First the radiosity-version, then the photon_map-version.

It seems like radiosity does take IOR and reflection into account..

cool :)

What we need now, is a way to "direct" the radiosity samples, so samples
will only be taken in the direction of other objects.  (Wouldn't be of much
use when a sky sphere is used though...)

cu!
--
ZK AKA SaD
http://www.povplace.be.tf
"The only thing that helps me maintain my slender grip on reality is the
friendship I share with my collection of singing potatoes."

Post a reply to this message

Attachments:
Download 'photonsradiosity3.jpg' (8 KB) Download 'photonsradiosity4.jpg' (9 KB)

Preview of image 'photonsradiosity3.jpg'

Preview of image 'photonsradiosity4.jpg'

Zeger Knaepen wrote:
>First the radiosity-version, then the photon_map-version.
>
>It seems like radiosity does take IOR and reflection into account..
>
>cool :)
>
>What we need now, is a way to "direct" the radiosity samples, so samples
>will only be taken in the direction of other objects.  (Wouldn't be of much
>use when a sky sphere is used though...)

All this is pretty interesting and all, but I doubt it's got practical
applications. If you want HDRI and photons and no light_sources just render
a first version with light_sources and use save_file in the photons block
(you can stop the render just after building the photon map, you don't need
actually render the scene). Then get rid of the lightsource/sources and
change save_file with load_file. Simple as that. I haven't followed all
this discussion so I apology if this was already suggested.

--
Jonathan.

Post a reply to this message

"JRG" <jrg### [at] hotmailcom> schreef in bericht
news:web.3e0fa5e4be56d4eb978071460@news.povray.org...
> All this is pretty interesting and all, but I doubt it's got practical
> applications. If you want HDRI and photons and no light_sources just render
> a first version with light_sources and use save_file in the photons block
> (you can stop the render just after building the photon map, you don't need
> actually render the scene). Then get rid of the lightsource/sources and
> change save_file with load_file. Simple as that. I haven't followed all
> this discussion so I apology if this was already suggested.

:)
I dind't know you could save photons :)

But actually these images were just a test to see if radiosity takes reflection
and ior into account, which it does :)

cu!
--
camera{location-z*3}#macro G(b,e)b+(e-b)*(C/50)#end#macro L(b,e,k,l)#local C=0
;#while(C<50)sphere{G(b,e),.1pigment{rgb G(k,l)}finish{ambient 1}}#local C=C+1
;#end#end L(y-x,y,x,x+y)L(y,-x-y,x+y,y)L(-x-y,-y,y,y+z)L(-y,y,y+z,x+y)L(0,x+y,
<.5,1,.5>,x)L(0,x-y,<.5,1,.5>,x)               // ZK http://www.povplace.be.tf

Post a reply to this message

This is where I disagree with POV-Ray's radiosity implementation.
Radiosity is for simulating diffuse interreflection. Simulating specular
interactions during the process of shooting radiosity rays is a waste and is
better done using POV-Ray's excellent photon mapping. Fortunately this was
easy to fix. I did it in my own personal patch, and it has resulted in some
nice speedups when doing two-pass renders (in scenes that contain
reflections etc). :)

George

> But actually these images were just a test to see if radiosity takes
reflection
> and ior into account, which it does :)
>

Post a reply to this message

This is IMHO :) I do appreciate all the work of the POV-Ray developers and
community, of course!

George



"George Pantazopoulos" <the### [at] attbicom*KILLSPAM*> wrote in message
news:3e0fb3f9$1@news.povray.org...
>     This is where I disagree with POV-Ray's radiosity implementation.
> Radiosity is for simulating diffuse interreflection. Simulating specular
> interactions during the process of shooting radiosity rays is a waste and
is
> better done using POV-Ray's excellent photon mapping. Fortunately this was
> easy to fix. I did it in my own personal patch, and it has resulted in
some
> nice speedups when doing two-pass renders (in scenes that contain
> reflections etc). :)
>
> George
>
> > But actually these images were just a test to see if radiosity takes
> reflection
> > and ior into account, which it does :)
> >
>
>
>

Post a reply to this message

Could you show some example pictures so wel can compare the results of the
regular 3.5 images with those of your patch? IMHO there is a very thin line
between diffuse interreflection and "normal" interreflection.

