No, Margus is right, it's not possible to simulate it with current features,
because transmit and filter don't have the same lighting model as is
necessary to create translucent effects (for Johannes, translucency is like
the sort of effect you get from shinning light on rice paper - the shadows,
and also like candle wax - the object actually seems to have a density).
While I guess this could be simulated with media it would be very difficult
to get the right effect and would be slow to render...

What I propose is a finish statement translucency keyword...  The way the
calculation is made is by taking into account what translucency really is...
that is, non-directional diffuse reflection... (*gulp*).  Now because the
diffuse color of a surface depends on the angle between the surface normal
and the light, what you do to get translucency is ignore the surface normal
alignment... (if that makes ANY sense to ANYONE I'll be happy, because it's
hard to explain...).

Um, yeah, so there you go... also if you combined with ray distance (into
the object) with another bias value you could get a "thickness" for the
translucency... like, 0.4 would be glue, 0.8 would be wax etc etc

Anyway, just an idea...

--
Lance.


---
For the latest 3D Studio MAX plug-ins, images and much more, go to:
The Zone - http://come.to/the.zone

Post a reply to this message

Lance Birch heeft geschreven in bericht <371041bf.0@news.povray.org>...
.......................
>What I propose is a finish statement translucency keyword...  The way the
>calculation is made is by taking into account what translucency really
is...
>that is, non-directional diffuse reflection... (*gulp*).

Now you have me confused, don't you mean "non-directional diffuse
transmission plus absorbsion"?
To me translucent is between transparent and opaque. Transparent is
glassclear and has a transmission of at least 50%, opaque has a transmission
smaller than 5% (0%), translucent has a transmission between 5% and 50% and
is not glassclear. Light passes through and is scatterd and absorbed (milky
glass).

> Now because the
>diffuse color of a surface depends on the angle between the surface normal
>and the light, what you do to get translucency is ignore the surface normal
>alignment... (if that makes ANY sense to ANYONE I'll be happy, because it's
>hard to explain...).

ingo

--
Met dank aan de muze met het glazen oog.

Post a reply to this message

Um, sort of, except that it is not necessarily how much light it lets
through, or how much you can see through it...

Opacity and transparency are messures of the same thing... (not to be
confused with transmission of course)  For example, you can give a texture a
transmission of 50%, and no matter how thick the object is, a certain amount
of light (50%) will always go through it... give it 5%, the object can be 1
unit or 1000000 units thick and 5% of the light will still get through.
What translucency does is affect the amount of light that can get through
based on the ray depth and also changes the color of the object based on a
density bias multiplier, applying this to the normal based surface lighting.

The important thing to remember is that the translucency must be object
surface normal independant to get the right effect... (only later is it
applied to the surface normal dependant shader)...

So yes, you're right about the absorbtion but you can't call it transmission
because although it will affect the transmittion of the surface, it has to
be based on the distance (the absorbtion, as you said) AND also affect the
surface itself...

The reason it's called non-directional diffuse reflection is not because it
reflects anything, but because the light reflects of it...  The
non-directional part comes from the fact that the multiplier is
non-directional but it can still be called diffuse because it affects the
diffuse of the object...

Basically to give a REALLY good example of translucency, it's just like
wax... the thicker the sheet of wax (or candle) the less light that will be
able to pass through it... AND, apart from that, the density (or
translucency bias) also affects the amount of light that can pass through...
for example, the denser the wax the less light will get through it...

Did any one get that or did I just make it more confusing?

--
Lance.


---
For the latest 3D Studio MAX plug-ins, images and much more, go to:
The Zone - http://come.to/the.zone

Post a reply to this message

Lance Birch heeft geschreven:
>Opacity and transparency are messures of the same thing... (not to be
>confused with transmission of course)  For example, you can give a texture
a
>transmission of 50%, and no matter how thick the object is, a certain
amount
>of light (50%) will always go through it... give it 5%, the object can be 1
>unit or 1000000 units thick and 5% of the light will still get through.

>What translucency does is affect the amount of light that can get through
>based on the ray depth and also changes the color of the object based on a
>density bias multiplier, applying this to the normal based surface
lighting.


That's what I meant with translucent is not glass clear (wrong choise of
words).

>The important thing to remember is that the translucency must be object
>surface normal independant to get the right effect... (only later is it
>applied to the surface normal dependant shader)...

