POV-Ray : Newsgroups : povray.advanced-users : Weird results scaling normals Server Time
26 Dec 2024 14:04:39 EST (-0500)
  Weird results scaling normals (Message 1 to 10 of 10)  
From: Nieminen Juha
Subject: Weird results scaling normals
Date: 24 Nov 1999 13:37:11
Message: <383c3057@news.povray.org>
Try this:

camera { location y*4-z*4 look_at 0 angle 35 }
light_source { <100,200,-300> 1 }
#declare Granite = texture { pigment { rgb x } normal { granite 1 } }
plane
{ y,0 texture { Granite }
}

  And now try this:

camera { location y*4-z*4 look_at 0 angle 35 }
light_source { <100,200,-300> 1 }
#declare Granite = texture { pigment { rgb x } normal { granite } }
plane
{ y,0
  texture
  { gradient x texture_map
    { [0 Granite scale .01][1 Granite scale .01]
    }
    scale 100
  }
}

  The result should be identical, but it isn't.
  I remember reading about problems when scaling normals. Is this the one?

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/


Post a reply to this message

From: Ken
Subject: Re: Weird results scaling normals
Date: 24 Nov 1999 13:39:41
Message: <383C308E.AA2D8ADA@pacbell.net>
Nieminen Juha wrote:

> 
>   The result should be identical, but it isn't.
>   I remember reading about problems when scaling normals. Is this the one?

Sounds familiar. If I am not mistaken there is a bug report addressing
this issue.

-- 
Ken Tyler -  1200+ Povray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/


Post a reply to this message

From: Nieminen Juha
Subject: Re: Weird results scaling normals
Date: 24 Nov 1999 13:47:23
Message: <383c32bb@news.povray.org>
Btw, this causes serious problems. Let me explain where did I find this
bug.
  I was trying to make a plane that looks like a terrain, with ground, woods
and rivers. I declared a ground texture (the granite normal is really good
for this), a water texture and a forest texture.
  To apply the forest to the ground, I made a bozo texture map with them.
Then I made a turbulenced gradient texture map with this and the
water to add rivers.
  Since the rivers where too close to each other, I scaled up the texture.
This of course scaled up all the subtextures as well. I wanted to scale only
the texture map and not the textures inside it. The only way of doing this
is to scale down the subtextures by the same amount.
  However, as we can see, this breaks the normals. They become really
weird. There's no way to fix this.

  Then I tried to achieve this by layering the textures, but for some
reason you can't layer patterned textures. What a problem!

  I ended up using separated planes for each layer. This is easy when
handling planes, but what about more complex objects? There should be an
easier way.

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/


Post a reply to this message

From: Gilles Tran
Subject: Re: Weird results scaling normals
Date: 24 Nov 1999 14:01:26
Message: <383C361B.8B9EA438@inapg.inra.fr>
May be a stupid remark, but what about scaling the normals within the normal
statement until they look right ?
G.


Post a reply to this message

From: Ken
Subject: Re: Weird results scaling normals
Date: 24 Nov 1999 14:16:59
Message: <383C392F.CCAA5C67@pacbell.net>
Gilles Tran wrote:
> 
> May be a stupid remark, but what about scaling the normals within the normal
> statement until they look right ?
> G.

 I'll tell you why for Warp this is unacceptable.

  Warp has a very analytical approach to scene design and expects that
when he does this it should do that. No fault in that I suppose except
when it does not behave as it is supposed to.

  Me, on the other hand, do not always try to predict the outcome based on
numerical fact. I just tweak things until they look good. Much as a painter
will try a different brush if the brush stroke looks wrong with the one
he is using.

  A clash between art and science.

-- 
Ken Tyler -  1200+ Povray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/


Post a reply to this message

From: Nieminen Juha
Subject: Re: Weird results scaling normals
Date: 24 Nov 1999 14:33:55
Message: <383c3da3@news.povray.org>
Ken <tyl### [at] pacbellnet> wrote:
:  I'll tell you why for Warp this is unacceptable.

  Because it doesn't work. As I said before, there's no fix for this.

camera { location y*4-z*4 look_at 0 angle 35 }
light_source { <100,200,-300> 1 }
#declare Granite = texture { pigment { rgb x } normal { granite scale .01 } }
plane
{ y,0
  texture
  { gradient x texture_map
    { [0 Granite][1 Granite]
    }
    scale 100
  }
}

  It's still wrong (and identical to the second scene in my first post).

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/


Post a reply to this message

From: Rune
Subject: Re: Weird results scaling normals
Date: 27 Nov 1999 20:41:11
Message: <38408837@news.povray.org>
Ken wrote:
>Nieminen Juha wrote:
>>
>>   The result should be identical, but it isn't.
>>   I remember reading about problems when scaling normals. Is this the
one?
>
>Sounds familiar. If I am not mistaken there is a bug report addressing
>this issue.

Yes, I posted a message in povray.general and later posted it to
povray.bugreports too.
Both posts are entitled "Normal scaling bug(s)".

Like Warp I think it's a very serious problem and I keep finding myself in
situations where I can't get the desired results because of this bug.

I really hope it will be fixed in the next version, but that will also
create a problem because it will break a very large percentage of all
existing scenes.

Greetings,

Rune

---
Updated November 19: http://rsj.mobilixnet.dk
Containing 3D images, stereograms, tutorials,
The POV Desktop Theme, 350+ raytracing jokes,
miscellaneous other things, and a lot of fun!


Post a reply to this message

From: TonyB
Subject: Re: Weird results scaling normals
Date: 28 Nov 1999 07:46:31
Message: <38412427@news.povray.org>
>I really hope it will be fixed in the next version, but that will also
>create a problem because it will break a very large percentage of all
>existing scenes.

Also the bug regarding the incorrect distribution of noise will cause this
problem, but what the heck, so long as the program works correctly. Perhaps
adding the option to use the old, incorrect methods using the #version
thingy would prevent problems?


Post a reply to this message

From: Nathan Kopp
Subject: Re: Weird results scaling normals
Date: 29 Nov 1999 00:03:12
Message: <38420910@news.povray.org>
This will be fixed in the next release of UVPov.

-Nathan

Nieminen Juha <war### [at] punarastascstutfi> wrote...
>   Try this:
>
> camera { location y*4-z*4 look_at 0 angle 35 }
> light_source { <100,200,-300> 1 }
> #declare Granite = texture { pigment { rgb x } normal { granite 1 } }
> plane
> { y,0 texture { Granite }
> }
>
>   And now try this:
>
> camera { location y*4-z*4 look_at 0 angle 35 }
> light_source { <100,200,-300> 1 }
> #declare Granite = texture { pigment { rgb x } normal { granite } }
> plane
> { y,0
>   texture
>   { gradient x texture_map
>     { [0 Granite scale .01][1 Granite scale .01]
>     }
>     scale 100
>   }
> }
>
>   The result should be identical, but it isn't.
>   I remember reading about problems when scaling normals. Is this the one?
>
> --
> main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
> ):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/


Post a reply to this message

From: TonyB
Subject: Re: Weird results scaling normals
Date: 29 Nov 1999 07:18:11
Message: <38426f03@news.povray.org>
Nathan! Nathan! Nathan! Nathan! Nathan! Nathan! Nathan! Wooo! =D


PS: Do you suppose you could add slope-dependant textures to it soon? I want
to render my cloud scene in one version instead of cutting out the clouds to
be able to work on the mountains. Pleeease.... pretty please.... with
(insert favorite topping) on top... please. :)


Post a reply to this message

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