POV-Ray : Newsgroups : povray.bugreports : artifacts on lathe object Server Time
1 Dec 2021 03:42:18 EST (-0500)
  artifacts on lathe object (Message 6 to 15 of 15)  
<<< Previous 5 Messages Goto Initial 10 Messages
From: David Mandelberg
Subject: Re: artifacts on lathe object
Date: 10 Oct 2017 16:20:00
Message: <web.59dd2b10d40b44f2146834bc0@news.povray.org>
"omniverse" <omn### [at] charternet> wrote:
> "David Mandelberg" <dav### [at] mandelbergorg> wrote:
> > "Bald Eagle" <cre### [at] netscapenet> wrote:
> > > "David Mandelberg" <dav### [at] mandelbergorg> wrote:
> > > > When I render the below scene, there are some spurious black artifacts near
the
> > > > center of the red vuvuzela. Without the translation and rotation, the
artifacts
> > > > go away. I'm using POV-Ray 3.7.0 on Ubuntu 16.04.
> > > >
> > >
> > > I think there may have been some issues with the bezier spline, that have been
> > > addressed post-v-3.7
> > > I didn't have a chance to render this in 3.7.1-beta.8, but I'd guess it might
> > > look better.
> > > I'd say either install the beta and try it, or use a different spline type.
> >
> > There's still an issue in 3.7.1-beta.9. I'll try 3.8.0-alpha.9322209 next, but
> > it'll take me a bit of time to get to that.
>
> Think you'll find it's the same... except be aware v3.8 has changed default
> ambient so if you render with directive #version 3.7; it will look different
> from using 3.8 in its place.

You're right. 3.8.0-alpha.9322209 has the same issue.

>
> At first I thought the lathe object was thinner but its just the missing ambient
> in shadow.
>
> Nothing new about lathe problems, there was something about negative control
> points for bezier_spline which I'm guessing doesn't actually relate to this
> trouble.


Post a reply to this message

From: Bald Eagle
Subject: Re: artifacts on lathe object
Date: 10 Oct 2017 19:50:00
Message: <web.59dd5b8cd40b44f25cafe28e0@news.povray.org>
"David Mandelberg" <dav### [at] mandelbergorg> wrote:
> When I render the below scene, there are some spurious black artifacts near the
> center of the red vuvuzela. Without the translation and rotation, the artifacts
> go away.

I'm not sure how you got them to go away - they certainly don't for me.

I've tried removing the translation and rotation, and moving the camera, the
light sources, etc.  Gave the finish a little specular reflection - still there.
I used a cubic_spline instead - still there.

I suppose I ought to mention that when i was deveoping a recent scene with a
bell - I used a SOR, and I saw the same sort of artefact.

<snippet>
#declare bellSpline = array [12] {
     < 3.4688, 0.0001>,
     < 3.4688, 0.0002>,
     < 3.0938, 0.4375>,
     < 2.9375, 0.6250>,
     < 2.4844, 1.3438>,
     < 2.1719, 2.6094>,
     < 2.0469, 4.4688>,
     < 1.9063, 4.6875>,
     < 1.7500, 4.8750>,
     < 1.0313, 5.0938>,
     < 0.0001, 5.1563>,
     < 0.0001, 5.1564>
}

.....


#declare Bell1 =
union {

 cylinder {<0, 0, 0>, <0, (5.0938/2)+1, 0> 0.125}
 sphere {0, 0.25}
 torus {0.125, 0.125/4 rotate x*90 translate -y*0.25}
 object {Handle translate -y*1}
 sor {
     arraySize+1,
  #for (B, 0, arraySize)
  #local NewSpline = bellSpline [B]/2;
  //#debug concat ("B = ", str (B, 3, 1), "| Point = ", vstr (2, bellSpline [B],
",", 5, 1), "\n")
  #declare Xmax = max (Xmax, NewSpline.x);
  #declare Ymax = max (Ymax, NewSpline.y);
  NewSpline,
  #end
     open
 }
 //texture { T_Gold_1B }
 texture { mtex }
 //pigment {rgb <1, 1, 0>}
}

</snippet>

So there's something going on - just in general.


