30 years on and I have a newbie problem.
I notice that my wood textures change depending on the camera angle. 
This is irksome.  i.e.
http://www.buckosoft.com/tteoac/video/frames/ttho1909.png
http://www.buckosoft.com/tteoac/video/frames/ttho1912.png

texture{
   pigment {
     P_WoodGrain1A
     color_map { M_Ash }
     rotate x*2*rand(FamRoomFloorRand)
     scale 1.5*inches*rand(FamRoomFloorRand)+0.75*feet
     rotate y*90
     translate <rand(FamRoomFloorRand)*100*inches,
                rand(FamRoomFloorRand)*100*inches,
                rand(FamRoomFloorRand)*100>
   }
}

I don't recall the dude who does the wood puzzles having this issue.

-- 
dik

Post a reply to this message

On 9-3-2017 6:11, dick balaska wrote:
> 30 years on and I have a newbie problem.
> I notice that my wood textures change depending on the camera angle.
> This is irksome.  i.e.
> http://www.buckosoft.com/tteoac/video/frames/ttho1909.png
> http://www.buckosoft.com/tteoac/video/frames/ttho1912.png
>
> texture{
>   pigment {
>     P_WoodGrain1A
>     color_map { M_Ash }
>     rotate x*2*rand(FamRoomFloorRand)
>     scale 1.5*inches*rand(FamRoomFloorRand)+0.75*feet
>     rotate y*90
>     translate <rand(FamRoomFloorRand)*100*inches,
>                rand(FamRoomFloorRand)*100*inches,
>                rand(FamRoomFloorRand)*100>
>   }
> }
>
> I don't recall the dude who does the wood puzzles having this issue.
>

Isn't that because of all those rand()? In an animation I would not be 
surprised if this happened. The trick would be to keep the texture 
declaration outside of the video loop I guess.

-- 
Thomas

Post a reply to this message

On 3/9/2017 7:59 AM, Thomas de Groot wrote:
> On 9-3-2017 6:11, dick balaska wrote:
>> 30 years on and I have a newbie problem.
>> I notice that my wood textures change depending on the camera angle.
>> This is irksome.  i.e.
>> http://www.buckosoft.com/tteoac/video/frames/ttho1909.png
>> http://www.buckosoft.com/tteoac/video/frames/ttho1912.png
>>
>> texture{
>>   pigment {
>>     P_WoodGrain1A
>>     color_map { M_Ash }
>>     rotate x*2*rand(FamRoomFloorRand)
>>     scale 1.5*inches*rand(FamRoomFloorRand)+0.75*feet
>>     rotate y*90
>>     translate <rand(FamRoomFloorRand)*100*inches,
>>                rand(FamRoomFloorRand)*100*inches,
>>                rand(FamRoomFloorRand)*100>
>>   }
>> }
>>
>> I don't recall the dude who does the wood puzzles having this issue.
>>
>
> Isn't that because of all those rand()? In an animation I would not be
> surprised if this happened. The trick would be to keep the texture
> declaration outside of the video loop I guess.
>

I suspect that too.
Mike what happens if you animate the scene without moving the camera?

-- 

Regards
     Stephen

Post a reply to this message

Am 09.03.2017 um 06:11 schrieb dick balaska:
> 30 years on and I have a newbie problem.
> I notice that my wood textures change depending on the camera angle.
> This is irksome.  i.e.
> http://www.buckosoft.com/tteoac/video/frames/ttho1909.png
> http://www.buckosoft.com/tteoac/video/frames/ttho1912.png
> 
> texture{
>   pigment {
>     P_WoodGrain1A
>     color_map { M_Ash }
>     rotate x*2*rand(FamRoomFloorRand)
>     scale 1.5*inches*rand(FamRoomFloorRand)+0.75*feet
>     rotate y*90
>     translate <rand(FamRoomFloorRand)*100*inches,
>                rand(FamRoomFloorRand)*100*inches,
>                rand(FamRoomFloorRand)*100>
>   }
> }
> 
> I don't recall the dude who does the wood puzzles having this issue.

I suspect what you're seeing isn't actually the wood grain, but a moiree
pattern created by that wood grain and the pixel grid.

Unfortunately, when the "frequency" of a pattern almost coincides with
the "frequency" of the pixel grid, POV-Ray's standard anti-aliasing
offers little help.

