In article <pgp### [at] tritonimagicode>,
 Christoph Hormann <chr### [at] gmxde> wrote:

> But materials can't be targets, only objects can.  The only way you can 
> implement this is to make every object somehow 'inherit' the photon 
> settings of its material.  But this would not be obvious to the user so 
> it will just cause confusion in most cases.

Why? It "inherits" the ior and media of its interior, the pigment of its 
texture, etc. There's nothing about photons that ties them specifically 
to the object, it doesn't own the photons. Making them an aspect of the 
material makes perfect sense to me. Glass objects have caustics...it's 
an aspect of glass, not of the objects.


> Most likely not.  For efficiency you will only use photons on those 
> objects where the caustics are visible in the scene.  Glass objects out 
> of direct view that are only visible as reflections in other shapes for 
> example will probably not use photons.

Even if they are out of direct view, their photons may be in view 
directly, or in reflections. And for the case where they really don't 
contribute, provide a mechanism for turning them off.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tagpovrayorg>
http://tag.povray.org/

Post a reply to this message

In article <402### [at] hotmailcom>,
 andrel <a_l### [at] hotmailcom> wrote:

> > I didn't quite follow you. Can you give an example?
> Well, if you have an object which is partly glass
> and partly some opaque material and you add 'target off'
> to the latter part. Will the resulting object collect photons
> (because of the glass part that switches it on, and
> it is inherited by the total object) or not because
> of the 'target of' in the other part. Will always the
> first declaration take precedence or the last (or
> even a middle one).

The parts of the object that are glass will be photon targets. The parts 
with the opaque material will not. You can't apply multiple materials to 
a simple object, so there's no problem. (well, there may be a problem 
with difference and intersection, but no more than there is now)

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tagpovrayorg>
http://tag.povray.org/

Post a reply to this message

In article <402### [at] hotmailcom>,
 andrel <a_l### [at] hotmailcom> wrote:

> Photons as a material property makes sense, but that would
> probably mean a large rewrite of the code (if even possible).

I highly doubt this. Actually, it would probably only require changes to 
parse.cpp...

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tagpovrayorg>
http://tag.povray.org/

Post a reply to this message

Ok, I seem to have misunderstood you somewhere.
I thought that because the photons are an object
thing now, you wanted to promote the photon setting
from any material in an object to the object itself.

   Andrel

Post a reply to this message

news:402### [at] hotmailcom...
> Ok, I seem to have misunderstood you somewhere.
> I thought that because the photons are an object
> thing now, you wanted to promote the photon setting
> from any material in an object to the object itself.

Oh, not at all. Sorry if I was unclear. If there was an object that is an
union of glass and wooden part, then the glass part should receive photons,
not everything. And if you, for any reason, don't want that glass part to
receive photons, then specify target off, or whatever. I just think it
should work just like caustics.

Post a reply to this message

Christopher James Huff wrote:
> 
> Why? It "inherits" the ior and media of its interior, the pigment of its 
> texture, etc. There's nothing about photons that ties them specifically 
> to the object, it doesn't own the photons. Making them an aspect of the 
> material makes perfect sense to me. Glass objects have caustics...it's 
> an aspect of glass, not of the objects.

I'd suggest you (and everyone else who thinks photons are just like 
(faked) caustics) to read up how they work before stating such 
conclusions.

The idea behind making photon settings part of the material certainly is 
to make it unnecessary for the user to think about which *objects* he 
wants how many photons for.  But since doing so will not actually change 
anything in the way POV-Ray distributes the photons this will not result 
in any actual improvement of the output quality or the render efficiency 
- it would just create the illusion for the user that he does not need 
to care about which objects should have photons.

What would really be interesting is a way for POV-Ray to automatically 
calculate the target values for the objects that receive photons based 
on their importance for the render result.  There have been various 
attempts for importance based photon distributions, a good introduction 
can (although focussed on global photons) be found in:

http://graphics.uni-ulm.de/Importance.pdf


Christoph

-- 
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 11 Jan. 2004 _____./\/^>_*_<^\/\.______

Post a reply to this message

> Oh, not at all. Sorry if I was unclear. If there was an object that is an
> union of glass and wooden part, then the glass part should receive photons,
> not everything. And if you, for any reason, don't want that glass part to
> receive photons, then specify target off, or whatever. I just think it
> should work just like caustics.
I think I understood what you wanted to achieve.
I had a problem with the way I thought it was supposed to
be realized. It is all clear to me now (I am a valid
target for photons again). Thanx,

    Andrel

Post a reply to this message

In article <vib### [at] tritonimagicode>,
 Christoph Hormann <chr### [at] gmxde> wrote:

> I'd suggest you (and everyone else who thinks photons are just like 
> (faked) caustics) to read up how they work before stating such 
> conclusions.