So, an interior statement instead of finish?
(media, Mie-scattering, fade_distance, I'll give it a try)


>Did any one get that or did I just make it more confusing?

All clear.

ingo
--
Met dank aan de muze met het glazen oog.

Post a reply to this message

Yep, a media should be able to do it, but what I'm suggesting is that by
building it into the texture statement (whether it be in finish or whatever)
and having a specific translucency function, it can be made to be VERY fast.

The same goes for fluorescence...

--
Lance.


---
For the latest 3D Studio MAX plug-ins, images and much more, go to:
The Zone - http://come.to/the.zone

Post a reply to this message

Lance Birch wrote:
> 
> Yep, a media should be able to do it, but what I'm suggesting is that by
> building it into the texture statement (whether it be in finish or whatever)
> and having a specific translucency function, it can be made to be VERY fast.
> 
> The same goes for fluorescence...
> 
> --
> Lance.

  By your very definition it is not a propety of the surface. Therefore
it should be an interior funtion of the solid object and not a surface
property like pigments, finishes, and textures. It is contrary to the
model and the way Pov should treat behaviour like this. If it is to be
implemented let's do it correctly or not at all.

-- 
Ken Tyler

mailto://tylereng@pacbell.net

Post a reply to this message

OK then, sorry, I haven't kept up with Media and Interior and stuff like
that... so I wasn't sure what to put it in... (the last thing I played with
was a halo, which has been totally outdated).

OK, I guess it should interior then... and fluorescence in the finish
statement.

--
Lance.


---
For the latest 3D Studio MAX plug-ins, images and much more, go to:
The Zone - http://come.to/the.zone

Post a reply to this message

You talked about depth in your description.  A texture has no depth.  That is
why we moved IOR and media into the interior statement.

Dale (POV-Team)

Lance Birch wrote:

> Yep, a media should be able to do it, but what I'm suggesting is that by
> building it into the texture statement (whether it be in finish or whatever)
> and having a specific translucency function, it can be made to be VERY fast.
>
> The same goes for fluorescence...
>
> --
> Lance.
>
> ---
> For the latest 3D Studio MAX plug-ins, images and much more, go to:
> The Zone - http://come.to/the.zone

Post a reply to this message

Yeah, sorry about that... I haven't kept up with interior and media etc etc
and I didn't know where it should go...

--
Lance.


---
For the latest 3D Studio MAX plug-ins, images and much more, go to:
The Zone - http://come.to/the.zone

Post a reply to this message

You talked of transmittance ('transmit'?) being 5% or 50% or 100%
regardless of thickness of the object having such. This doesn't appear
to be exactly true when considering different rgb values used. Well
known fact that 'filter 1' is 100% clear if used on 'rgb 1', and 100%
non transparent if 'filter 1' is used with 'rgb 0' instead.
However also 'transmit 0.5' of 'rgb 1' is clearer than if 'rgb 0' or so
it seems. So I'm not so sure of the true workings of these when
considering the "visible" appearances anyhow. I know it might have a lot
to do with the background object or texture seen through something like
this, but black has always obscured better than white whether filter or
transmit is used far as it looks to me.
Back to the translucent and fluorescent concepts. I do understand the
idea at least. A white wax candle stick is best for me to think of for
describing translucence. What is needed is the blurring of incoming
light and/or object textures as seen through or "into" it. I was
thinking of a highly diffusing normal in a texture which would
effectively scatter everything, and coupled with light fading. Of
course, I haven't checked this out yet.
Not saying I'm against your idea either, just so you know.
The fluorescence would be in need of a glow effect for sure if to be
done right at all (translucence actually glows too via outside light
sources). Emitting objects/textures I believe may have been assumed to
be covered by radiosity somewhat.
Well, I feel I'm just reiterating what's already been said.
All in all the answer probably is a media type thing all right, not
texture based. (Like I needed to write anything once again... heck)


Lance Birch wrote:
> 
> --
> Lance.
> 
> ---
> For the latest 3D Studio MAX plug-ins, images and much more, go to:
> The Zone - http://come.to/the.zone

-- 
 omniVERSE: beyond the universe
  http://members.aol.com/inversez/homepage.htm
 mailto:inv### [at] aolcom?Subject=PoV-News

Post a reply to this message