billd@celerity.uucp (Bill Davidson) (08/22/89)
I sent mail to Mark V about this but I never got an answer (I'm not absolutely sure he ever got my mail since it bounced with the first two addresses that I tried, the third didn't bounce but it may have fallen into email hell). Anyway, I'm trying to run ray on the nff files generated by Eric Haines programs as distributed from uoregon. The aspect ratios seem screwed up in general but especially on cylinders and cones. I'm not sure if it's a problem in the ray-tracer or in the nff files. Anyway, anything circular which is oriented horizontally will be stretched out. I'm computing these things in a square field with square pixels. Anyone out there know how to fix this (is there a way to set aspect ratio somewhere?). --Bill
markv@phoenix.Princeton.EDU (Mark T Vandewettering) (08/22/89)
In article <561@celit.UUCP> billd@celerity.UUCP (Bill Davidson) writes: >I sent mail to Mark V about this but I never got an answer (I'm >not absolutely sure he ever got my mail since it bounced with the >first two addresses that I tried, the third didn't bounce but it >may have fallen into email hell). Anyway, I'm trying to run ray on >the nff files generated by Eric Haines programs as distributed >from uoregon. The aspect ratios seem screwed up in general but >especially on cylinders and cones. I'm not sure if it's a problem >in the ray-tracer or in the nff files. Anyway, anything circular >which is oriented horizontally will be stretched out. I'm computing >these things in a square field with square pixels. Anyone out >there know how to fix this (is there a way to set aspect ratio >somewhere?). I never received any mail from you, but since I read this group normally, I'll take time here to respond. It has just recently come to my attention that my cylinder code is buggy, causing things to be quite oddly distorted. I must have introduced this bug prior to my posting, because I will swear that I tried this stuff out (yeah right mark, uh huh, tell it to the judge) In any case, I am working on a bug fix, but am pretty busy. When I have fixed it, I will post the bug fixes. I will also make a new version available from cs.uoregon.edu, which will also include the patches submitted to me by Eric Haines (thanks Eric, why didn't you catch these :-) There must be renewed interest in my raytracer after Siggraph :-) Mark
billd@fps.com (Bill Davidson) (08/24/89)
In article <10046@phoenix.Princeton.EDU> markv@phoenix.Princeton.EDU (Mark T Vandewettering) writes: >In article <561@celit.UUCP> I wrote: [problems sending mail to Mark V] >I never received any mail from you. I kind of suspected that. It was several months ago though so I wouldn't expect you to remember it if you had. [good news about fixes he's going to post] >There must be renewed interest in my raytracer after Siggraph :-) Actually, I haven't touched it for months but I've been trying to write a generalized color quantizer that reads and writes several formats with dithering (24 bit's to n bits where 24 >= n >= 1. (Jef P. when do we get COLOR PBM? :-). I was inspired into this project by the raytracer. I've been using Eric's nff files raytraced with Mark's raytracer as test images. I printed several on our imagen laser printer and everyone who looks at them comments about the distortion so I decided to try to get an answer on it again. --Bill