|
 |
What in particular are you interested in seeing about the code??
All the textures are found in various IRTC submissions I've made in the
past. The only new things is the arrangement of objects and as I always do,
it is all POV primatives.
What I can tell you is a bit about the technique I use in developing the
scene, which would not be apparent from the source.
If you look at the earlier posting and compare it to the current one, you
will see objects that in the earlier work are very basic, or simply a
primative. Specifically, compare the original SW Gate, nearest you in the
view... then look at the detail in the later picture. If you place objects
in a scene, utilizing a simple cube or relative object to the final object,
it becomes very simple to continue to develop the details on the object
outside of the final composition.
I maintain a second source code .POV file that I copy objects from the big
render into, add detail, render, rotate, etc. to my heart's content,
focusing on just the detail of that object. When I've created all the
detail I can come up with for that object on that round, I copy all the
objects from the test member and then replace them in the "master" render.
All objects are rendered "Zero-centered" meaning that they are all created
around the origin at 0,0,0. Placement of an object takes it from <0,0,0>
and translates it to its final location in the scene. Since in my testing,
I keep the object at 0,0,0, all objects in that render show up centered.
Once I move them to the master render, they are already transfered to their
final location by the earlier placement of the generic object.
For example:
In the master:
#declare Tower1=union{
box{<-5,0,-5><5,10,5>} // square tower
box{<-5.5,10,-5.5><5.5,12,5.5>}
}
object{Tower1 translate<-40,0,-40>} // translate the object to the corner
As you can see, I define an object via a CSG, but it is no more than two
cubes one on top of the other. I also put that object in a location in the
final scene via the object define below the definition. I can use several
"Tower1" object defines, defining the same tower on all four corners of the
castle (for example).
Now, If I cut just the definition for Tower1 from the master and using a
test.pov member, I can play with just the tower definition, add
crenallations to the top, difference out the windows, adjust texture, add a
foundation set of blocks, flagpoles, etc. and render over and over in my
test.pov member until I am satisfied with the result. I would then copy the
definition and any related defs from test.pov and replace the CSG for Tower1
with the new defs. I don't have to touch the "object" definition, leave it
alone. I re-run the entire master scene again and walla, new towers appear
in place of the previous cubes.
I repeat this process over and over again for every object in the scene.
Since in my test member, they all appear relative to 0,0,0, you can add
details and such to the individual objects right where you want them and
then with the whole lot bundled up as a CSG that replaces the previous one
in your master file, every component of that adjustment is exactly in place
even though it is translated elsewhere in your scene.
Again, if there is something specific in the render you are interested in,
please ask.
Regards,
Robert J Becraft
aka cas### [at] aol com
Alessandro Coppo <a.c### [at] iolit> wrote in message
news:3a466696@news.povray.org...
> It is becoming REALLY beautiful. Are you going to release the code
> (please!!!)?
>
> Alessandro Coppo
> a.c### [at] iolit
> www.geocities.com/alexcoppo/
>
>
>
Post a reply to this message
|
|
