Stephen wrote:

> The problem with making CSG or mesh models transparent, is the internal 
> parts show when it is partly transparent.

For plain CSG, can't this usually be avoided by combining
elements with "merge"? Of course I suppose this doesn't help for
meshes if the original mesh already has "internal" triangles.

Post a reply to this message

Christian Froeschlin <chr### [at] chrfrde> wrote:
> Stephen wrote:
>
> > The problem with making CSG or mesh models transparent, is the internal
> > parts show when it is partly transparent.
>
> For plain CSG, can't this usually be avoided by combining
> elements with "merge"? Of course I suppose this doesn't help for
> meshes if the original mesh already has "internal" triangles.

I have studied Bill's Code and all the hints he gave with his IRTC entry 2006
last year very carefully. So it is no wonder that I remember it quite well.
Fortunatelly Bill has posted the complete code of his entry and gave a lot of
hints. Many thanks for that to Bill, even after all the years!

IMO his main and wonderful idea is to create an eggshell from a figure and fill
this with a proper media. Having CSG objects every deviation from convexity is a
problem since you cannot simply scale down the figure a small amount to craft
thsi eggshell. This would only work in very convenient situations. With a mesh
you can move the vertices along the normals (Blender seems to able to do this,
but I used Wings for that) and you have an inner object to subtract from the
outer to gain this eggshell soon. Problems arise from vertices which are under
the surface in the original object (joints) or from holes. I remember well
having dived into Poser Jennies nose to close the holes there. Than I learned
about MeshLab and the problem of holes was fixed. So I hadn't to repair the hole
of "the mouth I don't speak with" to cite Till Eulenspiegel. Thanks to Thomas
for mentioning the works of Charles de Coster in another thread here.
I was unaware of his wonderful interpretation of the stuff so far.

But for the picture under discussion here: Really great work. And as ever, a
very good look for details.

Best regards,
Michael

Post a reply to this message

Le 17/09/2013 15:00, Ive a écrit :
> this means almost no time for POV-Ray.

This is good: small time to model many long to render beautiful scenes.
(or even an animation)

It's just: almost no time in front of computer, not for POV-Ray.

Post a reply to this message

> But I know, POV-Ray purists would consider this as cheating.

   I don't know if there is still any around here... ;)

   BTW, great work as usual. I hope to also make a comeback later this
month... my POV skills are starting to get rusty after almost a year
without poving.

--
Jaime

Post a reply to this message

Christian Froeschlin <chr### [at] chrfrde> wrote:
> Stephen wrote:
>
> > The problem with making CSG or mesh models transparent, is the internal
> > parts show when it is partly transparent.
>
> For plain CSG, can't this usually be avoided by combining
> elements with "merge"? Of course I suppose this doesn't help for
> meshes if the original mesh already has "internal" triangles.

There is more to it than that, Christian.
Imagine a simple box materialising. Do we want to see the hidden faces as it
goes from transparent to opaque? With a model of a police box all the internal
and hidden detail would show. If a person went from invisible to translucent to
visible, would you see their body under their clothes and what about all the
internal organs?
Now with a Tardis how do we see that the inside is larger than the outside?

100 Karma points to the person who can do a realistic (WTF) materialisation or
de-materialisation of a complex model. (With shadows and you get 500 Karma
points.)

Stephen

Post a reply to this message

On 18-9-2013 12:41, Stephen wrote:

> 100 Karma points to the person who can do a realistic (WTF) materialisation or
> de-materialisation of a complex model. (With shadows and you get 500 Karma
> points.)

/That/ would be an interesting TC-RTC Challenge indeed.

Thomas

Post a reply to this message

A few loose thoughts about this, just to get my ideas clear.

On 18-9-2013 12:41, Stephen wrote:
> There is more to it than that, Christian.
> Imagine a simple box materialising. Do we want to see the hidden faces as it
> goes from transparent to opaque?

No. It would be the /shape/ of the box, the outline if you want, 
becoming transparent.

> With a model of a police box all the internal
> and hidden detail would show.

It shouldn't. While becoming transparent the inside remains hidden from 
our eyes.

> If a person went from invisible to translucent to
> visible, would you see their body under their clothes and what about all the
> internal organs?

Again, invisible to the onlooker. Only the outward shape of the person 
is involved.

> Now with a Tardis how do we see that the inside is larger than the outside?

An interesting challenge would be to show both (open doors). I did 
something similar a very long time ago (>10 years) with a door to 
another dimension. I cheated somewhat: The door opened into space 
(galaxy.inc) and was made of a vertical plane mirror (reflecting space) 
with a differenced door. Behind the door was a landscape cut off by the 
vertical plane. I must have the code somewhere...

Thomas

Post a reply to this message

At least I found the image. Made in 2000.

Thomas

Post a reply to this message

Attachments:
Download 'portedepandore2.png' (474 KB)

Preview of image 'portedepandore2.png'

Thomas de Groot <tho### [at] degrootorg> wrote:
> A few loose thoughts about this, just to get my ideas clear.
>
> On 18-9-2013 12:41, Stephen wrote:
> > There is more to it than that, Christian.
> > Imagine a simple box materialising. Do we want to see the hidden faces as it
> > goes from transparent to opaque?
>
> No. It would be the /shape/ of the box, the outline if you want,
> becoming transparent.
>

That is what I think but we have grown up with special effects that show that
this is the way to do it.

> > With a model of a police box all the internal
> > and hidden detail would show.
>
> It shouldn't. While becoming transparent the inside remains hidden from
> our eyes.
>

It depends how you look at it, ;-)
If you change the transparency in Pov. You start seeing the internal structure.

> > If a person went from invisible to translucent to
> > visible, would you see their body under their clothes and what about all the
> > internal organs?
>
> Again, invisible to the onlooker. Only the outward shape of the person
> is involved.
>

But more interesting and medically valuable. :-)

> > Now with a Tardis how do we see that the inside is larger than the outside?
>
> An interesting challenge would be to show both (open doors). I did
> something similar a very long time ago (>10 years) with a door to
> another dimension. I cheated somewhat: The door opened into space
> (galaxy.inc) and was made of a vertical plane mirror (reflecting space)
> with a differenced door. Behind the door was a landscape cut off by the
> vertical plane. I must have the code somewhere...
>

I thought of using a df3 that depicted an internal scene. But then got
distracted with trying to use a proximity macro. Needless to say neither got
finished.

Stephen

Post a reply to this message

Jaime Vives Piqueres <jai### [at] ignoranciaorg> wrote:
>    BTW, great work as usual. I hope to also make a comeback later this
> month... my POV skills are starting to get rusty after almost a year
> without poving.

I'd love to see that.  Your rust will surely be better than mine.

Post a reply to this message

Attachments:
Download 'rusty_ball.jpg' (109 KB)

Preview of image 'rusty_ball.jpg'