 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi there,
The minimal scene file below should produce identical images for any
value of the scaling factor s, but it doesn't. s=1e-1 works fine, s=1e-3
produces individual spheres instead of a sweep, and s=1e-5 makes the
object fully invisible. The code for sphere_sweep doesn't seem to scale.
Using other values for the sphere_sweep tolerance and turning off the
bounding (-MB) doesn't solve the problem.
This might be related to http://bugs.povray.org/task/81, as I'm getting
the artifacts described there as well for intermediate scaling factors,
but the problems described here appear to be much more severe.
I'm using POV-Ray 3.7 on OpenSUSE 12.2.
Regards,
Roman
// START MINIMAL SCENE FILE
#version 3.7;
#declare s = 1e-3; // 1e-1: ok, 1e-3: individual spheres, 1e-5: invisible
camera {
location 7*z
look_at 0
scale s
}
light_source {
<20,25,40>, rgb 1
scale s
}
sphere_sweep {
cubic_spline
5,
< 2.4,-2.1,0>, 1
< 0.8,-0.9,0>, 1
<-0.6, 0.4,0>, 1
<-1.8, 1.7,0>, 1
<-2.8, 2.0,0>, 1
scale s
pigment { color rgb 1 }
}
// END MINIMAL SCENE FILE
Post a reply to this message
Attachments:
Download 'sphere_sweep_problem_1e-1.png' (3 KB)
Download 'sphere_sweep_problem_1e-3.png' (3 KB)
Download 'sphere_sweep_problem_1e-5.png' (1 KB)
Preview of image 'sphere_sweep_problem_1e-1.png'
Preview of image 'sphere_sweep_problem_1e-3.png'
Preview of image 'sphere_sweep_problem_1e-5.png'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I wonder if you can /scale/ a camera? In the docs section about
transforming the camera, this is /not/ mentioned as such, only rotate
and translate, so maybe there lies the problem.
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Le 13/12/2013 11:55, Roman a écrit :
> Hi there,
>
> The minimal scene file below should produce identical images for any
> value of the scaling factor s, but it doesn't. s=1e-1 works fine, s=1e-3
> produces individual spheres instead of a sweep, and s=1e-5 makes the
> object fully invisible. The code for sphere_sweep doesn't seem to scale.
>
> Using other values for the sphere_sweep tolerance and turning off the
> bounding (-MB) doesn't solve the problem.
>
> This might be related to http://bugs.povray.org/task/81, as I'm getting
> the artifacts described there as well for intermediate scaling factors,
> but the problems described here appear to be much more severe.
>
> I'm using POV-Ray 3.7 on OpenSUSE 12.2.
>
It is not related to 81, but 297
http://bugs.povray.org/task/297
(81 is about artefact (noise) in solver, whereas 297 is about epsilon)
A lot of shapes would not scale either. That's 297.
--
Just because nobody complains does not mean all parachutes are perfect.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 2013-12-13 13:50, Le_Forgeron wrote:
> It is not related to 81, but 297
>
> http://bugs.povray.org/task/297
>
> (81 is about artefact (noise) in solver, whereas 297 is about epsilon)
>
> A lot of shapes would not scale either. That's 297.
Thank you for clarifying this. To work around the problem, I'm currently
scaling my whole scene to a size that works for all objects. That's a
bit cumbersome though, and I'm worried that such a common error-free
scale won't exist in all cases. IMHO this is a serious limitation of
POV-Ray that should be fixed, if at all possible.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thomas de Groot <tho### [at] degroot org> wrote:
> I wonder if you can /scale/ a camera? In the docs section about
> transforming the camera, this is /not/ mentioned as such, only rotate
> and translate, so maybe there lies the problem.
>
> Thomas
Roman has discovered a new version of the word angle, using scale is the same
that using angle, I have just tried it with somme blob.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Roman <vet### [at] ethzch> wrote:
> On 2013-12-13 13:50, Le_Forgeron wrote:
>
> > It is not related to 81, but 297
> >
> > http://bugs.povray.org/task/297
> >
> > (81 is about artefact (noise) in solver, whereas 297 is about epsilon)
> >
> > A lot of shapes would not scale either. That's 297.
>
> Thank you for clarifying this. To work around the problem, I'm currently
> scaling my whole scene to a size that works for all objects. That's a
> bit cumbersome though, and I'm worried that such a common error-free
> scale won't exist in all cases. IMHO this is a serious limitation of
> POV-Ray that should be fixed, if at all possible.
