Hi there !

Today I developed a solution for real random numbers in the Povray-Code.
Not pseudo random...real random.

The Idea was how to include the time into Povray.
Now I have always a new picture, when I start a new rendering of it.

OK...if somebody else is also interested on real random numbers...please reply
to me.


Greetings to Povray-Users
Flynn

Post a reply to this message

Yes, I was thinking about involving time into it, too. But then again,
two different computers and rendering at the same time - will lead again
to the same result. So, while POV-Ray gives us a pseudo random, you give
us now a pseudo pseudo random.

But don't get me wrong - I was about to have it done the same way. And
it is a good idea.

If you want to go that way further, and want to work towards the third
pseudo, you could also involve computer specs to the time-depending
random. And also things that change frequently on computers, like amount
of free HDD space and size of random-selected files also being computed
into the pseudo pseudo pseudo random.

POV-Ray's random was intentionally meant to enable everyone to produce
the same result. But taking your point, yes, it would be nice if POV-Ray
4.0 could supply us with a real random, too. :-)

Post a reply to this message

Sven Littkowski <jam### [at] yahoocom> wrote:
> Yes, I was thinking about involving time into it, too. But then again,
> two different computers and rendering at the same time - will lead again
> to the same result. So, while POV-Ray gives us a pseudo random, you give
> us now a pseudo pseudo random.
>
> But don't get me wrong - I was about to have it done the same way. And
> it is a good idea.
>
> If you want to go that way further, and want to work towards the third
> pseudo, you could also involve computer specs to the time-depending
> random. And also things that change frequently on computers, like amount
> of free HDD space and size of random-selected files also being computed
> into the pseudo pseudo pseudo random.
>
> POV-Ray's random was intentionally meant to enable everyone to produce
> the same result. But taking your point, yes, it would be nice if POV-Ray
> 4.0 could supply us with a real random, too. :-)

Hi Sven !

Thanks for your funny message !
What you wrote, that there is a reason why we get pseudo-random to produce same


But !!!!! Yeeess...the must be a BUT !

I like to make pics using random numbers to see different results, and take the
best (what I think what the best is).
So I think both options are very useful.
And 2 Computers produces the same results when they start rendering the same
time...yes...when they have exactly the same time...down to
microseconds...yes...there is always a limit for us.
And you have really great ideas to get more better
pseudo-pseudo-pseudo-pseudo-numbers using files and HDD space. Thanks for it !!!

Bye
Flynn :-)

Post a reply to this message

"Flynn" <pot### [at] gmxnet> wrote:
> Sven Littkowski <jam### [at] yahoocom> wrote:
> > Yes, I was thinking about involving time into it, too. But then again,
> > two different computers and rendering at the same time - will lead again
> > to the same result. So, while POV-Ray gives us a pseudo random, you give
> > us now a pseudo pseudo random.
> >
> > But don't get me wrong - I was about to have it done the same way. And
> > it is a good idea.
> >
> > If you want to go that way further, and want to work towards the third
> > pseudo, you could also involve computer specs to the time-depending
> > random. And also things that change frequently on computers, like amount
> > of free HDD space and size of random-selected files also being computed
> > into the pseudo pseudo pseudo random.
> >
> > POV-Ray's random was intentionally meant to enable everyone to produce
> > the same result. But taking your point, yes, it would be nice if POV-Ray
> > 4.0 could supply us with a real random, too. :-)
>
> Hi Sven !
>
> Thanks for your funny message !
> What you wrote, that there is a reason why we get pseudo-random to produce same

>
> But !!!!! Yeeess...the must be a BUT !
>
> I like to make pics using random numbers to see different results, and take the
> best (what I think what the best is).
> So I think both options are very useful.
> And 2 Computers produces the same results when they start rendering the same
> time...yes...when they have exactly the same time...down to
> microseconds...yes...there is always a limit for us.
> And you have really great ideas to get more better
> pseudo-pseudo-pseudo-pseudo-numbers using files and HDD space. Thanks for it !!!
>
> Bye
> Flynn :-)

There's also rand.inc which has several different random number generators.

I used it myself in my packing crate image to randomize the surface on the
iso-planks

http://news.povray.org/povray.binaries.images/thread/%3C50765d2d%40news.povray.org%3E/?ttop=404540&toff=600

and again to randomize the distribution of particles in my spray-paint macro:

http://news.povray.org/web.5083ad99befa8986d97ee2b90%40news.povray.org

Regards,
A.D.B.

Post a reply to this message

Ah...OK...

thanks...
Flynn

Post a reply to this message

Am 26.01.2016 um 03:18 schrieb Flynn:

> Today I developed a solution for real random numbers in the Povray-Code.
> Not pseudo random...real random.
> 
> The Idea was how to include the time into Povray.
> Now I have always a new picture, when I start a new rendering of it.
> 
> OK...if somebody else is also interested on real random numbers...please reply
> to me.

While I hate to be a spoil-sport, I think you might be interested to
hear that POV-Ray 3.7 already has intrinsic support for this, via the
following idiom ;)

    #declare RandomSource = seed(now * FACTOR);

where FACTOR is the fraction of a day after which you are guaranteed to
get a new random sequence; e.g. FACTOR=24 gives you a new sequence every
hour, while FACTOR=24*60*60 gives you new values every second. (Larger
values may or may not work as expected, depending on the platform.)

I just noticed that this has been poorly documented though, with the
"now" keyword being only mentioned in the context of the "datetime()"
function.

Post a reply to this message

I know the "know" keyword from PHP. Interesting. Good to know now.   :-)

Post a reply to this message

"Anthony D. Baye" <Sha### [at] spamnomorehotmailcom> wrote:

>
> There's also rand.inc which has several different random number generators.
>

As far as I can tell, those generators are all pseudo-random too -- well,
pseudo-pseudo-pseudo....random ;-)  By using lots of tricks to get the final
values, but still based on seed(...) and rand(...)

Post a reply to this message

Am 27.01.2016 um 23:18 schrieb Kenneth:
> "Anthony D. Baye" <Sha### [at] spamnomorehotmailcom> wrote:
> 
>>
>> There's also rand.inc which has several different random number generators.
>>
> 
> As far as I can tell, those generators are all pseudo-random too -- well,
> pseudo-pseudo-pseudo....random ;-)  By using lots of tricks to get the final
> values, but still based on seed(...) and rand(...)

Actually those aren't (pseudo-)random number generators in the usual
sense, nor are they intended to be, but rather macros that use a
pseudo-random number generator to generate pseudo-random numbers with
particular non-linear distributions.

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Am 27.01.2016 um 23:18 schrieb Kenneth:

> > Kennth wrote:
> >
> > As far as I can tell, those generators are all pseudo-random too -- well,
> > pseudo-pseudo-pseudo....random ;-)  By using lots of tricks to get the final
> > values, but still based on seed(...) and rand(...)
>
> Actually those aren't (pseudo-)random number generators in the usual
> sense, nor are they intended to be, but rather macros that use a
> pseudo-random number generator to generate pseudo-random numbers with
> particular non-linear distributions.

Ah yes, I see what you mean. (I've never had a need --that I know of! -- for
using such non-linear distributions. I guess I need to do some research about
such things.)

Post a reply to this message