CAD Development : 3017


CAD / Development / 3017	3016 \| 3018

Subject:	Re: Hidden surface removal, and vertex order in part/primitive DAT files
Newsgroups:	lugnet.cad.dev
Date:	Mon, 4 Oct 1999 16:27:27 GMT
Viewed:	659 times

On Sun, 3 Oct 1999 21:17:06 GMT, Jean-Pierre PARIS <Paris.Jean-Pierre@wanadoo.fr> wrote: > Gary Williams wrote: > > > > [snip] > > Maybe it would be simpler to just manually set the vertex order of the first > > polygon, let the program fix that polygon's neighbors and its neighbors' > > neighbors, and if any polygons are left then ask the user to set the vertex > > order of one of those remaining, and repeat the process. > > I really don't like programs that need user interactions for such job! Neither do I, but I also know how hard it is to visually check a part for errors. > Thus I tried to cast enough rays so that all the faces are being > oriented correctly. How does that work? Is it based on what the majority of the rays deterimine? I probably need to re-read some of the messages in this thread, to get a better idea of this process. > Now all the std bricks are OK, without any > assumption about the fact that they are mainly composed of 2 box5 and a > lot of studs. The general algorithm is working well here. So do you 'drill into' primitives? > If the pressure of all Ldrawers grows, I'll try to take time, and check > my code against more parts. If someone has good ideas of difficult > bricks, please let me know! Do you want suggestions of difficult bricks to test against? 2560 (Boat Base 8 x 16) is pretty ugly. It's a big file, there are lots of nooks and crannies, and I had to cut apart various primitives to make it all go. If you test on this brick, I can also send you the final set of files, un-inlined. > > There are only 1600+ part files, after all...we could use a group of > > volunteers to divide the task, and get every part fixed in a week's time. :) > > Get ready with a Java dev. kit, and I can send my code to you!!! Package it so the fixer doesn't need a JDK, and send it to a lot of people. Steve

Message has 3 Replies:

		Re: Hidden surface removal, and vertex order in part/primitive DAT files
(...) I choose rays so that they may orient one face. With this face I use a sort of flood-fill algorithm (please try to find on of my previuos message) to orient al neighbors. Once a face is oriented, I don't look at it anymore. But, your idea (...) (25 years ago, 5-Oct-99, to lugnet.cad.dev)

		Re: Hidden surface removal, and vertex order in part/primitive DAT files
(...) Ok, I can announce that my code actually orient ~10000 faces of 2560, BUT 4. In a previous version, 40 faces remain unoriented. After I checked all distances between the ~12000 of the part, I have discovered that some distances were very (...) (25 years ago, 15-Oct-99, to lugnet.cad.dev)

		Re: Hidden surface removal, and vertex order in part/primitive DAT files
Hello! Eventually I finished orienting (automatically) all faces of bricks 10 and 2560! For those who are interested by the results, please have a look at (URL) the same place, one'll find the corrected 2560. Yes, this part as small bugs: 16 faces (...) (25 years ago, 21-Nov-99, to lugnet.cad.dev)

Message is in Reply To:

		Re: Hidden surface removal, and vertex order in part/primitive DAT files
(...) I really don't like programs that need user interactions for such job! Thus I tried to cast enough rays so that all the faces are being oriented correctly. Now all the std bricks are OK, without any assumption about the fact that they are (...) (25 years ago, 3-Oct-99, to lugnet.cad.dev)

56 Messages in This Thread:

Entire Thread on One Page:: Nested: All | Brief | Compact | Dots
Linear: All | Brief | Compact
This Message and its Replies on One Page:: Nested: All | Brief | Compact | Dots
Linear: All | Brief | Compact

Custom Search