Last one for now. Looks like the media-filled mesh object reacts to its
environment (the hair object) in a strange way, but the places where there
are no artifacts are pretty decent and the render time is acceptable. Other
tests are necessary of course. Possibly this could be useful on Poser
characters seen from afar.

Here the media used. The character was later scaled by 2.58.
media{
    scattering {1, rgb 600*<253,202,177>/255 extinction 1}
    intervals 1
    samples 50
    method 3
}


G.

--

**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

Post a reply to this message

Attachments:
Download 'testtrans5.jpg' (26 KB)

Preview of image 'testtrans5.jpg'

Getting closer. Have you tried your material with *solid* objects (not
meshes), just to see if it works the same way.
BTW it looks promising.

--
Jonathan.

"Gilles Tran" <tra### [at] inapginrafr> ha scritto nel messaggio
news:3bcb486a@news.povray.org...
> Last one for now. Looks like the media-filled mesh object reacts to its
> environment (the hair object) in a strange way, but the places where there
> are no artifacts are pretty decent and the render time is acceptable.
Other
> tests are necessary of course. Possibly this could be useful on Poser
> characters seen from afar.
>
> Here the media used. The character was later scaled by 2.58.
> media{
>     scattering {1, rgb 600*<253,202,177>/255 extinction 1}
>     intervals 1
>     samples 50
>     method 3
> }
>
>
> G.
>
> --
>
> **********************
> http://www.oyonale.com
> **********************
> - Graphic experiments
> - POV-Ray and Poser computer images
> - Posters
>
>
>
>

Post a reply to this message

3bcb4a50@news.povray.org...
> Getting closer. Have you tried your material with *solid* objects (not
> meshes), just to see if it works the same way.
> BTW it looks promising.
>
Well basically it's Kari's milk glass so we know it's working all right at
least on simple primitives. I tested a box, a cylinder and a sphere without
problems but a blob gives me weird results. I've no experience in using
media in blobs though.
...
Later : after some further tests, it seems that increasing the intervals
improves things quite a bit (I suppose that must have been obvious for those
who know how media works...) but I still can't get rid of the wireframe
artifacts (see below).

G.
--

**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

Post a reply to this message

Attachments:
Download 'testtrans6.jpg' (24 KB)

Preview of image 'testtrans6.jpg'

While designing my candles, I made some tests without the actual surface
(i.e. using rgbt 1 and no specular). It helps to *tune* media settings. Do
you get those wireframe artifacts with a completely transparent surface too?

--
Jonathan.

"Gilles Tran" <tra### [at] inapginrafr> ha scritto nel messaggio
news:3bcb5374@news.povray.org...

> 3bcb4a50@news.povray.org...
> > Getting closer. Have you tried your material with *solid* objects (not
> > meshes), just to see if it works the same way.
> > BTW it looks promising.
> >
> Well basically it's Kari's milk glass so we know it's working all right at
> least on simple primitives. I tested a box, a cylinder and a sphere
without
> problems but a blob gives me weird results. I've no experience in using
> media in blobs though.
> ...
> Later : after some further tests, it seems that increasing the intervals
> improves things quite a bit (I suppose that must have been obvious for
those
> who know how media works...) but I still can't get rid of the wireframe
> artifacts (see below).
>
> G.
> --
>
> **********************
> http://www.oyonale.com
> **********************
> - Graphic experiments
> - POV-Ray and Poser computer images
> - Posters
>
>
>
>

Post a reply to this message

Actually, I was going to suggest that you swap intervals and samples. You
should also be able to make do with much less than 50. My candle uses 5.

Post a reply to this message

The idea of a "misty being" is a cool one; I just don't like seeing mouth parts
innards, especially if it looks like artifact of  your modelling routine.
That's why I didn't like the first one.

