POV-Ray : Newsgroups : povray.programming : Infinite coherent noise? Server Time
28 Mar 2024 10:17:28 EDT (-0400)
  Infinite coherent noise? (Message 1 to 8 of 8)  
From: stbenge
Subject: Infinite coherent noise?
Date: 21 Mar 2008 17:54:19
Message: <47e43c9b$1@news.povray.org>
Hi,

I was wondering how POV is able to render coherent noise. Is it truly 
infinite, or just nearly? I've never seen a repeat, but perhaps it 
occurs just past the limit of observation.

I'm coding a simple demo of randomly generated topography, with the 
ability to scroll around the landscape. The only noise generators I've 
found reluctantly admit that the noise is "nearly infinite". I want an 
absolutely infinite landscape, but I don't know if it's even possible. 
Thanks~

Sam


Post a reply to this message

From: Mike Williams
Subject: Re: Infinite coherent noise?
Date: 21 Mar 2008 20:30:46
Message: <4G11TJC15F5HFwYn@econym.demon.co.uk>
Wasn't it stbenge who wrote:
>Hi,
>
>I was wondering how POV is able to render coherent noise. Is it truly 
>infinite, or just nearly? I've never seen a repeat, but perhaps it 
>occurs just past the limit of observation.
>
>I'm coding a simple demo of randomly generated topography, with the 
>ability to scroll around the landscape. The only noise generators I've 
>found reluctantly admit that the noise is "nearly infinite". I want an 
>absolutely infinite landscape, but I don't know if it's even possible.

It's actually worse than that. Any finite computing system has a maximum 
number that its arithmetic system can represent. That places an absolute 
limit on the size of any landscape that the computer can represent, 
whether it involves random elements or not.

It's possible to create hardware random number generators that are truly 
random, using quantum effects, but they'd be useless for generating 
randomly generated topology because they wouldn't produce the same 
random numbers when you re-render the scene.

-- 
Mike Williams
Gentleman of Leisure


Post a reply to this message

From: stbenge
Subject: Re: Infinite coherent noise?
Date: 23 Mar 2008 00:59:50
Message: <47e5f1d6@news.povray.org>
Mike Williams wrote:
> Wasn't it stbenge who wrote:
>> Hi,
>>
>> I was wondering how POV is able to render coherent noise. Is it truly 
>> infinite, or just nearly? I've never seen a repeat, but perhaps it 
>> occurs just past the limit of observation.
>>
>> I'm coding a simple demo of randomly generated topography, with the 
>> ability to scroll around the landscape. The only noise generators I've 
>> found reluctantly admit that the noise is "nearly infinite". I want an 
>> absolutely infinite landscape, but I don't know if it's even possible.
> 
> It's actually worse than that. Any finite computing system has a maximum 
> number that its arithmetic system can represent. That places an absolute 
> limit on the size of any landscape that the computer can represent, 
> whether it involves random elements or not.

Well, I just made a random landscape generator, and it's huge. I'm sure 
it repeats eventually, but I haven't seen it happen.

> It's possible to create hardware random number generators that are truly 
> random, using quantum effects, but they'd be useless for generating 
> randomly generated topology because they wouldn't produce the same 
> random numbers when you re-render the scene.

I think we are still a ways off from seeing a commercial quantum 
computer. Is there a way to achieve quantum effects apart from the 
corresponding hardware?

Sam


Post a reply to this message

From: Mike Williams
Subject: Re: Infinite coherent noise?
Date: 23 Mar 2008 04:22:48
Message: <$xOEunBYFi5HFwq7@econym.demon.co.uk>
Wasn't it stbenge who wrote:
>Mike Williams wrote:
>> Wasn't it stbenge who wrote:
>>> Hi,
>>>
>>> I was wondering how POV is able to render coherent noise. Is it 
>>>truly  infinite, or just nearly? I've never seen a repeat, but 
>>>perhaps it  occurs just past the limit of observation.
>>>
>>> I'm coding a simple demo of randomly generated topography, with the 
>>>ability to scroll around the landscape. The only noise generators 
>>>I've  found reluctantly admit that the noise is "nearly infinite". I 
>>>want an  absolutely infinite landscape, but I don't know if it's even >>>possible.
>>  It's actually worse than that. Any finite computing system has a 
>>maximum  number that its arithmetic system can represent. That places 
>>an absolute  limit on the size of any landscape that the computer can 
>>represent,  whether it involves random elements or not.
>
>Well, I just made a random landscape generator, and it's huge. I'm sure 
>it repeats eventually, but I haven't seen it happen.

Well that's what you'd expect with "nearly infinite". I wouldn't expect 
to see any repetition in a landscape the size of the Earth.

>> It's possible to create hardware random number generators that are 
>>truly  random, using quantum effects, but they'd be useless for 
>>generating  randomly generated topology because they wouldn't produce 
>>the same  random numbers when you re-render the scene.
>
>I think we are still a ways off from seeing a commercial quantum 
>computer. Is there a way to achieve quantum effects apart from the 
>corresponding hardware?

