Subject:
|
Re: Formalizing a method for handling synthesizable parts
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Sat, 12 Oct 2002 00:28:19 GMT
|
Viewed:
|
574 times
|
| |
|
|
Philo,
I like your idea of multiple resolutions. It could save a lot of space.
I think I that I don't have a minimal solution to the number of segments
used. Some fine tuning might make a big difference in file space consumed.
The stuff that is out there now, may not be optimal.
One problem I have is with electric cables. I chose to have across
section that is to coincident circles so the cable looked more realistic.
One problem with this is that the cable twist is not part of the algorithm.
If you need a twist, you are out of luck.
I've got the mechanism in place where the band plane can be any
orientation (not just in the xy plane).
I'm able to synthesize the straight line tangent segments between pulleys,
and now I'm wrestling with the arc around the pully. I will be done soon.
Next is chain/conveyor, then tread.
This still leaves springs and strings. I have not even thought about
these much yet.
Kevin
In lugnet.cad.dev, Philippe Hurbain writes:
> In lugnet.cad.dev, Kevin Clague writes:
> > Oop.s I reworked the organization of the code and ended up messing this up.
> > 1.0.0.2 is uploaded now.
> Works fine now ;o)
> My tests were far from exhaustive though !
>
>
> > I'm not sure what we can do about the total number of triangles or quads. I
> > suppose we could introduce more sets of hose cross sections that are longer.
> > Double the length, halve the number of quads.
>
> I guess there is no perfect solution: modelling curved parts with some
> smoothness requires lots of polygons. Fortunately machine power increases!
>
> One possibility though would be to have a dual set of primitives, one hi-res
> and the other lo-res, swichable according to rendering quality (similar to
> lo-quality studs)
>
> Philo
|
|
Message is in Reply To:
 | | Re: Formalizing a method for handling synthesizable parts
|
| (...) Works fine now ;o) My tests were far from exhaustive though ! (...) I guess there is no perfect solution: modelling curved parts with some smoothness requires lots of polygons. Fortunately machine power increases! One possibility though would (...) (22 years ago, 6-Oct-02, to lugnet.cad.dev)
|
37 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