Post a reply to this message

3bcb56cd@news.povray.org...
> While designing my candles, I made some tests without the actual surface
> (i.e. using rgbt 1 and no specular). It helps to *tune* media settings. Do
> you get those wireframe artifacts with a completely transparent surface
too?
>
It's a good idea. And yes, the wireframe artifacts show up too with a
completely transparent surface. I guess that a close up on the part where
the triangles are small (like the lips and nose) would reveal them too.

G.

--

**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

Post a reply to this message

3bcb7bf1@news.povray.org...
> Actually, I was going to suggest that you swap intervals and samples. You
> should also be able to make do with much less than 50. My candle uses 5.
No, it doesn't work in that case. Lowering the samples number just creates
more visible and uglier artifacts. Increasing them (up to 200) doesn't fix
the wireframe artifacts either, and just makes the render crawl.
It was just a test anyway. Temporary conclusion is that using this technique
with meshes has serious limitations but that it can be very valuable in
certain circumstances.

G.


--

**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

>

Post a reply to this message

I am no Rhino expert, but wouldnt this mesh problem be solved with a smooth
command ?
That would add more triangles to the mesh but as i see the problem it would
help a lot.

Normally i would not recommend a product that costs US$ 975 (iirc) , but
there's a free trial version with 25 saves. I think 25 it's enough for
smoothing some poses and trying it in POV, ahn ?

Will

PS: The URL is www.rhino3d.com .


"Gilles Tran" <tra### [at] inapginrafr> escreveu na mensagem
news:3bcc5cfc$1@news.povray.org...

> 3bcb7bf1@news.povray.org...
> > Actually, I was going to suggest that you swap intervals and samples.
You
> > should also be able to make do with much less than 50. My candle uses 5.
> No, it doesn't work in that case. Lowering the samples number just creates
> more visible and uglier artifacts. Increasing them (up to 200) doesn't fix
> the wireframe artifacts either, and just makes the render crawl.
> It was just a test anyway. Temporary conclusion is that using this
technique
> with meshes has serious limitations but that it can be very valuable in
> certain circumstances.
>
> G.
>
>
> --
>
> **********************
> http://www.oyonale.com
> **********************
> - Graphic experiments
> - POV-Ray and Poser computer images
> - Posters
>
> >
>
>

Post a reply to this message

3bcd1581@news.povray.org...
> I am no Rhino expert, but wouldnt this mesh problem be solved with a
smooth
> command ?
> That would add more triangles to the mesh but as i see the problem it
would
> help a lot.

It seems a good solution. It's also feasible with the free version of Amapi
but I'd loose the uv mapping information because Amapi 4 doesn't recognise
it. Would the uv info be kept after the smoothing in Rhino (I haven't used
Rhino since 1998) ? If it's the case it's worth trying.

G.

--

**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters

>
> Normally i would not recommend a product that costs US$ 975 (iirc) , but
> there's a free trial version with 25 saves. I think 25 it's enough for
> smoothing some poses and trying it in POV, ahn ?
>
> Will
>
> PS: The URL is www.rhino3d.com .
>
>
> "Gilles Tran" <tra### [at] inapginrafr> escreveu na mensagem
> news:3bcc5cfc$1@news.povray.org...

> > 3bcb7bf1@news.povray.org...
> > > Actually, I was going to suggest that you swap intervals and samples.
> You
> > > should also be able to make do with much less than 50. My candle uses
5.
> > No, it doesn't work in that case. Lowering the samples number just
creates
> > more visible and uglier artifacts. Increasing them (up to 200) doesn't
fix
> > the wireframe artifacts either, and just makes the render crawl.
> > It was just a test anyway. Temporary conclusion is that using this
> technique
> > with meshes has serious limitations but that it can be very valuable in
> > certain circumstances.
> >
> > G.
> >
> >
> > --
> >
> > **********************
> > http://www.oyonale.com
> > **********************
> > - Graphic experiments
> > - POV-Ray and Poser computer images
> > - Posters
> >
> > >
> >
> >
>
>

Post a reply to this message