Post a reply to this message

Here is a more formal test I did a couple of months ago (before school got
really hectic :) )

George


http://home.attbi.com/~themaxx/povray/xrtest.html

"Apache" <apa### [at] yahoocom> wrote in message
news:3e0fb65b$1@news.povray.org...
> Could you show some example pictures so wel can compare the results of the
> regular 3.5 images with those of your patch? IMHO there is a very thin
line
> between diffuse interreflection and "normal" interreflection.
>
>

Post a reply to this message

"George Pantazopoulos" <the### [at] attbicom*KILLSPAM*> schreef in bericht
news:3e0fb3f9$1@news.povray.org...
>     This is where I disagree with POV-Ray's radiosity implementation.
> Radiosity is for simulating diffuse interreflection. Simulating specular
> interactions during the process of shooting radiosity rays is a waste and is
> better done using POV-Ray's excellent photon mapping. Fortunately this was
> easy to fix. I did it in my own personal patch, and it has resulted in some
> nice speedups when doing two-pass renders (in scenes that contain
> reflections etc). :)
>
> George

Hmm, I disagree actually.  For example: take a room with a light source (a
bright object, it won't work with a real light_source) and a mirror.  An object
in that room should have 2 shadows: one from the real light source, and one from
the virtual light source, the reflection of the real light source.  Using
radiosity as it is now, you can simulate this behaviour, but this isn't possible
in any way if radiosity doesn't take reflections into account.

I realise it would be mostly unnoticable effects, but I believe it's those
effects that makes a difference between a photorealistic rendering and an
obvious synthetic image.

cu!
--
camera{location-z*3}#macro G(b,e)b+(e-b)*(C/50)#end#macro L(b,e,k,l)#local C=0
;#while(C<50)sphere{G(b,e),.1pigment{rgb G(k,l)}finish{ambient 1}}#local C=C+1
;#end#end L(y-x,y,x,x+y)L(y,-x-y,x+y,y)L(-x-y,-y,y,y+z)L(-y,y,y+z,x+y)L(0,x+y,
<.5,1,.5>,x)L(0,x-y,<.5,1,.5>,x)               // ZK http://www.povplace.be.tf

Post a reply to this message

:-)

I just started a "faked photons" radiosity scene....

Post a reply to this message

Actually, these two situations solve similar but different parts of the
global illumination problem.  Photons in POV only solve light transport from
light_source objects to the first diffuse surface.  POV's photons stop when
they hit a diffuse surface.  POV's radiosity, on the other hand, tracks
light transport from diffuse (or ambient) surfaces to other diffuse
surfaces.  Along the way, the light might be reflected or refracted.
Basically, POV's radiosity picks up where photons and direct lighting stop.

Because of this, using photons and HRDI together with the suggested method
(save_file and load_file) will provide an inaccurate (though probably
good-looking) render, because you would be using both photons and radiosity
to solve the same part of the light transport model.

-Nathan


"George Pantazopoulos" <the### [at] attbicom*KILLSPAM*> wrote...
>     This is where I disagree with POV-Ray's radiosity implementation.
> Radiosity is for simulating diffuse interreflection. Simulating specular
> interactions during the process of shooting radiosity rays is a waste and
is
> better done using POV-Ray's excellent photon mapping. Fortunately this was
> easy to fix. I did it in my own personal patch, and it has resulted in
some
> nice speedups when doing two-pass renders (in scenes that contain
> reflections etc). :)
>
> George
>
> > But actually these images were just a test to see if radiosity takes
> reflection
> > and ior into account, which it does :)
> >
>
>
>

Post a reply to this message