I'm trying with low scale on sphere_sweep and it appears discontinuities when
approching a scale value of 0.005. Scaling the camera has no effect on the
sphere_sweep, it just act like angle.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Le_Forgeron <lef### [at] freefr> wrote:
> Le 13/12/2013 11:55, Roman a écrit :
> > Hi there,
> >
> > The minimal scene file below should produce identical images for any
> > value of the scaling factor s, but it doesn't. s=1e-1 works fine, s=1e-3
> > produces individual spheres instead of a sweep, and s=1e-5 makes the
> > object fully invisible. The code for sphere_sweep doesn't seem to scale.
> >
> > Using other values for the sphere_sweep tolerance and turning off the
> > bounding (-MB) doesn't solve the problem.
> >
> > This might be related to http://bugs.povray.org/task/81, as I'm getting
> > the artifacts described there as well for intermediate scaling factors,
> > but the problems described here appear to be much more severe.
> >
> > I'm using POV-Ray 3.7 on OpenSUSE 12.2.
> >
>
> It is not related to 81, but 297
>
> http://bugs.povray.org/task/297
>
> (81 is about artefact (noise) in solver, whereas 297 is about epsilon)
>
> A lot of shapes would not scale either. That's 297.
>
> --
Me again! After some reading I think the problem is related to problem #243 for
pov3.7 and 92 for Pov3.6. A too little scale make artifacts. No solution with
SDL. I have tested the file *.pov from #243 with Pov3.6 and 3.7 and the two
differents artifacts appears.
I maybe will try with UberPov, but it is actually buzy in work.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 13.12.2013 21:23, schrieb Fractracer:
> Me again! After some reading I think the problem is related to problem #243 for
> pov3.7 and 92 for Pov3.6. A too little scale make artifacts. No solution with
> SDL. I have tested the file *.pov from #243 with Pov3.6 and 3.7 and the two
> differents artifacts appears.
> I maybe will try with UberPov, but it is actually buzy in work.
Don't bother; UberPOV is no different than official POV-Ray 3.7.0 when
it comes to sphere_sweep.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 13-12-2013 20:31, Fractracer wrote:
> Thomas de Groot <tho### [at] degrootorg> wrote:
>> I wonder if you can /scale/ a camera? In the docs section about
>> transforming the camera, this is /not/ mentioned as such, only rotate
>> and translate, so maybe there lies the problem.
>>
>> Thomas
>
> Roman has discovered a new version of the word angle, using scale is the same
> that using angle, I have just tried it with somme blob.
>
>
That is strange indeed. In my simple world view I would think that angle
and scale are two very different processes. Even if both give identical
results, I would like an explanation about how that comes to be.
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 2013-12-13 14:35, Roman wrote:
> Thank you for clarifying this. To work around the problem, I'm currently
> scaling my whole scene to a size that works for all objects. That's a
> bit cumbersome though, and I'm worried that such a common error-free
> scale won't exist in all cases. IMHO this is a serious limitation of
> POV-Ray that should be fixed, if at all possible.
I'm still struggling with this issue. So I've rescaled everything to a
size that works for sphere_sweep, but now some of my other objects look
wrong. More precisely, their materials look wrong. I have narrowed it
down to the minimal scene file below, which shows a sphere with a
non-zero filter value in the pigment. Different scales result in
different images. What gives? I'm confused.
-Roman
// START MINIMAL SCENE FILE
#version 3.7;
// 1e-2, 1e-3, 1e-4, 1e-5 all produce different images,
// 1e-6 even produces a parse error!
#declare s = 1e-6;
camera { location 3*s*z look_at 0 }
light_source { s*<10,5,20>, rgb 1 }
sphere { 0, s pigment { color rgb <1,0,0> filter 0.5 } }
// END MINIMAL SCENE FILE
Post a reply to this message
Attachments:
Download 'scale_problem_1e-2.png' (3 KB)
Download 'scale_problem_1e-3.png' (3 KB)
Download 'scale_problem_1e-4.png' (3 KB)
Download 'scale_problem_1e-5.png' (1 KB)
Preview of image 'scale_problem_1e-2.png'
Preview of image 'scale_problem_1e-3.png'
Preview of image 'scale_problem_1e-4.png'
Preview of image 'scale_problem_1e-5.png'
|
|
| |
| |
|
|
|
|
| |
|
|
