Subject:
|
Re: Aspect ratio (via L3P)
|
Newsgroups:
|
lugnet.cad.ray
|
Date:
|
Fri, 28 Jun 2002 17:18:55 GMT
|
Viewed:
|
982 times
|
| |
|
|
Sproaticus wrote...
> In lugnet.cad.ray, Lars C. Hassing writes:
>
> > There is still time for more requests.
>
> Lars,
>
> Thanks for your good work, and I have just a few requests, in order if
> decreasing importance:
>
> - option to include a file at end and/or begin of .pov file (e.g. for
> megapov's edge-detection feature, etc.)
How about
-ib<filename> include file at beginning of POV file
-ie<filename> include file at end of POV file
> - placing camera block, BoundingBox info into variables for use by included
> macros
Good idea. I have already used variables in the revised camera definition, like
L3Location and L3LookAt, and I could add:
#declare L3ModelBBoxMin = <99,99,99>;
#declare L3ModelBBoxMax = <99,99,99>;
#declare L3ModelBBoxCenter = (L3ModelBBoxMin+L3ModelBBoxMax)/2;
#declare L3ModelBBoxSize = L3ModelBBoxMax-L3ModelBBoxMin;
You can access e.g. the model height by L3ModelBBoxSize.y
Can you think of other variables ?
> - option to output models and/or submodels as separate .inc or .pov files
I need to understand the reason for this request, because I can imagine
some problems with managing these files. E.g. color usage.
How would you use the files ?
> re: shadowless lights & include camera block -- woo hoo, you read my mind!
> :-D :-D
>
> ... in other news ...
>
> I have noticed in some models with some wildly-rotated subparts (mecha for
> example) there tends to be a lot of dead space to the side or top or bottom
> of the image. This dead space goes away if I inline the datfile before
> running l3p. I think it's a nested bounding-box issue. If l3p could offer
> an option to inline a file in RAM before converting it to POV, that would be
> cool -- but I hesitate to ask for it since I don't feel it really falls
> within the scope of l3p's intent.
You are correct. Both about the cause of the problem and about the scope of L3P :-)
L3P works on one file at the time (it doesn't walk the tree), and when it processes the
model, the parts and submodels have been thrown away and cannot be used in
the automatic camera positioning calculation.
Some day when there are no more L3P16 users around I will change the algorithm
and let L3P have the whole tree in RAM and the problem will be solved.
Hm, come to think of it, I might be able to inline model and submodels
but not parts/primitives, without too much memory usage.
But it's a balance, you wouldn't be able to process DatsVille (having many submodels)
with L3P16. But then again, if you want to do DatsVille you would probably need
a newer computer anyway and you can use the 32-bit L3P.
Triangular parts like wings can also give some dead space. To get rid of that L3P would
have to do a full tree walk to visit to every triangle/quad and use their vertices.
But with todays computers it probably won't be an issue.
> Just a few thoughts, and again, thanks for the great stuff.
You're welcome, and thanks for the suggestions.
/Lars
|
|
Message has 1 Reply:
 | | Re: Aspect ratio (via L3P)
|
| (...) Awesome. (...) These are the only variables I can think of. I think it might also be useful to expose the formulas for determining the camera and light positions via macros, but I can't think of any real-world needs at the moment. (...) Well, (...) (22 years ago, 29-Jun-02, to lugnet.cad.ray)
|
Message is in Reply To:
| | Re: Aspect ratio (via L3P)
|
| (...) Lars, Thanks for your good work, and I have just a few requests, in order if decreasing importance: - option to include a file at end and/or begin of .pov file (e.g. for megapov's edge-detection feature, etc.) - placing camera block, (...) (22 years ago, 26-Jun-02, to lugnet.cad.ray)
|
20 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
|
|
|
|