What you would need to do is to enforce some stochastic oversampling,
for example by adding a bit of focal blur. Alternatively, you might want
to try UberPOV's anti-aliasing mode 3.

Post a reply to this message

Am 2017-03-09 02:59, also sprach Thomas de Groot:

> Isn't that because of all those rand()? In an animation I would not be
> surprised if this happened. The trick would be to keep the texture
> declaration outside of the video loop I guess.
>

No, those are so each board is different. You know,
loop {
   box {
     0,1
     texture {
       wood
       translate <rand,rand,rand>
     }
   }
}
The TV unit has no rands, and it "sparkles" as the camera moves past it.
This tile floor is built similarly with T_Stone8 and lots of rands
http://www.buckosoft.com/tteoac/video/frames/ttho0109.png
and it doesn't suffer the same "random" fate.

-- 
dik

Post a reply to this message

Am 2017-03-09 09:21, also sprach clipka:
>
> I suspect what you're seeing isn't actually the wood grain, but a moiree
> pattern created by that wood grain and the pixel grid.

https://xkcd.com/1814/

It sure looks and acts like a moire.

But, but, but, why is it only an issue when I change the angle of the 
camera on the object, and not the distance.  And why do I only see this 
with woods at any scale and not stones or other fractals at any scale.

It seems odd.

 > What you would need to do is to enforce some stochastic oversampling,
 > for example by adding a bit of focal blur. Alternatively, you might want
 > to try UberPOV's anti-aliasing mode 3.

Focal blur is out of the question. UberPOV, hmmm...

-- 
dik

Post a reply to this message

Am 23.03.2017 um 06:25 schrieb dick balaska:
> Am 2017-03-09 09:21, also sprach clipka:
>>
>> I suspect what you're seeing isn't actually the wood grain, but a moiree
>> pattern created by that wood grain and the pixel grid.
> 
> https://xkcd.com/1814/
> 
> It sure looks and acts like a moire.
> 
> But, but, but, why is it only an issue when I change the angle of the
> camera on the object, and not the distance.

Hmm... dunno about that. You might have to provide a genuine animation
where we can see the thing in action.

> And why do I only see this
> with woods at any scale and not stones or other fractals at any scale.

'Cause wood tends to be more regular than stones or fractals?

The more noise you have in a pattern, the less likely you are to get a
moire effect.

Post a reply to this message

Le 17-03-23 à 01:25, dick balaska a écrit :
> Am 2017-03-09 09:21, also sprach clipka:
>>
>> I suspect what you're seeing isn't actually the wood grain, but a moiree
>> pattern created by that wood grain and the pixel grid.
>
> https://xkcd.com/1814/
>
> It sure looks and acts like a moire.
>
> But, but, but, why is it only an issue when I change the angle of the
> camera on the object, and not the distance.  And why do I only see this
> with woods at any scale and not stones or other fractals at any scale.
>
> It seems odd.
>
>> What you would need to do is to enforce some stochastic oversampling,
>> for example by adding a bit of focal blur. Alternatively, you might want
>> to try UberPOV's anti-aliasing mode 3.
>
> Focal blur is out of the question. UberPOV, hmmm...
>
You also get a moiré effect with onion, gradient and checker. If you use 
the cells pattern the moiré is almost inexistant.
wood, checker and onion are periodic pattern, cells is not.
Most stone are also aperiodic. fractals are also aperiodics.
Periodic or repeating patterns, when scalled small enough cause moiré, 
while irregular patterns don't.
By small enough, I mean that the details are pixel sized or smaller.

Post a reply to this message

On 03/23/2017 01:25 AM, dick balaska wrote:
> Am 2017-03-09 09:21, also sprach clipka:
>>
>> I suspect what you're seeing isn't actually the wood grain, but a moiree
>> pattern created by that wood grain and the pixel grid.
>
> https://xkcd.com/1814/
>
> It sure looks and acts like a moire.
>
> But, but, but, why is it only an issue when I change the angle of the
> camera on the object, and not the distance.  And why do I only see this
> with woods at any scale and not stones or other fractals at any scale.
>

Thought just hit me, if mostly angle dependent, using the aoi pattern to 
gradually change from a non-banded wood texture (average-wood-color) to 
the normal banded wood texture I suspect would help.

Rendering larger images - no AA - and shrinking to the final size 
perhaps a path too.

Bill P.

Post a reply to this message