133136.4718@athena.mit.edu>
- Sender: Robin Green
- Followup-To: rjg@uk.ac.bham.cs
- Organization: School of Computer Science, University of Birmingham, UK
- Lines: 28
- Nntp-Posting-Host: christopher
-
- OK, finally a question to tax you, cos it's taxing me.
-
- I want to produce a fully stochastic raytracer, integrating across the picture
- plane by jittering pixels, across light source areas and across time. These are
- the easy bits. I want to integrate across the reflection function, that is I
- want to jitter a ray using IMPORTANCE SAMPLING so that it produces a series
- of n jittered rays each with a probability of direction proportional to their
- associated reflectance value. How do I do this?
-
- All I want is to be able to produce glossy surfaces and an aparrent effect of
- 'diffuse transparancy' which I take to mean Tupperware. Please note I am not
- asking for a package to solve all my worldly problems, provide 3rd world
- relief, cure cancer, produce colour overheads and make an entertaining lunchtime
- friend.
-
- As I see it, the reflectance function can be a very complex shape roughly
- maximum at the reflected rays direction but certainly not symmetrical or even
- a volume of rotation. Erk.
-
- Please email me some insights into this problem. It's just another example of
- SIGGRAPH papers saying '...which is left as an exercise for the student...'
- God I HATE that!!
-
- - Thank you for reading this. If you have enjoyed this plea for help, please
- share it with your friends, your family or even your pets. No.1 in an
- occasional series...
-
- - RJGreen.
-