Subject:
|
Re: Seams like a problem
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Thu, 12 Oct 2000 02:27:44 GMT
|
Viewed:
|
776 times
|
| |
|
|
"Lars C. Hassing" <lch@ccieurope.com> wrote in message
news:G2AJD0.KAC@lugnet.com...
> Travis Cobbs wrote...
> > Thanks for this info, Lars. I'll be updating LDView prior to releasing • v1.0
> > to take these things into account. I'm probably going to take rule 2 • and
> > recursively apply it up the line to verify that none of the ancestors • are
> > parts. Does anybody see any problems with this?
>
> Yes, I think it is unncessary. Why would you do that?
If you have a part which references a subpart which references another
subpart (not likely, true), you don't want to shrink either one of the
subparts. However, my suggested solution is not a good way; you're right.
I wasn't setting the part flag on subparts, but this is probably wrong. I
need to just not shrink subparts. By setting them to be parts anyway,
though, only their parent part would have to be checked at the time of
shrinkage. The trick is that I don't have a seamable flag which is separate
from my part flag.
> Besides it would hurt performance. With the current scheme I set a
> Seamable flag on LT1 references once during input. Then if seams
> are turned on I can simply test the flag when traversing the tree.
I actually do all the seam-stuff as pre-processing (which is why it has to
reload the model when you change the value of the seam width in LDView).
I'll probably be changing this somewhat in the future, but not right now.
> > I'm a little confused here, though, because it seems to me that rule 2 • would
> > prevent the seams from appearing in subparts, as long as they are truly
> > that: SUBparts (although sub-subparts would have problems with a
> > single-level check for your parent being a part). If someone were to
> > reference a subpart directly, then seams would appear, of course. I • think
> > the recursive check would prevent spurious seams for all subparts. Am I
> > missing something here?
>
> I'm not sure what you're aiming at. Can you give an example?
I think I was misunderstanding the situation. I thought that it was being
reported that L3P/L3Lab were putting seams between subparts, when that
apparently only happens when you open the part directly, causing it to be
unaware that the top-level model is a part.
--Travis Cobbs (tcobbs@san.REMOVE.rr.com)
|
|
Message has 1 Reply:
 | | Re: Seams like a problem
|
| (...) The first time I noticed it was while rendering in POV-ray a part for which the .pov file was generated via L3P. The part consisted of four identical subparts meeting at (0,0,0). I mirrored the subpart in the manner of 8-8sphe.dat, so that the (...) (24 years ago, 12-Oct-00, to lugnet.cad.dev)
|
Message is in Reply To:
| | Re: Seams like a problem
|
| Travis Cobbs wrote... (...) Yes, I think it is unncessary. Why would you do that? Besides it would hurt performance. With the current scheme I set a Seamable flag on LT1 references once during input. Then if seams are turned on I can simply test the (...) (24 years ago, 11-Oct-00, to lugnet.cad.dev)
|
16 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
|
|
|
|