Well you certainly can't write software that's sensitive to quantum 
fluctuations on a deterministic digital computer, so you either need 
some hardware that is sensitive to quantum fluctuations, or an external 
feed of data that is generated by such hardware. Such external data 
feeds include things like UK Premium Bond winning numbers and Lottery 
winning numbers.

[UK Premium Bond numbers use the signal noise in transistors. Each 
impact of the balls in a lottery machine doubles the angle of 
reflection, so after a few hundred impacts a difference in initial 
conditions at the quantum level would be amplified sufficiently to cause 
a different ball to be selected.]

We're not talking "quantum computers" here. That's something completely 
different.

-- 
Mike Williams
Gentleman of Leisure


Post a reply to this message

From: ingo
Subject: Re: Infinite coherent noise?
Date: 26 Mar 2008 14:23:18
Message: <Xns9A6DCF6062D01seed7@news.povray.org>
in news:47e43c9b$1@news.povray.org stbenge wrote:

> The only noise generators I've 
> found reluctantly admit that the noise is "nearly infinite".

Could you use data from http://www.random.org?

ingo


Post a reply to this message

From: stbenge
Subject: Re: Infinite coherent noise?
Date: 26 Mar 2008 15:55:04
Message: <47eab828@news.povray.org>
ingo wrote:
> in news:47e43c9b$1@news.povray.org stbenge wrote:
> 
>> The only noise generators I've 
>> found reluctantly admit that the noise is "nearly infinite".
> 
> Could you use data from http://www.random.org?
> 
I don't think so. I need determinate numbers, ones which I can expect to 
see every time I punch in a new seed value.

Sam


Post a reply to this message

From: stbenge
Subject: Re: Infinite coherent noise?
Date: 26 Mar 2008 15:57:52
Message: <47eab8d0@news.povray.org>
Mike Williams wrote:
> Wasn't it stbenge who wrote:

>> Well, I just made a random landscape generator, and it's huge. I'm 
>> sure it repeats eventually, but I haven't seen it happen.
> 
> Well that's what you'd expect with "nearly infinite". I wouldn't expect 
> to see any repetition in a landscape the size of the Earth.

Using what as a measuring stick? 1 pixel = 1 atom?

>>> It's possible to create hardware random number generators that are 
>>> truly  random, using quantum effects, but they'd be useless for 
>>> generating  randomly generated topology because they wouldn't produce 
>>> the same  random numbers when you re-render the scene.
>>
>> I think we are still a ways off from seeing a commercial quantum 
>> computer. Is there a way to achieve quantum effects apart from the 
>> corresponding hardware?
> 
> Well you certainly can't write software that's sensitive to quantum 
> fluctuations on a deterministic digital computer, so you either need 
> some hardware that is sensitive to quantum fluctuations, or an external 
> feed of data that is generated by such hardware. Such external data 
> feeds include things like UK Premium Bond winning numbers and Lottery 
> winning numbers.
> 
> [UK Premium Bond numbers use the signal noise in transistors. Each 
> impact of the balls in a lottery machine doubles the angle of 
> reflection, so after a few hundred impacts a difference in initial 
> conditions at the quantum level would be amplified sufficiently to cause 
> a different ball to be selected.]
> 
> We're not talking "quantum computers" here. That's something completely 
> different.

We're talking about chaos theory, yes? Nothing is truly random, rather 
everything is a result of cause and effect. But try to predict it :)

Sam


Post a reply to this message

From: Mike Williams
Subject: Re: Infinite coherent noise?
Date: 27 Mar 2008 00:19:27
Message: <Byu7ufNV1y6HFw0i@econym.demon.co.uk>
Wasn't it stbenge who wrote:
>Mike Williams wrote:
>> Wasn't it stbenge who wrote:
>
>>> I think we are still a ways off from seeing a commercial quantum 
>>>computer. Is there a way to achieve quantum effects apart from the 
>>>corresponding hardware?
...
>>  We're not talking "quantum computers" here. That's something 
>>completely  different.
>
>We're talking about chaos theory, yes? Nothing is truly random, rather 
>everything is a result of cause and effect. But try to predict it :)

We're talking real quantum effects (well at least I was) which most 
current theories consider to be truly random. Chaos theory tells us that 
we can build a machine that is so sensitive to initial conditions that 
quantum events occurring in the atoms on the surface of the lottery 
balls would cause different lottery outcomes.

Quantum computers work by causing a particle to be in a superposition of 
many quantum states, each state representing a possible value that might 
be the answer to some question, and somehow causing the superposition to 
collapse to the state which solves the question that's being asked. 
That's sort of going in the opposite direction from what we want.

-- 
Mike Williams
Gentleman of Leisure


Post a reply to this message

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