Several combinations of threshold, strength
scaling and rotations seem to do this to my
blobs  :((

Can anyone explain what is going in here.
See code below.


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html


#version 3.1;

blob {
  threshold 0.01
    cylinder {
      -y, y, 1
      strength 1
      scale <2, 0.5, 1.0>
      rotate 20*z
    }
  sturm
  pigment { color rgb <0, 1, 1> }
  translate -(x + z)/2
}

blob {
  threshold 0.1
    sphere {
      <0, 0, 0>, 0.7
      strength 1
      scale <2.0, 0.5, 1.0>
      rotate 30*z
    }
  sturm
  pigment { color rgb <1, 1, 0> }
  translate z*3/2
}

background { color rgb <1, 1, 1>/2 }

light_source {
  <10, 8, -5>
  color rgb <1, 1, 1>
}

camera {
  location <2, 3, 2>
  look_at <0, 0, 0>
}

Post a reply to this message

Attachments:
Download 'brokenblob01.jpg' (8 KB)

Preview of image 'brokenblob01.jpg'

"Tor Olav Kristensen" <tto### [at] onlineno> wrote in message
news:39B4064A.5F6FC3E3@online.no...
|
| Several combinations of threshold, strength
| scaling and rotations seem to do this to my
| blobs  :((
|
| Can anyone explain what is going in here.

Can't explain it but if threshold for the cylinder is 0.075 and for the
sphere 0.21 they both seem okay at that point.
Changing the scale to 5 times larger (and applying *5 to camera location,
light position and the translations of sphere and cylinder too) doesn't fix
them.  I was trying to see if there were too small because that's a know
problem with the primitives if they get too small.  These aren't in the
range of size that would be a problem normally to begin with though.

Bob

Post a reply to this message

It has to do with the scale... try changing the offending scale to a
negative... i.e... <-x,y,z> vs <x,y,z> why this is i don't know... It happened
to me when using the <matrix> w/ blobs...?

Bob Hughes wrote:

> "Tor Olav Kristensen" <tto### [at] onlineno> wrote in message
> news:39B4064A.5F6FC3E3@online.no...
> |
> | Several combinations of threshold, strength
> | scaling and rotations seem to do this to my
> | blobs  :((
> |
> | Can anyone explain what is going in here.
>
> Can't explain it but if threshold for the cylinder is 0.075 and for the
> sphere 0.21 they both seem okay at that point.
> Changing the scale to 5 times larger (and applying *5 to camera location,
> light position and the translations of sphere and cylinder too) doesn't fix
> them.  I was trying to see if there were too small because that's a know
> problem with the primitives if they get too small.  These aren't in the
> range of size that would be a problem normally to begin with though.
>
> Bob

Post a reply to this message

Bob Hughes wrote:

> "Tor Olav Kristensen" <tto### [at] onlineno> wrote in message
> news:39B4064A.5F6FC3E3@online.no...
> |
> | Several combinations of threshold, strength
> | scaling and rotations seem to do this to my
> | blobs  :((
> |
> | Can anyone explain what is going in here.
>
> Can't explain it but if threshold for the cylinder is 0.075 and for the
> sphere 0.21 they both seem okay at that point.
> Changing the scale to 5 times larger (and applying *5 to camera location,
> light position and the translations of sphere and cylinder too) doesn't fix
> them.  I was trying to see if there were too small because that's a know
> problem with the primitives if they get too small.  These aren't in the
> range of size that would be a problem normally to begin with though.

Thank you for looking into the problem Bob.

I have some blob images that have "faults" in
them because of this (methinks).

And even if I could make the faults go away with
simpler blobs (by altering settings as you describe
above), I wouldn't know how to tweak the settings
for the more complicated blobs I made.

I too was afraid that I used to marginal values,
so therefore I tried to find more "normal" values
that also damaged the blobs.

I'm not sure yet if this problem qualifies as a bug.

Are there any TAG-people out there that can
cast some light on this problem  ?


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

SomeGuy wrote:

> It has to do with the scale... try changing the offending scale to a
> negative... i.e... <-x,y,z> vs <x,y,z> why this is i don't know... It happened
> to me when using the <matrix> w/ blobs...?

Thank you for this information.

Do you have any examples of blobs that was destroyed
and that could be fixed that way?


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

In article <39B4440C.35D5F8BB@online.no>, Tor Olav Kristensen 
<tto### [at] onlineno> wrote:

> I'm not sure yet if this problem qualifies as a bug.

It might simply be a result of the rendering method..."fixing" it might 
be either very difficult or result in a rendering slow down.
You might be able to get around the problem by using an isosurface, just 
use the blob pattern as the isosurface function. You might even get a 
speed up...and just for fun, apply some warps to the pigment function. 
:-)
I have found that turbulence makes a nicely bumpy corroded-metal surface.


> Are there any TAG-people out there that can
> cast some light on this problem  ?

Hmm, I suppose I should come up with some witty remark about casting 
light in a raytracer, maybe including photons, but I can't think of 
anything right now. :-)

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message

On Mon, 04 Sep 2000 22:30:02 +0200, Tor Olav Kristensen
<tto### [at] onlineno> wrote:

>
>Several combinations of threshold, strength
>scaling and rotations seem to do this to my
>blobs  :((
>
>Can anyone explain what is going in here.
>See code below.

Looks like a bug in the internal bounding hierarchy for the blob
objects, as disabling it seems to fix the problem.

I hope this gives enough insight to those who know where to look for
the bug in the source code.


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] usanet
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message

Peter Popov wrote:

> On Mon, 04 Sep 2000 22:30:02 +0200, Tor Olav Kristensen
> <tto### [at] onlineno> wrote:
>
> >
> >Several combinations of threshold, strength
> >scaling and rotations seem to do this to my
> >blobs  :((
> >
> >Can anyone explain what is going in here.
> >See code below.
>
> Looks like a bug in the internal bounding hierarchy for the blob
> objects, as disabling it seems to fix the problem.
>
> I hope this gives enough insight to those who know where to look for
> the bug in the source code.

Thank you Peter.

Are you saying that it is possible to tell POV-Ray
(without recompiling) to disable internal bounding ?

And that the defects disappears when you're doing this ?

If so, then I'm a happy man again !


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

Peter Popov wrote:
> 
> On Mon, 04 Sep 2000 22:30:02 +0200, Tor Olav Kristensen
> <tto### [at] onlineno> wrote:
> 
> >
> >Several combinations of threshold, strength
> >scaling and rotations seem to do this to my
> >blobs  :((
> >
> >Can anyone explain what is going in here.
> >See code below.
> 
> Looks like a bug in the internal bounding hierarchy for the blob
> objects, as disabling it seems to fix the problem.
> 

Agreed, at least for the bug coming from the bounding hierarchy.
I did not try to disable it, I find the bug instead.

> I hope this gives enough insight to those who know where to look for
> the bug in the source code.
> 

Bug found, and a correction has now been proposed in povray.general

(and I even made a web page in french about it, but you'd better be patient 
or have a T1 connection (lot of pictures))
http://grimbert.cjb.net/pov/correct/blob.html

Have a nice day!

Post a reply to this message

On Fri, 08 Sep 2000 02:56:59 +0200, Tor Olav Kristensen
<tto### [at] onlineno> wrote:

>Thank you Peter.
>
>Are you saying that it is possible to tell POV-Ray
>(without recompiling) to disable internal bounding ?

hierarchy off

Does that help? Oh, and BTW it's in the docs :))

>And that the defects disappears when you're doing this ?

Yup.

>If so, then I'm a happy man again !

Glad to hear that.


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] usanet
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message