Subject:
|
Re: Some Words To BFC
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Wed, 5 Apr 2000 15:01:00 GMT
|
Reply-To:
|
rui.martins@link.NOSPAMpt
|
Viewed:
|
2655 times
|
| |
|
|
On Wed, 5 Apr 2000, John VanZwieten wrote:
> Rui Martins <Rui.Martins@link.pt> wrote in message
> [...SNIP...]
> > This assumption can be made, since all our *.dat files are used to define
> > solid objects (it should be solid, lego parts are, only stickers don't • apply),
> > which tells us that any face can be clipped if it is facing away from the
> user.
>
> Surely you don't mean _all_ our _*.dat_ files! All but a couple of the
> primitives do not represent solid objects, and this is the central problem
> with allowing them to be BFC'd in a non-certified part.
If you assume (which is correct for lego parts) that every *.dat file you use to
define a part, belongs to a whole object (the part) which is a solid object (it
should be) then you can say that any of those subfiles can have their faces
clipped, since they belong to a solid object, as long as they are marked
apropriatly to enable clipping.
> There is no way you
> can tell just from the winding if a 4-4cyli.dat is supposed to face out or
> in.
Right ! But since we have winding and the rule that the faces should point
up/outside then you know, unless we use it inversed without informing the
renderer.
[...SNIP...]
> > Only the unwanted inverts ! (the ones which don't use an invert matrix),
> > which are NOT so many.
>
> Why do you say not so many? Don't all the n x n bricks and plates use a box
> primitive for the outside and the same box primitive for the inside? So if
> we certify the box5.dat without fixing all the bricks and plates, all their
> undersides will dissappear when using BFC.
What you say is true !
But I mean that these cases are simply converted, because you can use a
reference with an inverted matrix, since the object is simetrical.
The cases which are "not so many" are the ones where we can't do this.
[...SNIP...]
> > Not necessarily ! only a few files will show this problem, but every file
> > already made can have IMEDIATE BENEFITS from the revised/compliant *.dat • files,
> > as soon as they are available.
>
> I beg to differ. Some files will show immediate benefits, while others will
> be really messed up.
Is my English that bad, isn't that what I'm saying, but you let the user choose
if he want's BFC or NOT.
> > > Otherwise, someone using MLCad would render a model, only
> > > to find out that a couple parts used are not BFC compliant but inverted a
> > > primitive that is compliant.
> >
> > In this case (which I already tought about) we just have to define as • default
> > that overall clipping (render program option) is DISABLED, which will give • the
> > user the same behaviour of today, if the user want's to increase • performance, he
> > can turn the new feature on (ENABLE), knowing the POSSIBLE consequences if • used
> > with non-BFC compliant files
>
>
> IMHO, BFC isn't worth the work if it's not going to produce reliable results
> along with increased performance. Having even 10% of he parts in my model
> rendered incorrectly is not acceptable.
You can always disable clipping (same output as today) or you can implement a
mark feature (in the renderer or in the DAT file?) which tells the renderer that
it should NOT apply BFC to all files derived from that one. But I think this is
unnecessary.
We should have some view ahead, currently files are NOT conformant, but in some
near future they will be, then you will NOT want the added overhead required to
support the strange cases.
See ya
Rui Martins
|
|
Message has 1 Reply:
 | | Re: Some Words To BFC
|
| (...) In other words, every face in a subfile will have other faces in the part-construction that will form the other side of the solid, right? That sounds correct. (...) There are only a couple of non-symmetrical primitives, but they are used (...) (25 years ago, 5-Apr-00, to lugnet.cad.dev)
|
Message is in Reply To:
| | Re: Some Words To BFC
|
| Rui Martins <Rui.Martins@link.pt> wrote in message news:Pine.GSU.4.10.1...0@is-sv... (...) saying, (...) subfiles (...) those (...) fixed. (...) restrictions, (...) it. (...) until (...) the (...) allows (...) a (...) apply), (...) user. (...) (...) (25 years ago, 5-Apr-00, to lugnet.cad.dev)
|
61 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
|
|
|
|
