Subject:
|
Re: Several replies to CLIPPING / WINDING
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Tue, 19 Oct 1999 12:51:57 GMT
|
Viewed:
|
431 times
|
| |
|
|
Rui Martins wrote in message ...
> On Mon, 18 Oct 1999, Leonardo Zide wrote:
>
> > Rui Martins wrote:
> > >
> > > Well, the ideia is to minimize the state change, but if you have to make
> > > some, that is not so drastic. On some hardware, some state/option
> > > changes, force a flush in the hardware buffers, which breaks up the
> > > pipeline & eventually hurts performance, but this is not the case.
> >
> > I understand that you would be allowed to have as many "0 CCW"/"0 CW"
> > lines as you want. This will hurt performance and of course, it would be
> > better if you could use only 1 call per part.
>
> In fact you can, use as many as you like, but that it's not the goal.
> if you read my previous mail, following this one I'am replying, you got
> the ideia that no WINDING meta command is the ideia, if the .DAT is well
> optimized.
>
> If someone want's to, he can make a very heavy .DAT file, by doing
> to many state changes, or even do state changes with no usefullness,
> just to bog down the drawer program, we can' fight a bad behaved
> builder. But in the other way, no one would see is files, no one likes
> to wait too much.
Wait a minute, you were not going to actually send winding state changes
to the graphics engine?
Simply keep track of the state yourself and send either vertices 1,2,3
or 3,2,1.
The part author can use as many WINDING CW/CCW as he likes
- it won't hurt performance a bit!
> > To use a vertex array you also need to have all vertexes indexed,
> > which is not the case of the DAT files and one of the reasons why I have
> > my own file format.
>
> To Drawer program, can do this job of indexing.
> But I know that it is difficult, when some programs put 2,0001 or
> 1,99998 instead of 2, for example, this could be a goal for improvement
> in ldraw compatible editors.
I have already done that. After reading a part into memory I index it.
It typically reduces the number of vertices and thus transformations
to 20-50%. However, as the math currently takes less than 10% of the
rendering time in L3, the total saving is only 2-5%
I don't think it is that difficult: a part may use 2.0001 or 1.99998 in
stead of 2, but not for the same vertex. If it did there would be gaps,
which of course should be fixed.
/Lars
> ( VI/NOTEPAD are the best for the JOB! 8)
vi rules! :)
|
|
Message has 2 Replies:
 | | Re: Several replies to CLIPPING / WINDING
|
| (...) NO, I said that, to be used when optimizing the existent files, not when rendering the optimized files. (...) I never said that ! (...) I think that depends on the host processor that you have and your specific graphics board (2D/3D) ;) (...) (...) (25 years ago, 20-Oct-99, to lugnet.cad.dev)
|
Message is in Reply To:
| | Re: Several replies to CLIPPING / WINDING
|
| (...) In fact you can, use as many as you like, but that it's not the goal. if you read my previous mail, following this one I'am replying, you got the ideia that no WINDING meta command is the ideia, if the .DAT is well optimized. If someone want's (...) (25 years ago, 18-Oct-99, to lugnet.cad.dev)
|
15 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
|
|
|
|