Post a reply to this message

From: omniverse
Subject: Re: artifacts on lathe object
Date: 10 Oct 2017 22:45:00
Message: <web.59dd84e0d40b44f29c5d6c810@news.povray.org>
"Bald Eagle" <cre### [at] netscapenet> wrote:
> "David Mandelberg" <dav### [at] mandelbergorg> wrote:
> > When I render the below scene, there are some spurious black artifacts near the
> > center of the red vuvuzela. Without the translation and rotation, the artifacts
> > go away.
>
> I'm not sure how you got them to go away - they certainly don't for me.
>
> I've tried removing the translation and rotation, and moving the camera, the
> light sources, etc.  Gave the finish a little specular reflection - still there.
> I used a cubic_spline instead - still there.
>
> I suppose I ought to mention that when i was deveoping a recent scene with a
> bell - I used a SOR, and I saw the same sort of artefact.

It's when the camera is perpendicular to sides (within vertical extent, ignoring
actual surface curvature) facing camera, so if it gets tilted away enough for
either the base or cap imaginary infinite reach outward to miss the camera it
doesn't show.

I was checking doing that; it can make a line of surface color outward into
space away from the surface, along with the colorless mess. The extent seems to
get confined to the maximum x point, so I don't know if that's a result of
bounding or what but I didn't see it go beyond sides of a box with same
dimensions. Or more precisely, kept within a cylinder of maximum x dimension.

Your bell again below, able to be added directly into lathe scene file and
rendered to show its trouble with this.


#declare bellSpline = array [12] {
     < 3.4688, 0.0001>,
     < 3.4688, 0.0002>,
     < 3.0938, 0.4375>,
     < 2.9375, 0.6250>,
     < 2.4844, 1.3438>,
     < 2.1719, 2.6094>,
     < 2.0469, 4.4688>,
     < 1.9063, 4.6875>,
     < 1.7500, 4.8750>,
     < 1.0313, 5.0938>,
     < 0.0001, 5.1563>,
     < 0.0001, 5.1564>
}

#declare arraySize=11;
#declare Xmax=3.4688;
#declare Ymax=5.1564;

//#declare Bell1 =
union {
 cylinder {<0, 0, 0>, <0, (5.0938/2)+1, 0> 0.125}
 sphere {0, 0.25}
 torus {0.125, 0.125/4 rotate x*90 translate -y*0.25}
// object {Handle translate -y*1}
 sor {
     arraySize+1,
  #for (B, 0, arraySize)
  #local NewSpline = bellSpline [B]/2;
  #declare Xmax = max (Xmax, NewSpline.x);
  #declare Ymax = max (Ymax, NewSpline.y);
  NewSpline,
  #end
     open
 }
 //texture { T_Gold_1B }
// texture { mtex }
 pigment {rgb <1, 1, 0>}
 rotate <-2,0,0>
}


Post a reply to this message

From: omniverse
Subject: Re: artifacts on lathe object
Date: 10 Oct 2017 23:05:00
Message: <web.59dd8954d40b44f29c5d6c810@news.povray.org>
"omniverse" <omn### [at] charternet> wrote:
Of course I notice after posting my last reply that simply adding keyword sturm
to the sor bell fixes it completely, at least from what I can see from the
example.

Unfortunate that the same can't be said for lathe, which doesn't seem affected
by sturm at all.


Post a reply to this message

From: clipka
Subject: Re: artifacts on lathe object
Date: 11 Oct 2017 04:45:54
Message: <59ddda42@news.povray.org>
Am 11.10.2017 um 04:41 schrieb omniverse:

> I was checking doing that; it can make a line of surface color outward into
> space away from the surface, along with the colorless mess. The extent seems to
> get confined to the maximum x point, so I don't know if that's a result of
> bounding or what but I didn't see it go beyond sides of a box with same
> dimensions. Or more precisely, kept within a cylinder of maximum x dimension.

The confinement you're seeing there is the lathe's internal bounding
mechanism kicking in, which does indeed use cylinders (instead of boxes)
to bound each individual segment.


