that would be great i would like to see the texture details... =]

Tor Olav Kristensen wrote:

> Gleb wrote:
> >
> > Looks nice.
>
> Thanks.
>
> > Will it form a whole sphere when combined
> >
> > with the y-rotated other textured one?
>
> No. I'm sorry if the title of my prior
> posts have confused you people.
>
> This shape uses another PhiFn() than the one
> I used for the shape in my "The remains of an
> iso sphere: Take 2" image.
>
> I think that I'll soon post some source code
> for this image to p.t.s-f.
>
> Tor Olav

Post a reply to this message

Tor Olav Kristensen wrote:
>...
> (Because there seems to be some
> calculations that are not calculated
> correctly by POV-Ray. Increasing
> max_gradient or accuracy did not help
> at all.)
>...

Attached are two images that shows
the defect in the shape.

IIRC Riemann16_s1.jpg was rendered with the
max_gradient set to 4 and for Riemann16_s2.jpg
it was set to 250.

(The images was rendered at different
resolutions and with different light settings.)

I find the "tongue" that appears in the _s2
image quite mysterious.


Tor Olav

Post a reply to this message

Attachments:
Download 'riemann16_s1.jpg' (13 KB) Download 'riemann16_s2.jpg' (5 KB)

Preview of image 'riemann16_s1.jpg'

Preview of image 'riemann16_s2.jpg'

Christoph Hormann wrote:
>...
> If that surface structure is a normal you could make it part of the
> isosurface geometry.  Apart from that, do you have 'normal on' in
> radiosity?

I didn't think of that. Now I have started experimenting
with that. Thank you for the suggestion.

And yes, I had the normal on in radiosity.

>...
> > I should probably mention that I had
> > to "hand correct" the finished image
> > somewhat at the sphere's "south pole".
> >
> > (Because there seems to be some
> > calculations that are not calculated
> > correctly by POV-Ray. Increasing
> > max_gradient or accuracy did not help
> > at all.)
> 
> I think it's quite obvious that spiral structures lead to infinite
> gradients which can't be rendered correctly under all circumstances.

Yes that was my first thought too.

But as this shape is symmetric around both
its x and z axes (as Anton Sherwood mentioned
too), there should be infinite gradients at
the opposite pole as well.

But I have not experienced any problems at
that pole yet.


Tor Olav

Post a reply to this message

Anton Sherwood wrote:
> 
> Tor Olav:
> > > (Because there seems to be some
> > > calculations that are not calculated
> > > correctly by POV-Ray. Increasing
> > > max_gradient or accuracy did not help
> > > at all.)
> 
> Slime:
> > If you defined the spiral correctly, that's because the spiral
> > never ends, the detail increases towards the poles infinitely.
> > That is, if I understood the description of the shape that I
> > read a while ago correctly.
> 
> In an equiangular spiral (reverse Mercator projection of oblique
> parallels), that would happen at both poles; TOK said only one.

I don't remember what reverse Mercator projection is,
so I don't quite get what you are saying here.

But I guess that you are thinking the same as me:
That the problem should surface at the other pole as
well, because the gradients should be symmetric too.

 
> What's the formula?

I have posted the source to povray.text.scene-files now:

http://news.povray.org/povray.text.scene-files/21784/
news://news.povray.org/3C549C14.A9009E4%40online.no


> (Love the textures.)

I just modified one that Gilles Tran once made.
"Radiosity test #2", Aug 2000:

http://news.povray.org/search/?s=Radiosity+test+%232&b=Search


Tor Olav

Post a reply to this message

"R. Suzuki" wrote:
> 
> "Tor Olav Kristensen" wrote in message
> > I'm not sure if this image is finished
> > or not now.
> >
> > What do you think ?
> 
> Nice image.

Thank you.


> > I should probably mention that I had
> > to "hand correct" the finished image
> > somewhat at the sphere's "south pole".
> >
> > (Because there seems to be some
> > calculations that are not calculated
> > correctly by POV-Ray. Increasing
> > max_gradient or accuracy did not help
> > at all.)
> 
> I tried to render your scene file posted in p.t.s-f
> last month.  But I could not find any problem
> on the image.  Did you use the same function?


You are right: The last scene file did not
produce any errors in the image.

For this image a used a different function
for the rotation. See my recent post to
povray.text.scene-files for the source code:

http://news.povray.org/povray.text.scene-files/21784/
news://news.povray.org/3C549C14.A9009E4%40online.no


Tor Olav

Post a reply to this message

"Tor Olav Kristensen" <tor### [at] onlineno> wrote in message
news:3C54A942.617AD2B5@online.no...
[snip]
> Attached are two images that shows
> the defect in the shape.
>
> I find the "tongue" that appears in the _s2
> image quite mysterious.
>
>
> Tor Olav

Perhaps this defect is only on the inner surfaces so the upper pole
looks OK from outside?

Alf

Post a reply to this message

Tor Olav Kristensen wrote:
> 
> But as this shape is symmetric around both
> its x and z axes (as Anton Sherwood mentioned
> too), there should be infinite gradients at
> the opposite pole as well.
> 
> But I have not experienced any problems at
> that pole yet.

Well, there's no guarantee that insufficient max_gradient leads to
artefacts.  

To avoid such problems you could cut off the spirals at some distance to
the poles, it would probably not be very visible and should solve the
problem.

Christoph 

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Alf Peake wrote:
> 
> "Tor Olav Kristensen" <tor### [at] onlineno> wrote in message
> news:3C54A942.617AD2B5@online.no...
> [snip]
> > Attached are two images that shows
> > the defect in the shape.
> >
> > I find the "tongue" that appears in the _s2
> > image quite mysterious.
> >
> >
> > Tor Olav
> 
> Perhaps this defect is only on the inner surfaces so the upper pole
> looks OK from outside?

Alf, the defect is visible on the outside too.

I have actually turned the shape upside down
so that the defect should be less visible.
(I.e. in the shadows inside.)


Tor Olav

Post a reply to this message

"Tor Olav Kristensen" <tor### [at] hotmailcom> wrote in message
news:3C565CA8.BF44B3C6@hotmail.com...
>
> Alf Peake wrote:
> >
> > Perhaps this defect is only on the inner surfaces so the upper
pole
> > looks OK from outside?
>
> Alf, the defect is visible on the outside too.
>

Sorry. My error. I understood you only had a problem with the one
pole.

Alf

Post a reply to this message

> Anton Sherwood wrote:
> > In an equiangular spiral (reverse Mercator projection of oblique
> > parallels) . . .

Tor Olav Kristensen wrote:
> I don't remember what reverse Mercator projection is,

I meant inverse: from the plane back to the sphere.


-- 
Anton Sherwood, http://www.ogre.nu/blog/blog.htm

Post a reply to this message