Subject:
|
Re: [ldraw.org] L3P Question
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Thu, 15 Apr 1999 01:02:07 GMT
|
Viewed:
|
698 times
|
| |
|
|
On Wed, 14 Apr 1999 14:05:07 GMT, "Lars C. Hassing" <lch@ccieurope.com> wrote:
> > Now, for the camera views.
> >
> > First, screw all the other methods of figuring out how to
> > place a camera. Stop trying to fathom X,Y,Z coordinates
> > with your fingers, and put your left hand down. :-)
> > Forget the lattitudes and longitudes, and translations and
> > vectors and all that other $%#!.
> >
> > Once you have the .pov file made, go back into LEdit. Place
> > a light.dat about where you want the camera to be. Then
> > place another one where you want the camera to look_at or point.
> >
> > Save the file. Open it in a text editor. The first 3
> > coordinates of the first light dat directly correspond to
> > the camera location coordinates in the pov file.
> > copy and past them from the .dat to the .pov.
> >
> > Do the same with the second light.dat's coordinates. Paste
> > them into the cameras Look_At coordinates.
> >
> > Make sure you paste into the proper places.
> >
> > Render and thank me later. :-) (1)
> > Or better yet, thank Lars for using the same coordinate
> > system that makes this possible. (2)
>
> Oh no, no, no!
> What is this? Is it that difficult to use the latitude/longitude?
> Using the above copy-n-paste will bypass L3P's efforts to place
> the model in the center of the image with equal amount of space
> around it. You risk cutting off part of the model.
Lars, I didn't intend for this to be a "fix" of any type. And the
latitude/longitude method is a good one to use. If I sit there thinking about
it enough, I can do it.
And the automatic centering of the model is a fantastic feature that I love.
I just offered this method as a fairly simple way of directly finding xyz
coordinates to use in an existing pov file - using the simple expedient of
moving a point around in LEdit.
> I used the globe concept because I assumed it was easy to grasp.
> No worry about X Y Z coordinates, just imagine your model inside
> a globe. I even made an illustrative image (at least I thought so):
> http://www.netby.net/Nord/Mandelvej/Hassing/l3p_i/L3GLOBE2.jpg
>
> Could you (or others) please try to explain what the difficult part is.
>
> How did I fail to make the description in
> http://www.netby.net/Nord/Mandelvej/Hassing/l3p.html#l3globe
> intelligible and explanatory?
>
> That said, I admit L3P lacks a way of specifying the look_at
> position, which currently is set to the center of the model.
> But the camera position may also be set by -cc<x>,<y>,<z> and
> light.dat may be recognized by the -l options (see also -ld).
> /Lars
No, I can follow your explanation. And I can even envision a model in my head
and extrapolate where to change the camera coordinates to. Your system works
fine, and I am in no way denigrating it. Sorry if it seemed that way. I was a
bit insensitive in my remarks. :-(
My method, using LEdit, just allows me to be lazy. I can move the point
(light.dat is just handy for that) around to where I think it should be, then
look at the coordinates and use those numbers. No straining of my brain
required. :-)
-- Terry K --
|
|
Message is in Reply To:
 | | RE: [ldraw.org] L3P Question
|
| (...) I agree on that. I really was in doubt whether or not to add the logo, but after looking at many renderings I found it to increase realism very much (as do textures in general, compared to "flat" faces). I also considered writing e.g. "LARS" (...) (25 years ago, 14-Apr-99, to lugnet.cad.dev)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
