How can I get a smooth random number, ie, it varies from the last random
number by no more than a certain amount? (I'm making a tree macro and
want a way to twist the limbs.)

Post a reply to this message

Larry Fontaine wrote:
> 
> How can I get a smooth random number, ie, it varies from the last random
> number by no more than a certain amount? (I'm making a tree macro and
> want a way to twist the limbs.)

I'll send you a possible solution by email. I would post the source but
it's a bit longish for this group.

-- 
Ken Tyler
1100+ Povray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html

Post a reply to this message

On Sat, 09 Oct 1999 22:32:33 -0500, Larry Fontaine <lfo### [at] isdnet>
wrote:

>How can I get a smooth random number, ie, it varies from the last random
>number by no more than a certain amount? (I'm making a tree macro and
>want a way to twist the limbs.)

There's a smooth random generator in the Throroughly Useful Macros by
John VanSickle. Search Ken's links :)


Peter Popov
ICQ: 15002700

Post a reply to this message

Peter Popov wrote:
> 
> On Sat, 09 Oct 1999 22:32:33 -0500, Larry Fontaine <lfo### [at] isdnet>
> wrote:
> 
> >How can I get a smooth random number, ie, it varies from the last random
> >number by no more than a certain amount? (I'm making a tree macro and
> >want a way to twist the limbs.)
> 
> There's a smooth random generator in the Throroughly Useful Macros by
> John VanSickle. Search Ken's links :)
> 
> Peter Popov
> ICQ: 15002700

I already sent it to him :)

-- 
Ken Tyler -  1100+ 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

I'm not sure what you're after. The smooth rand macro is one solution.
I can also send you a Perlin noise macro I created. It takes a float as the
parameter and returns values within a certain limit. Close parameters give
statistically close values. The result is similar to turbulence - you have
control over octaves and omega - but the noise is 1D, i.e. it returns a float.
Let me know if you're interested, because I'd have to clean up the source before
posting it.

Margus

Larry Fontaine wrote:
> 
> How can I get a smooth random number, ie, it varies from the last random
> number by no more than a certain amount? (I'm making a tree macro and
> want a way to twist the limbs.)

Post a reply to this message

Larry Fontaine <lfo### [at] isdnet> wrote:
: How can I get a smooth random number, ie, it varies from the last random
: number by no more than a certain amount?

  If you mean a linear distribution, that's pretty simple:

#declare Seed = seed(0);
#declare Number = rand(Seed);
#macro NextNumber(maxvariation)
  #declare Number = Number-maxVariation+maxvariation*rand(Seed)*2;
  Number
#end

-- 
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

Nieminen Juha wrote:
> 
>   If you mean a linear distribution, that's pretty simple:
> 

AFAIK, this is normal distribution.
I made two little macros that do normal distribution for floats and 3D vectors
(it's trivial to extend it to nD vectors). They take the mean, the maximum
deviation and a predeclared seed as parameters.
The macros are available from the Macroscope (rand_ext and v_rand_ext)

Margus

Post a reply to this message

Margus Ramst <mar### [at] peakeduee> wrote:
: AFAIK, this is normal distribution.

  No, normal distribution is distributed in the form e^(-x^2). You can get
an approximation by calculating the average of at least three random numbers
(the more the numbers, the better the approximation, but three is usually
close enough).

-- 
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

Nieminen Juha wrote:
> 
> Margus Ramst <mar### [at] peakeduee> wrote:
> : AFAIK, this is normal distribution.
> 
>   No, normal distribution is distributed in the form e^(-x^2). You can get
> an approximation by calculating the average of at least three random numbers
> (the more the numbers, the better the approximation, but three is usually
> close enough).
> 

The more random numbers summed, the smaller the standard deviation.
PoD.

Post a reply to this message

PoD wrote:
> 
> The more random numbers summed, the smaller the standard deviation.
> PoD.

But that's just a relative measure, anyway. You can increase dispersion by
scaling with the mean as the origin.

Margus

Post a reply to this message