POV-Ray : Newsgroups : povray.binaries.images : sphere_sweep artefact image Server Time: 19 Oct 2019 10:12:42 GMT
  sphere_sweep artefact image (Message 1 to 10 of 11)  
Goto Latest 10 Messages Next 1 Messages >>>
From: Stephen
Subject: sphere_sweep artefact image
Date: 7 Mar 2018 13:23:19
Message: <5a9fe7c7@news.povray.org>
I’ve run into a problem using a large sphere_sweep. I get a horizontal, 
straight line artefact.
You can change the location of the artefact by raising or lowering 
either the camera location or look_at vector.
Happens in both ver 3.6 and 3.7
I posted the pov and ini files in povray.binaries.scene-files

-- 

Regards
     Stephen


Post a reply to this message


Attachments:
Download 'hilbert_curve_01a2_0000.png' (972 KB)

Preview of image 'hilbert_curve_01a2_0000.png'
hilbert_curve_01a2_0000.png


 

From: William F Pokorny
Subject: Re: sphere_sweep artefact image
Date: 7 Mar 2018 13:51:53
Message: <5a9fee79$1@news.povray.org>
On 03/07/2018 08:23 AM, Stephen wrote:
> 
> I’ve run into a problem using a large sphere_sweep. I get a horizontal, 
> straight line artefact.
> You can change the location of the artefact by raising or lowering 
> either the camera location or look_at vector.
> Happens in both ver 3.6 and 3.7
> I posted the pov and ini files in povray.binaries.scene-files
> 

Thanks for posting the scene and ini file. It looks like this is one of 
the sphere_sweep artifacts Christoph fixed in the updates for 
release/v3.8.0 and so too in the master (v3.8.1) branch.

Aside: this is the sort of artifact I've sometimes been able to 
ameliorate by using an odd number of pixels in a direction perpendicular 
to the artifact - trick works better at low resolutions. In this case an 
odd number of pixels vertically just changes to 3.7.0 artifact some...

Bill P.


Post a reply to this message

From: Stephen
Subject: Re: sphere_sweep artefact image
Date: 7 Mar 2018 15:00:33
Message: <5a9ffe91$1@news.povray.org>
On 07/03/2018 13:51, William F Pokorny wrote:
> On 03/07/2018 08:23 AM, Stephen wrote:
>>
>> I’ve run into a problem using a large sphere_sweep. I get a 
>> horizontal, straight line artefact.
>> You can change the location of the artefact by raising or lowering 
>> either the camera location or look_at vector.
>> Happens in both ver 3.6 and 3.7
>> I posted the pov and ini files in povray.binaries.scene-files
>>
> 
> Thanks for posting the scene and ini file. It looks like this is one of 
> the sphere_sweep artifacts Christoph fixed in the updates for 
> release/v3.8.0 and so too in the master (v3.8.1) branch.
> 
Thanks for the quick reply, Bill. I am pleased that it might be a known 
artefact. It looks like the artefact we used to get with SORs when the 
view crossed the horizon.
I'll wait until

> Aside: this is the sort of artifact I've sometimes been able to 
> ameliorate by using an odd number of pixels in a direction perpendicular 
> to the artifact - trick works better at low resolutions. In this case an 
> odd number of pixels vertically just changes to 3.7.0 artifact some...
> 

Maybe I should have said I was using ver 3.7.0. (I've yet to update my 
machine since I had a HDD crash.)

Aside 1
An odd number of pixels doesn't go well with some video codecs.


Aside 2
I was just interested in the 3d shape of the Hilbert curve.
When Maurice and his Blender team fix the Pov file importer. I will 
upload an obj file, if anyone is interested. Or if anyone is still using 
Bishop3D, a *.bsp one . I managed to import it into that. Oops! that's 
the exported file I uploaded but B3D and import it.
It is much easer for me to inspect it with an OpenGl display.

