"Divine Wind" <div### [at] blueyondercouk> wrote:
> ..... then once the char or other object is "improved
> then re-enter it back into the main code.
This is very easy to do in pov-ray. it is usually don by keeping each object
or groep of objects in seperate files, and then "include" (link to) these
files
from the main file, only the needed parts. but there are other ways to do it
as well.

> The idea of limiting the render time using a principle of is the object
> behind the camera or not, and is there a refelective object infront of the
> camera seems to me to be a good idea.
should not make mutch difference. pov-ray automatically groeps (parts of)
objects into larger blocks. if the whole block is outside the camera view,
pov-ray will skip all parts in that block automatically. pov-ray will also
make this work properly when objects are visable in reflextions.
Typically the only reason for disabeling objects outside the camera view is
when
you run out of memory. (if you run out of memory then you have something
else on your mind, becouse the images will render to slow for an animation
:-)

> Also simplifing the object if it is small and a large distance off.
This is very easy to do in pov-ray: let povray calculate the distance
between
the camera and the object, and select one of the versions, or draw nothing
at
all. for animations, you could also replace distant objects with a flat
bitmap image of the object, and rotate it towards the camera automatically.
(but the shadows will look very strange)

> My base idea is to creat an island where the story will take place this
> means that I will not have to create any other things in the future for the
> landscape. This will have to be modified as the storie  unfolds.
>
> Thanks for the idea of blender. Where can I get it from?
www.blender.org (i have not used it in a long time)

> About the actual animation. is DTA to be found on the POV site a good idea
> for the small scenes. Then moving to adobe premier to put the scenes to
> gether with a good idea?. Do you (japp) or any one else have any experance
> in doing/useing these tools?. If not which ones do you use?
i never looked into DTA, anyone else ? (i not a big fan of adobe product, i
usually use some kind of dos/commandlin (and now linux) tool to convert
images
into movies. the free "aviedit" (just google for it) program also works but
can't handle long animations (about 200(?) frames max) "ffempeg" tools work
ok,
there are many discussions about this at povray.binaries.utilities.

i would set up a new "main" povray file for each shot, and include all the
needed objects by adding links to other files. textures are defined in yet
another file. (maybe a file with simple and a file with complicated
textures with the same texture names, so you can link to the file with
simple textures when rendering tests.) the camera movement and the position
of all the moving objects (during the shot) can then be defined in the main
file (for each shot),
while the actual objects are defined in there own file, so you can easaly
re-use the "object" file for other shots.

you can also make special main pov-ray files, that have only a fixed camera
and some simple light and link to a single object, as a tool for when
creating or
changing objects.

you normaly waht to define the camera position at the top of the main
pov-ray file, and include all other files after that. this way you can use
the camera position in all files you include. (like: for calculating the
distance beteen the camera and an object)
This also makes another trick possible: the included files can detect if a
camera position was defined. if not, the included file can dicide to use a
test camer and light. this way you can render the include files on ther own
when
working on the details of an object.

if you use file names like <shot_nr><frame_nr>.png (s001f001.png) then you
can
sort the files alfabetacally and convert them to movie file, but can still
make changes to the length (nuber of frames) of each shot.
(number shots like 10,20,30, so you can insert more shots later)
you can also "edit" the movie after rendering just by renaming the frames of
part of a shot to a different shot.

(i once made a pov-ray file that rendered a black image, but also produced a
dos batch file, that could rename large numbers of images.
it's in  povray.binaries.utilities as "pov -> ffmpeg -> ogg theora tool.")

as you can see, povray is very flexible. basically the "script" is run for
each frame of your animation, and the script can make calculations and
decisions for each frame before actually rendering it.

there are many ways to do the same thing in pov-ray. just use what looks
best, and if it does not work, is to hard work, or is to slow, just ask.
ppl here will usually be able to sugest a different way to do the same
thing.

have fun,
jaap.

Post a reply to this message