 | | Question: polygons filling a t-shape —Niels Karsdorp
|
| | | Hi, I read somewhere (forgot where) that quads and triangles should not pass by vertices, because this may cause gaps in some renderers. For L-shapes, the poligons should be created as on the right image: (URL) But how should T-shapes (and H-shapes, (...) (21 years ago, 3-Dec-03, to lugnet.cad, FTX)
|
| | | |
| | | | | Re: Question: polygons filling a t-shape —Travis Cobbs
|
| | | | (...) I suspect they will create artifacts in just about any renderer, but the severity will vary from one to another. (...) The above looks fine. You could do it with one less LDraw primitive while still preventing artifacts if you did it like so (...) (21 years ago, 4-Dec-03, to lugnet.cad, FTX)
|
| | | | |
| | | | | | Re: Question: polygons filling a t-shape —Larry Pieniazek
|
| | | | (...) Great diagrams, thanks! But I thought that (rgarding quad 2 in the last diagram) collinear points were considered an error by the checkers, so that when you submit your part to parts tracker it would be flagged??? (21 years ago, 4-Dec-03, to lugnet.cad, FTX)
|
| | | | |
| | | | | | Re: Question: polygons filling a t-shape —Travis Cobbs
|
| | | | | (...) I'm not sure if it does or not. After posting the previous post, I realized that I was mistaken about my alternate version being "ok". I just realized that it could result in the same rendering artifacts that using a triangle for polygon 2 (...) (21 years ago, 4-Dec-03, to lugnet.cad, FTX)
|
| | | | | |
| | | | | | | Re: Question: polygons filling a t-shape —Steve Bliss
|
| | | | | | (...) Thanks for pointing this out -- I've been bugged for awhile about the "colinear quads are bad just because L3P says they're bad" situation, here's a very good reason to not use them. Steve (21 years ago, 5-Dec-03, to lugnet.cad)
|
| | | | | | |
| | | | | | Re: Question: polygons filling a t-shape —Lars C. Hassing
|
| | | | (...) Yes, L3P detects and warns about the collinear points of the quad. In the case of splitting the T why don't you just use triangles? The renderers do that anyway, but then you can take control of the splitting. I believe quads are mainly a (...) (21 years ago, 4-Dec-03, to lugnet.cad)
|
| | | | |
| | | | | | Re: Question: polygons filling a t-shape —Steve Bliss
|
| | | | (...) Actually, with LDraw code, reducing triangle pairs to quads reduces that section of code by 35% (more-or-less). If you've got a lot of surfaces, that's not an insignificant savings of bytes. Steve (21 years ago, 5-Dec-03, to lugnet.cad)
|
| | | | |
