I hope that this hasn't been brought up before, I've checked the known
bugs and didn't see it. I get a very strange result when I render the
following scene in Pov3.5:

camera {
  location  <0.0, 2, -4>
  look_at   <0.0, 0.0,  0.0>
}
	
light_source { <-30, 30, -30>
  color rgb 1
}
	
plane { y, -.001
	pigment {
		checker
		color rgb 1
		color rgb .2
	}
}
	
cylinder { 0, y, 1 
  texture { pigment { color rgbt <1,1,1,.8> } }
  scale <.5,1,.5>
}

The critical value seems to be the -.001 with the plane.  If I change it
to, for example -.00100001 everything works as it should.

	~Kevin Wampler~

Post a reply to this message

Kevin Wampler wrote:
> 
> I hope that this hasn't been brought up before, I've checked the known
> bugs and didn't see it. I get a very strange result when I render the
> following scene in Pov3.5:
> 
> [...]

It seems strange indeed, but nothing new, it already occurs with 3.1.

Christoph

-- 
POV-Ray tutorials, IsoWood include,                 
TransSkin and more: http://www.tu-bs.de/~y0013390/  
Last updated 15 Jul. 2002 _____./\/^>_*_<^\/\.______

Post a reply to this message

Kevin Wampler wrote:
>
> I get a very strange result when I render the
> following scene in Pov3.5:

I got stripes of noise on the cylinder. This produces noise
on the whole plane. AMD TB 1 GHz Win98 SE.

camera{}
        
light_source{5 rgb 1}

plane{y,-1.001 pigment{checker rgb 1 rgb .2} rotate -10}



_____________
Kari Kivisalo

Post a reply to this message

Confirmed here as well. W2000, 1GHz, 256Mb, POVWin 3.5
Try moving the camera about as well.
<0.0, 3, -4> gives very strange results

-tgq

Post a reply to this message

"Christoph Hormann" <chr### [at] gmxde> wrote in message
news:3D3BE060.B4CEC84F@gmx.de...
>
> Kevin Wampler wrote:
> >
> > I hope that this hasn't been brought up before, I've checked the known
> > bugs and didn't see it.
>
> It seems strange indeed, but nothing new, it already occurs with 3.1.

It was found before all right, not very long before 3.5 was final. I don't
recall if a bug report was made but it was something posted to the beta test
group I believe. Odd thing how it affects the entire surface like that, just
like coincident surfaces. Behaves as though copies of the cylinder, and
plane (as you can see using -1.001 like Kari used), exist.

Post a reply to this message

hughes b wrote:
> 
> "Christoph Hormann" <chr### [at] gmxde> wrote in message
> news:3D3BE060.B4CEC84F@gmx.de...
> >
> > Kevin Wampler wrote:
> > >
> > > I hope that this hasn't been brought up before, I've checked the known
> > > bugs and didn't see it.
> >
> > It seems strange indeed, but nothing new, it already occurs with 3.1.
> 
> It was found before all right, not very long before 3.5 was final. I don't
> recall if a bug report was made but it was something posted to the beta test
> group I believe. Odd thing how it affects the entire surface like that, just
> like coincident surfaces. Behaves as though copies of the cylinder, and
> plane (as you can see using -1.001 like Kari used), exist.

For me it does even more, through the sides (but not the bottom) of the
cylinder the colors of the checkered plane are reversed, so the black
parts look white and the white parts look black, see
http://www.u.arizona.edu/~wamplerk/misc/bugtest.jpg

Post a reply to this message

well, I looked into this a bit further, and it looks like it's that the
boundary between the block change in the checker pattern is at -.001 (I
assume it's here so that a checkered plane at 0 will look ok).  This
seems pretty obvious now, and I'm kicking myself for not having realized
it earlier.  I suppose it's pretty unavoidable, so I won't post to
bugreports.

Post a reply to this message

"Kevin Wampler" <wam### [at] uarizonaedu> wrote in message
news:3D3DC1D5.1D916CFC@u.arizona.edu...
> well, I looked into this a bit further, and it looks like it's that the
> boundary between the block change in the checker pattern is at -.001 (I
> assume it's here so that a checkered plane at 0 will look ok).  This
> seems pretty obvious now, and I'm kicking myself for not having realized
> it earlier.  I suppose it's pretty unavoidable, so I won't post to
> bugreports.

Oh. Makes sense. Forgot about that kind of thing, gradient pattern too.
Thing about it is though, why is the surface of an object not using the
pattern affected also? Acts like the pattern supersedes surfaces or
something.

Post a reply to this message

hughes b wrote:
> Oh. Makes sense. Forgot about that kind of thing, gradient pattern too.
> Thing about it is though, why is the surface of an object not using the
> pattern affected also? Acts like the pattern supersedes surfaces or
> something.

I don't think it does.  I think that when the ray gets traced through
the cylinder, it accumulates a bit of floating point error on the way,
and that this gives rise to odd effects when it intersects the plane,
and you can see these artifacts through the transparent cylinder.  If
you make the cylinder opaque, the noise on it disappears.

Post a reply to this message