-- 

Regards
     Stephen


Post a reply to this message

From: Stephen
Subject: Re: sphere_sweep artefact image
Date: 7 Mar 2018 15:55:12
Message: <5aa00b60@news.povray.org>
On 07/03/2018 15:00, Stephen wrote:
>> Thanks for posting the scene and ini file. It looks like this is one 
>> of the sphere_sweep artifacts Christoph fixed in the updates for 
>> release/v3.8.0 and so too in the master (v3.8.1) branch.
>>
> Thanks for the quick reply, Bill. I am pleased that it might be a known 
> artefact. It looks like the artefact we used to get with SORs when the 
> view crossed the horizon.
> I'll wait until

Well I couldn't wait ;) I downloaded 
povray-3.8.0-alpha.9475849-av541-Win64.7z

Nice one Christoph :)


-- 

Regards
     Stephen


Post a reply to this message


Attachments:
Download 'hilbert_curve_01a2m_0000.png' (922 KB)

Preview of image 'hilbert_curve_01a2m_0000.png'
hilbert_curve_01a2m_0000.png


 

From: Alain
Subject: Re: sphere_sweep artefact image
Date: 7 Mar 2018 20:50:01
Message: <5aa05079$1@news.povray.org>
Le 18-03-07 à 08:23, Stephen a écrit :
> 
> I’ve run into a problem using a large sphere_sweep. I get a horizontal, 
> straight line artefact.
> You can change the location of the artefact by raising or lowering 
> either the camera location or look_at vector.
> Happens in both ver 3.6 and 3.7
> I posted the pov and ini files in povray.binaries.scene-files
> 

Old bug that was present in version 3.5 and probably earlier.
Seems to have been fixed in the current beta, and maybe in UberPov.


Post a reply to this message

From: Stephen
Subject: Re: sphere_sweep artefact image
Date: 7 Mar 2018 21:26:32
Message: <5aa05908$1@news.povray.org>
On 07/03/2018 20:51, Alain wrote:
> Le 18-03-07 à 08:23, Stephen a écrit :
>>
>> I’ve run into a problem using a large sphere_sweep. I get a 
>> horizontal, straight line artefact.
>> You can change the location of the artefact by raising or lowering 
>> either the camera location or look_at vector.
>> Happens in both ver 3.6 and 3.7
>> I posted the pov and ini files in povray.binaries.scene-files
>>
> 
> Old bug that was present in version 3.5 and probably earlier.

I've not used it much since then.

> Seems to have been fixed in the current beta, and maybe in UberPov.

It does. :-)

-- 

Regards
     Stephen


Post a reply to this message

From: Stephen
Subject: Re: sphere_sweep artefact image
Date: 8 Mar 2018 06:00:07
Message: <5aa0d167@news.povray.org>
On 07/03/2018 15:00, Stephen wrote:
> Maybe I should have said I was using ver 3.7.0. (I've yet to update my 
> machine since I had a HDD crash.)

I compared the time taken to render the file and the 3.8 alpha is slower.

3.7.0 -  7m 11s
3.8 alpha -  8m 3s

-- 

Regards
     Stephen


Post a reply to this message

From: Stephen
Subject: Re: sphere_sweep artefact image
Date: 8 Mar 2018 06:09:43
Message: <5aa0d3a7@news.povray.org>
On 08/03/2018 06:00, Stephen wrote:
> 3.8 alpha

Forgot to put which alpha

3.8.0-alpha.9475849+av541.msvc14.win

-- 

Regards
     Stephen


Post a reply to this message

From: clipka
Subject: Re: sphere_sweep artefact image
Date: 8 Mar 2018 09:51:38
Message: <5aa107aa@news.povray.org>
Am 08.03.2018 um 07:00 schrieb Stephen:
> On 07/03/2018 15:00, Stephen wrote:
>> Maybe I should have said I was using ver 3.7.0. (I've yet to update my
>> machine since I had a HDD crash.)
> 
> I compared the time taken to render the file and the 3.8 alpha is slower.
> 
> 3.7.0 -  7m 11s
> 3.8 alpha -  8m 3s
> 
Yes, the elimination of the sphere sweep artifacts was paid for in
sphere sweep performance.


Post a reply to this message

From: Stephen
Subject: Re: sphere_sweep artefact image
Date: 8 Mar 2018 10:09:57
Message: <5aa10bf5$1@news.povray.org>
On 08/03/2018 09:51, clipka wrote:
> Am 08.03.2018 um 07:00 schrieb Stephen:
>> On 07/03/2018 15:00, Stephen wrote:
>>> Maybe I should have said I was using ver 3.7.0. (I've yet to update my
>>> machine since I had a HDD crash.)
>>
>> I compared the time taken to render the file and the 3.8 alpha is slower.
>>
>> 3.7.0 -  7m 11s
>> 3.8 alpha -  8m 3s
>>
> Yes, the elimination of the sphere sweep artifacts was paid for in
> sphere sweep performance.
> 

As long as you know. I'm happy.

-- 

Regards
     Stephen


Post a reply to this message

Goto Latest 10 Messages Next 1 Messages >>>

Copyright 2003-2008 Persistence of Vision Raytracer Pty. Ltd.