Subject:
|
Re: Detemining Bad Vertex Sequence
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Wed, 11 Sep 2002 17:52:18 GMT
|
Viewed:
|
406 times
|
| |
|
|
In lugnet.cad.dev, Erik Olson writes:
> > The problem is that it is very slow ( lots of cross products and scalar
> > products ! ).
>
> Does it need to be any faster?
Are you talking about the algorithm or the program?
If you are talking about the algorithm, you are right! Determining the bad
vertex sequence isn't the problem of my program because I have only to run it
in the beginning, and only for each quad. I asked this question because I was
just wondering if there are better algorithms than the one that I found.
However, if you are talking about the program, I think that it needs to be
fast enough that someone creating a new part doesn't get bored waiting
for it to end. For example, it is taking about 28 minutes to certify the "2560
Boat Base 8 x 16" http://guide.lugnet.com/partsref/search.cgi?q=2560
Another reason to improve the performace, it's because debug a program that
takes to long to end is very tiring.
Anyway, it is taking this horrible time to end because I'm using a
statistical sampling of rays to detect if the polygon is pointing inside or
outside. So, I have to cast thousands and thousands of rays to make a good
sampling. I think that the performance will get better when I implement some of
the bounding box algorithms.
play well,
zanette.
|
|
Message is in Reply To:
12 Messages in This Thread:
   
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
