hal@datacube.UUCP (11/11/88)
Does anyone out there care to recommend a good reference on
raytracing?
There've been several programs posted (I'm still hoping to
catch MTV if he ever posts again), but they're a little hard to walk
though. I have a multiprocessor environment available, and I'd like
to try to partition one of these raytracers to run in it.
----------------------------- Hal Moroff -------------------------------
Datacube Inc. / Applications Group
4 Dearborn Road
Peabody, MA 01960
USA
VOICE: (508) 535-6644 FAX: (508) 535-5643 TWX: (710) 347-0125
UUCP: hal@datacube.COM [ihnp4 | mirror]!datacube!hal
{seismo,cbosgd,cuae2,mit-eddie}!mirror!datacube!halmarkv@uoregon.uoregon.edu (Mark VandeWettering) (11/13/88)
In article <9700001@datacube> hal@datacube.UUCP writes: > Does anyone out there care to recommend a good reference on >raytracing? Right now, there seems to be a lack of really good references about raytracing, other than pouring over lots of individual articles. "Procedural Elements of Computer Graphics" by Rogers is a very good book, but I am unsure whether it will tell you "enough" to get going on raytracing. You should probably get a copy of Turner Whitted's original article "An Improved Illumination Model for Shaded Display" in the June 80 CACM. Between that and Rogers, you should be able to hack together a rudimentary raytracer, and learn alot in the process. The best reference by far is a set of the Siggraph course notes called "Introduction to RayTracing". I have heard rumor that this is soon to be a real live book by next year. Pick one up if you can, excellent reference. The key that I have found to making all this work is to not try to implement what you don't fully understand. It makes code impossible to debug. Even if someone else gives you an algorithm, you should try to understand it rather than just copy it. > There've been several programs posted (I'm still hoping to >catch MTV if he ever posts again), but they're a little hard to walk >though. I have a multiprocessor environment available, and I'd like >to try to partition one of these raytracers to run in it. My program is indeed a bit hard to walk through. It is getting too big for novices to understand easily, especially if you don't have access to the now extensive set of references that I have gathered. I too have thought about a multiprocessor implementation of a raytracer. It shouldn't be too difficult, but then again, I haven't done it yet either :-) Concerning a new release of my raytracer: see a followup posting to appear here later (within a day). >----------------------------- Hal Moroff ------------------------------- > Datacube Inc. / Applications Group > 4 Dearborn Road > Peabody, MA 01960 > USA >VOICE: (508) 535-6644 FAX: (508) 535-5643 TWX: (710) 347-0125 >UUCP: hal@datacube.COM [ihnp4 | mirror]!datacube!hal > {seismo,cbosgd,cuae2,mit-eddie}!mirror!datacube!hal
johnm@trsvax.UUCP (11/14/88)
> Does anyone out there care to recommend a good reference on >raytracing? Try: Procedural Elements for Computer Graphics by David F. Rogers McGraw Hill ISBN 0-07-053534-5 This is the only book that I have ever bought sight unseen (based only upon the recommendations of others in this newsgroup), I haven't regretted it for a second. I highly recommend it. This book gives loads of information about ray tracers, scan line methods, painter's algorithms, etc. and gives you the flow charts and descriptions of data structures that you need to implement them. John Munsch