Post a reply to this message

From: clipka
Subject: Re: artifacts on lathe object
Date: 11 Oct 2017 04:57:02
Message: <59dddcde@news.povray.org>
Am 09.10.2017 um 23:16 schrieb David Mandelberg:
> When I render the below scene, there are some spurious black artifacts near the
> center of the red vuvuzela. Without the translation and rotation, the artifacts
> go away. I'm using POV-Ray 3.7.0 on Ubuntu 16.04.

I wouldn't be surprised if these artefacts were related to a similar
issue with 3rd-order polynomials (e.g. bezier or cubic) in the sphere
sweep primitive.


Post a reply to this message

From: clipka
Subject: Re: artifacts on lathe object
Date: 11 Oct 2017 04:57:48
Message: <59dddd0c$1@news.povray.org>
Am 11.10.2017 um 05:00 schrieb omniverse:
> "omniverse" <omn### [at] charternet> wrote:
> Of course I notice after posting my last reply that simply adding keyword sturm
> to the sor bell fixes it completely, at least from what I can see from the
> example.
> 
> Unfortunate that the same can't be said for lathe, which doesn't seem affected
> by sturm at all.

It should be though, as it does pass the flag to the root solver.


Post a reply to this message

From: William F Pokorny
Subject: Re: artifacts on lathe object
Date: 16 Oct 2017 08:05:09
Message: <59e4a075$1@news.povray.org>
On 10/11/2017 04:57 AM, clipka wrote:
> Am 09.10.2017 um 23:16 schrieb David Mandelberg:
>> When I render the below scene, there are some spurious black artifacts near the
>> center of the red vuvuzela. Without the translation and rotation, the artifacts
>> go away. I'm using POV-Ray 3.7.0 on Ubuntu 16.04.
> 
> I wouldn't be surprised if these artefacts were related to a similar
> issue with 3rd-order polynomials (e.g. bezier or cubic) in the sphere
> sweep primitive.
> 

I too think this likely the case.

Christoph, do you want to update the sphere_sweep github issue :

https://github.com/POV-Ray/povray/issues/147

to include this lathe case?

Or, should we create a new github issue for it and reference #147 as 
likely related?

Bill P.


Post a reply to this message

From: clipka
Subject: Re: artifacts on lathe object
Date: 16 Oct 2017 10:05:01
Message: <web.59e4bc04d40b44f2160105a20@news.povray.org>
William F Pokorny <ano### [at] anonymousorg> wrote:
> On 10/11/2017 04:57 AM, clipka wrote:
> > Am 09.10.2017 um 23:16 schrieb David Mandelberg:
> >> When I render the below scene, there are some spurious black artifacts near the
> >> center of the red vuvuzela. Without the translation and rotation, the artifacts
> >> go away. I'm using POV-Ray 3.7.0 on Ubuntu 16.04.
> >
> > I wouldn't be surprised if these artefacts were related to a similar
> > issue with 3rd-order polynomials (e.g. bezier or cubic) in the sphere
> > sweep primitive.
> >
>
> I too think this likely the case.
>
> Christoph, do you want to update the sphere_sweep github issue :
>
> https://github.com/POV-Ray/povray/issues/147
>
> to include this lathe case?
>
> Or, should we create a new github issue for it and reference #147 as
> likely related?
>
> Bill P.

I think it would be cleaner to create a new issue. We don't know at this point
whether they are genuinely related -- and I'm pretty sure each needs its own
separate fix anyway.


Post a reply to this message

From: William F Pokorny
Subject: Re: artifacts on lathe object
Date: 17 Oct 2017 09:35:28
Message: <59e60720$1@news.povray.org>
On 10/16/2017 10:02 AM, clipka wrote:
> William F Pokorny <ano### [at] anonymousorg> wrote:
>...
> I think it would be cleaner to create a new issue. We don't know at this point
> whether they are genuinely related -- and I'm pretty sure each needs its own
> separate fix anyway.
> 
OK. See: https://github.com/POV-Ray/povray/issues/335


Post a reply to this message

<<< Previous 5 Messages Goto Initial 10 Messages

Copyright 2003-2021 Persistence of Vision Raytracer Pty. Ltd.