POV-Ray : Newsgroups : povray.general : Possible Bug: Vanishing Isosurface Server Time
25 Dec 2024 01:54:29 EST (-0500)
  Possible Bug: Vanishing Isosurface (Message 1 to 3 of 3)  
From: Geodesic
Subject: Possible Bug: Vanishing Isosurface
Date: 11 Jun 2006 11:25:00
Message: <web.448c35565dbaf033d43fc2de0@news.povray.org>
Hi,
Rendering a number of identical isosurfaces, I noticed that one of them was
'missing'.  The problem seems to be based on a number of things - the
placement of  the object, the camera position, and bizarrely, a rotation
about the axis of symmetry which should produce no noticeable difference.
Checking the messages thread, it claims that the maximum gradient found was
0, and also lists all intersection tests with the isosurface as failing.

The other isosurfaces appeared exactly as they should, even in areas that
overlap where the 'vanished' isosurface should be.

I'm using version 3.6.1, on Windows 2000

camera {

    //orthographic

    location <4,8,10>

    direction <0,0,8>

    right <16,0,0>

    up <0,12,0>

    look_at <0,0,0>

}

#declare PentaToroid=object{

    isosurface{

        function{

            pow(sqrt(pow(x,2) + pow(z,2)) - 3 + (pow(z,5) -
10*pow(z,3)*pow(x,2) + 5*z*pow(x,4)) / pow(sqrt(pow(x,2) + pow(z,2)),5),2)
+ pow(y - (pow(x,5) - 10*pow(x,3)*pow(z,2) + 5*x*pow(z,4)) /
pow(sqrt(pow(x,2) + pow(z,2)),5),2) - 0.04

        }

        contained_by{box{<4.2,1.2,4.2><-4.2,-1.2,-4.2>}}

        evaluate 0.64*270,1.25,0.7

        all_intersections

        accuracy 0.001

        texture{T_Wood1}

    }

}
object{PentaToroid translate y*5 rotate y*0 rotate x*63.4333 rotate y*285
texture{T_Wood32}}

I find the object reappears for angles <250 or >320.
Thanks


Post a reply to this message

From: Mike Williams
Subject: Re: Possible Bug: Vanishing Isosurface
Date: 11 Jun 2006 15:01:03
Message: <cB60$PAYeGjEFwQW@econym.demon.co.uk>
Wasn't it Geodesic who wrote:
>Hi,
>Rendering a number of identical isosurfaces, I noticed that one of them was
>'missing'.  The problem seems to be based on a number of things - the
>placement of  the object, the camera position, and bizarrely, a rotation
>about the axis of symmetry which should produce no noticeable difference.
>Checking the messages thread, it claims that the maximum gradient found was
>0, and also lists all intersection tests with the isosurface as failing.
>
>The other isosurfaces appeared exactly as they should, even in areas that
>overlap where the 'vanished' isosurface should be.
>
>I'm using version 3.6.1, on Windows 2000

Do you get a warning displayed in the messages pane? I get this:

 Shutdown Warning: Evaluate found a maximum gradient of 0.000 and an 
 average gradient of 0.000. The maximum gradient variation was 0.000.
 It is recommended to adjust the parameters of 'evaluate' to:
 First parameter less than 0.000
 Second parameter less than 1.000 and greater than 1.0
 Third parameter greater than 1.000 and less than 1.0

I never really understood "eval", but if you use "maxgradient 250"
you'll see the object. POVRay will still put a warning suggesting a
higher max_gradient, but that's because there are tiny regions where y
is large and x and z are close to zero where the gradient becomes very
large indeed.

-- 
Mike Williams
Gentleman of Leisure


Post a reply to this message

From: Geodesic
Subject: Re: Possible Bug: Vanishing Isosurface
Date: 12 Jun 2006 12:50:01
Message: <web.448d9a0cd595742c4524b9330@news.povray.org>
> Do you get a warning displayed in the messages pane? I get this:
>
>  Shutdown Warning: Evaluate found a maximum gradient of 0.000 and an
>  average gradient of 0.000. The maximum gradient variation was 0.000.
>  It is recommended to adjust the parameters of 'evaluate' to:
>  First parameter less than 0.000
>  Second parameter less than 1.000 and greater than 1.0
>  Third parameter greater than 1.000 and less than 1.0

Yeah, exactly that.

> I never really understood "eval", but if you use "maxgradient 250"
> you'll see the object. POVRay will still put a warning suggesting a
> higher max_gradient, but that's because there are tiny regions where y
> is large and x and z are close to zero where the gradient becomes very
> large indeed.

Hmm.  I changed to using eval because of speed issues with just using
maxgradient, but I'll try it.
Thanks


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.