Hi,
	Is there a way with POV-Ray 3.1g to declare multiple cameras for the same
parse?  To accumulate the data for each camera and then give a result of the
combination?  In other words, is it possible to have antialiasing, motion blur,
composition, etc... I know there is a setting for antialiasing, but if there is
a way to deal with an accumulation buffer, let me know!

Thanks,
	Simon
-- 
+-------------------------+----------------------------------+
| Simon Lemieux           | Website : http://www.666Mhz.net  |
| Email : Sin### [at] 666Mhznet | POV-Ray, OpenGL, C++ and more... |
+-------------------------+----------------------------------+

Post a reply to this message

On Sat, 28 Oct 2000 08:51:08 -0400, Simon Lemieux
<lem### [at] yahoocom> wrote:

>Hi,
>	Is there a way with POV-Ray 3.1g to declare multiple cameras for the same
>parse?  To accumulate the data for each camera and then give a result of the
>combination?  In other words, is it possible to have antialiasing, motion blur,
>composition, etc... I know there is a setting for antialiasing, but if there is
>a way to deal with an accumulation buffer, let me know!
>
>Thanks,
>	Simon

No, there is not. However, you can render multiple frames with the
different cameras and then average the resulting images. I know it
works as I've done it, maybe you can even find the POV file that I
used to make these tricks somewhere in the povray.*.scene-files
groups.


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] usanet
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message

> No, there is not. However, you can render multiple frames with the
> different cameras and then average the resulting images. I know it
> works as I've done it, maybe you can even find the POV file that I
> used to make these tricks somewhere in the povray.*.scene-files
> groups.

That is exaclty what I thougth...  but then I realized that One povray file
should only contain One camera... and thought, if it contained more cameras, the
renderer could render each one separatly, and creating a file
name_of_project-camera_name.number_of_frame.file_format -> 
test.Cam1.001.tga
test.Cam2.001.tga
test.Cam1.002.tga
test.Cam2.002.tga
and so on...

the camera parameter could have at least one other parameter, name, and it could
even have another parameter for integrated accumulation, accum.
like this:
camera {
	name "Cam1"
	location <-1,0,0>
	look_at <0,0,10>
	accum 0.5  //  1.0 should be 100% opaque, while 0% should be 100%
transparent...
}
camera {
	name "Cam2"
	location <1,0,0>
	look_at <0,0,10>
	accum 0.5
}

This setting should produce the same focus as eyes actually do plus the distance
focus, implementing these camera in a robot should give it the distance between
it and the looked_at object...

Does this sound interesting?

-- 
+-------------------------+----------------------------------+
| Simon Lemieux           | Website : http://www.666Mhz.net  |
| Email : Sin### [at] 666Mhznet | POV-Ray, OpenGL, C++ and more... |
+-------------------------+----------------------------------+

Post a reply to this message

This could work. If you use MegaPov you can also render it as an animation with a
#switch statement, using the frame_number as the parameter, so each frame would give
you a separate tga file to work with.

As for adding the "name" option, the development team is looking at ways to add onto
the language, and I would like to see something a bit closer to object oriented
language come in so I could set something to be at "cam1.lookat" or in the camera
statement have "look_at MySphere.center".

Josh

Simon Lemieux wrote:

> > No, there is not. However, you can render multiple frames with the
> > different cameras and then average the resulting images. I know it
> > works as I've done it, maybe you can even find the POV file that I
> > used to make these tricks somewhere in the povray.*.scene-files
> > groups.
>
> That is exaclty what I thougth...  but then I realized that One povray file
> should only contain One camera... and thought, if it contained more cameras, the
> renderer could render each one separatly, and creating a file
> name_of_project-camera_name.number_of_frame.file_format ->
> test.Cam1.001.tga
> test.Cam2.001.tga
> test.Cam1.002.tga
> test.Cam2.002.tga
> and so on...
>
> the camera parameter could have at least one other parameter, name, and it could
> even have another parameter for integrated accumulation, accum.
> like this:
> camera {
>         name "Cam1"
>         location <-1,0,0>
>         look_at <0,0,10>
>         accum 0.5  //  1.0 should be 100% opaque, while 0% should be 100%
> transparent...
> }
> camera {
>         name "Cam2"
>         location <1,0,0>
>         look_at <0,0,10>
>         accum 0.5
> }
>
> This setting should produce the same focus as eyes actually do plus the distance
> focus, implementing these camera in a robot should give it the distance between
> it and the looked_at object...
>
> Does this sound interesting?
>
> --
> +-------------------------+----------------------------------+
> | Simon Lemieux           | Website : http://www.666Mhz.net  |
> | Email : Sin### [at] 666Mhznet | POV-Ray, OpenGL, C++ and more... |
> +-------------------------+----------------------------------+

--
Josh English -- Lexiphanic Lethomaniac
eng### [at] spiritonecom
The POV-Ray Cyclopedia http://www.spiritone.com/~english/cyclopedia/

Post a reply to this message

> This could work. If you use MegaPov you can also render it as an animation with a
> #switch statement, using the frame_number as the parameter, so each frame would give
> you a separate tga file to work with.
> 
> As for adding the "name" option, the development team is looking at ways to add onto
> the language, and I would like to see something a bit closer to object oriented
> language come in so I could set something to be at "cam1.lookat" or in the camera
> statement have "look_at MySphere.center".

I figured something for this:
let say the renderer (my own tool) knows how many cameras there is -> nCams;
and that both the tool and the pov file know what is the min/max clock ->
mClock;

then you could say that mClock is for Cam1, mClock*2 is for Cam2, and so on...
You can have a switch to see wich camera it is and give different settings for
each one... and then the tool could do the accumulation itself...  I would
personnally use OpenGL for this, since it can be accelerated and offline.

What I wanted to know was if POV-Ray alread had such a feature, if not it's a
matter of time 'til I define it myself...

Thanks,
	Simon
-- 
+-------------------------+----------------------------------+
| Simon Lemieux           | Website : http://www.666Mhz.net  |
| Email : Sin### [at] 666Mhznet | POV-Ray, OpenGL, C++ and more... |
+-------------------------+----------------------------------+

Post a reply to this message