rebeltaz wrote:
> I used POV ten or fifteen years ago and it was a Windows-based GUI system. I
> have long since forgotten everything that I knew, and I have moved on to Linux.
> Since there aren't many, if any, GUI modeling programs for Linux, I was looking
> at the tutorials on writing the code in a text editor. While I am perfectly
> capable of writing code using a text editor in numerous BASIC, C++, PHP and HTML
> languages, that's more of an abstract process.
> 
> What I am curious about is hard to explain. I guess... how do you 'see' what you
> are designing when you are only working with code. How do you know that the
> camera goes here. How do you that this shape is at these coordinates. I guess I
> am having a hard time understanding how you plot an image that you can't see.
> 
> I hope what I am asking makes sense. I'm not asking for a tutorial on how to
> write the code. More of the mentality behind writing the code. Or maybe I am
> missing something?
> 
> Thanks for your patience in reading this (since you got this far :)
> 
> 

A small idiosyncratic note related to positioning the camera.  I try to 
leave the main focus of my scene at the origin or maybe directly above 
the origin. So I point the camera there.  I then position the camera, 
and sometimes the lights too, utilizing the vrotate command.  Personally 
I have found this a more intuitive way to control the distance and angle 
of the camera to a subject.

e. g.
camera {
   location  vrotate ( <0,0,-10>, <20,30,0> )
   look_at   <0.0, 0.0, 0.0>
}	
This would place the camera at 10 units distance along the -z axis then 
rotate it around the subject, upward 20 degrees, then to the left 30 
degrees.


It is not for every situation, of course. Sometimes you know you want to 
locate the camera 'over behind the couch' or 'at the eye height of a 
figure in the doorway' and have it look at 'the center of the table', 
and so you do that with simple vectors like all the other objects.

Post a reply to this message