Completely irrelevant, and somewhat insulting. I understand how photons 
work. Yes, they are computed differently than faked caustics, that 
doesn't matter one bit. If transparency, reflection, and refraction are 
material attributes, it only makes sense for photons to be so as well.


> The idea behind making photon settings part of the material certainly is 
> to make it unnecessary for the user to think about which *objects* he 
> wants how many photons for.  But since doing so will not actually change 
> anything in the way POV-Ray distributes the photons this will not result 
> in any actual improvement of the output quality or the render efficiency 
> - it would just create the illusion for the user that he does not need 
> to care about which objects should have photons.

The point is to make it more convenient to write scenes. The fact that 
it won't improve efficiency or quality is entirely aside from the 
point...it isn't intended to do so in the first place.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tagpovrayorg>
http://tag.povray.org/

Post a reply to this message

Christopher James Huff wrote:
> 
>>I'd suggest you (and everyone else who thinks photons are just like 
>>(faked) caustics) to read up how they work before stating such 
>>conclusions.
> 
> 
> Completely irrelevant, and somewhat insulting. I understand how photons 
> work. [...]

Since you wrote:

> There's nothing about photons that ties them specifically 
> to the object, it doesn't own the photons.

I had serious doubts about this...


Christoph

-- 
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 11 Jan. 2004 _____./\/^>_*_<^\/\.______

Post a reply to this message

In addition to what Christopher said...

Christoph Hormann wrote:
> I'd suggest you (and everyone else who thinks photons are just like 
> (faked) caustics) to read up how they work before stating such conclusions.

I think I know how photons work and yes, the aim is to produce (in case 
of refractive photons) more accurate caustics. I never claimed that 
"photons are just like (faked) caustics". But the goal is very similar 
in many cases.

> The idea behind making photon settings part of the material certainly is 
> to make it unnecessary for the user to think about which *objects* he 
> wants how many photons for.  But since doing so will not actually change 
> anything in the way POV-Ray distributes the photons this will not result 
> in any actual improvement of the output quality or the render efficiency 
> - it would just create the illusion for the user that he does not need 
> to care about which objects should have photons.

I'm starting to repeat myself, but here goes:

1. The suggestion was to make it easier to alter material 
characteristics. The _result_ of using photons is a material property: 
more realistic caustics and light reflections. Only the calculation 
method differs greatly from other properties and non-visible objects 
have an impact on total render time. I don't understand why this can be 
a reason to make the usage of photons more cumbersome than other 
material properties. Of course there should allways be ways to control 
the amount of time/memory consumed for this calculation.

2. The quality actually might improve as user might omit the usage of 
photons on an object that is not visible whereas reflected of refracted 
photons might still be visible in the field of view - even via 
reflections. The possibility of this happening might be reduced if 
photons were a material property. (And there is no point putting objects 
into scene which don't affect the output in any way...)

3. Why should POV-SDL have "baby-sitting" (no offense meant) features? I 
allways considered it to be quite powerful tool and it is up to user to 
understand what he is doing. It would have been equally odd, if (say, 10 
years ago) you would have been forced to specify refraction for each 
objects individually without having the possibility to inculde it to 
texture declaration: "They take very long to render and you must 
consider every object very carefully..." There are documents that 
explaine the usage of various features and I think users can judge 
themselves how to use them.

> What would really be interesting is a way for POV-Ray to automatically 
> calculate the target values for the objects that receive photons based 
> on their importance for the render result.  There have been various 
> attempts for importance based photon distributions, a good introduction 
> can (although focussed on global photons) be found in:

That would indeed be a good thing. Has any attempts made to implement 
such a thing in Pov-RAY?

Severi

Post a reply to this message