Subject:
|
Re: BFC: LITS 2
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Fri, 31 Mar 2000 20:51:23 GMT
|
Highlighted:
|
(details)
|
Viewed:
|
1842 times
|
| |
|
|
Fourth in the series, starting with message #41 in the "Line in the
Sand" thread. Notice that I scanned past a number of messages before I
got another 'open issue' hit. This posting covers up to & through
message #83, the end of the thread.
In <http://www.lugnet.com/cad/dev/?n=3517>, Lars points out:
> 1 16 -4 0 0 0 0 0 0 0 4 1-4disc.dat
> The determinant is zero, which BTW causes POVRay to halt
> ("singular matrix"). L3P has to fix these matrices or POVRay
> would not render many parts at all!
> As part of certifying a DAT file I believe this kind of subfile
> reference should be fixed as well, like into
> 1 16 -4 0 0 0 1 0 0 0 4 1-4disc.dat
> "L3P -check -w3" will help spot the problems.
This point hasn't been discussed since, I believe. This is very valid,
these "singular transformation matrices" should be fixed.
BTW, this gives the CERTIFY tag something to define besides clipping and
winding. In order to cleanly certify a file, it should *not* have any
subfile commands with singular matrices.
Even so, CERTIFY is still functionally redundant...
------------------------------------------------------------------------
One other thing. Starting somewhere around message #3424
<http://www.lugnet.com/cad/dev/?n=3424>, Lars and I got to discussing
what purpose would be served by intentionally disabling clipping within
a file (via the NOWIND option, if that one is preserved in the spec,
otherwise using NOCLIP).
I contended that decorations will need to be flagged as double-sided, so
that they will show through the part when it is used as transparent.
Lars responded that no clipping can be performed on transparent
quads/tris, because some renderers will need to use the backside objects
to provide more depth to the transparent areas, so the spec should
explicitly require that BFC be disabled when rendering a
transparent-colored part.
I replied that that was an unreasonable solution because a part-file may
have hard-coded transparent sections, even though the reference to the
file is a solid color. Think of rendering <part:2634> in white. The
renderer would have to prescan the file, to find the hard-coded
transparent sections. And then the entire file would have BFC disabled.
Lars pointed out that we should not require parts-authors to code
anything that the rendering engine can figure out for itself.
And that was the end of intelligent discussion on that point.
------------------------------------------------------------------------
A side issue of that discussion: can transparent surfaces be clipped?
Steve
|
|
Message has 2 Replies:
 | | Re: BFC: LITS 2
|
| Steve: (...) No. - Or rather: That depends on how good looking a rendering you want. Thinking a bit more about it, I think we can get just as good result _with_ clipping of transparent surfaces as without. Yes, transparent surfaces can be clipped! (...) (24 years ago, 1-Apr-00, to lugnet.cad.dev)
| | | Re: BFC: LITS 2
|
| (...) I believe that the only problem is that these matrixes can't be inverted ! I'not sure if placing a one on the specific place has exactly the same graphic behaviour ? can someone confirm ? (...) Use my proposal: 0 CERTIFY BFC MTX where MTX is (...) (24 years ago, 2-Apr-00, to lugnet.cad.dev)
|
Message is in Reply To:
| | BFC: LITS 2
|
| Rui has stated that there are outstanding issues with the BFC proposal that are not included in the issues list at the start of that proposal. With this message, I am going to review the messages from the "Line in the Sand" thread, which started at (...) (24 years ago, 30-Mar-00, to lugnet.cad.dev)
|
24 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
|
|
|
|
