1. Let's try this ONE more time :-)
For people that cant read English, i post this one more
time:
constant n=#10000
function Rand32()
--no reason why you cant make this more inline...
atom a,b,c
a=rand(n)-1
b=rand(n)-1
c=a*n+b
return c
end function
Notes:
1. For two added rnd numbers the distribution should be the same.
It's multiplication that changes the distribution.
2. Two rand sequences from the same generator that show no
correlation, when added, form a new rand sequence that will
show no correlation to another rand sequence generated in
exactly the same way.
3. Multiplication, such as rand1*rand2 changes the distribution greatly.
4. Note that to generate 'a' above you cant use rand(n).
Take care,
Al
And, good luck with your Euphoria programming!
My bumper sticker: "I brake for LED's"
2. Re: Let's try this ONE more time :-)
On Mon, 05 Jul 2004 11:51:21 -0700, Al Getz <guest at RapidEuphoria.com>
wrote:
>For people that cant read English, i post this one more time:
And people say I need to try decaffeinated. 
)
3. Re: Let's try this ONE more time :-)
Pete Lomax wrote:
>
> On Mon, 05 Jul 2004 11:51:21 -0700, Al Getz <guest at RapidEuphoria.com>
> wrote:
>
> >For people that cant read English, i post this one more time:
> And people say I need to try decaffeinated. )
>
>
There were at least TWO (count em, 2> ) posts suggesting
a method of generating 32 bit random numbers which were
perfectly fine, and yet people were still bickering over
what to do about generating a 32 bit rand number ??
Makes me wonder about the intellegence level in this group,
any IQ's over 20 here, or did i miss something somewhere...
Take care,
Al
And, good luck with your Euphoria programming, if you can find
the keyboard, that is!
My bumper sticker: "I brake for LED's"
4. Re: Let's try this ONE more time :-)
On Mon, 05 Jul 2004 17:24:00 -0700, Al Getz <guest at RapidEuphoria.com>
wrote:
>Makes me wonder about the intellegence level in this group,
>any IQ's over 20 here, or did i miss something somewhere...
Charm on overdrive, I see.
5. Re: Let's try this ONE more time :-)
Al Getz wrote:
>
> Pete Lomax wrote:
> >
> > On Mon, 05 Jul 2004 11:51:21 -0700, Al Getz <guest at RapidEuphoria.com>
> > wrote:
> >
> > >For people that cant read English, i post this one more time:
> > And people say I need to try decaffeinated. )
> >
> >
> There were at least TWO (count em, 2> ) posts suggesting
> a method of generating 32 bit random numbers which were
> perfectly fine, and yet people were still bickering over
> what to do about generating a 32 bit rand number ??
Bickering? Suggesting alternates is normal in programming. Why do you
think this is evidence of bickering?
There are a huge number of ways to generate a 32-bit random number. So
what's the problem if people suggest some alternatives?
> Makes me wonder about the intellegence level in this group,
> any IQ's over 20 here, or did i miss something somewhere...
Apparently so. For example, "cant" is really spelled "can't";
the 1-st person singular pronoun is supposed to be spelled "I" and not
"i"; and adding two random numbers does not produce a third random number.
Let's say we are trying for random numbers from 1 to 100. If we
are using two independant random number generators 'randA' and 'randB',
and use something like floor(randA(100) + randB(100)) / 2) to do this.
This seems reasonable, but consider this...
What is the probability of getting 25? Which is two random numbers adding
up to 50 then dividing by 2? Quite high, actually. We have 1+49, 2+48,
3+47, ... 49+1; which 49 out of a possible 10000 different sums.
Now how about the probability of 1? (two randoms adding up to 2).
1+1; which is 1 out of 10000.
Thus we have very poor distribution of numbers if we just use the
additive method. And the story is much the same for multipliction,
though the distribution curve is a different shape.
For concatenation though, its more like this...
Formula --> (randA(10)-1) * 10 + (randB(10)-1) + 1
The probability for 25 ('2' followed by '5') is (1/10 * 1/10) ==> 1/100.
The probability for 1 ('0' followed by '1') is (1/10 * 1/10) ==> 1/100.
Etc for all numbers in the range. A uniform probability distribution.
> Take care,
Good advice, doctor.
--
Derek Parnell
Melbourne, Australia
6. Re: Let's try this ONE more time :-)
Hi Al,
You wrote:
----------
> From: Al Getz <guest at RapidEuphoria.com>
> To: EUforum at topica.com
> Subject: Let's try this ONE more time 
> Sent: 5 jul 2004 y. 22:51
>
> posted by: Al Getz <Xaxo at aol.com>
>
> For people that cant read English,
If people cant read English, then any new
your attempt to bring the truth in the last
instance in English looks like crazy next one.
Da, ne tak li? --> Yes, isn't it?
Try another language please, great Mumbo Jumbo
can be good ... I think ... I do not know ...
Russian is good, Juergen loves German, I think ...
What is your favourite second language?
> i post this one more time:
>
> constant n=#10000
>
> function Rand32()
> --no reason why you cant make this more inline...
> atom a,b,c
>
> a=rand(n)-1
> b=rand(n)-1
> c=a*n+b
> return c
> end function
> Notes:
> 1. For two added rnd numbers the distribution
> should be the same.
Sum of two random numbers with same flat distribution
gives the number with triangle distribution
(Simpson's distribution).
What do you mean under "two added rnd numbers" ?
> It's multiplication that changes the distribution.
Yes, multiplication changes the distribution too.
> 2. Two rand sequences from the same generator that
> show no correlation, when added, form a new rand
> sequence that will show no correlation to another
> rand sequence generated in exactly the same way.
Correlation and distribution are the different things.
We need the flat distribution in [0,#FFFFFFFF] range,
as far as I can see in the first Juergen's post.
> 3. Multiplication, such as rand1*rand2
> changes the distribution greatly.
Yes, changes.
> 4. Note that to generate 'a' above you cant use rand(n).
Shoot me, I can not understand this your note ...
> Take care,
> Al
>
> And, good luck with your Euphoria programming!
Good Luck!
Regards,
Igor Kachan
kinz at peterlink.ru
7. Re: Let's try this ONE more time :-)
Derek Parnell wrote:
> Al Getz wrote:
>>
>> Pete Lomax wrote:
>>>
>>> On Mon, 05 Jul 2004 11:51:21 -0700, Al Getz <guest at RapidEuphoria.com>
>>> wrote:
>>>
>>> For people that cant read English, i post this one more time:
>>> And people say I need to try decaffeinated. )
>>>
>>>
>> There were at least TWO (count em, 2> ) posts suggesting
>> a method of generating 32 bit random numbers which were
>> perfectly fine, and yet people were still bickering over
>> what to do about generating a 32 bit rand number ??
>
> Bickering? Suggesting alternates is normal in programming. Why do you
> think this is evidence of bickering?
>
> There are a huge number of ways to generate a 32-bit random number. So
> what's the problem if people suggest some alternatives?
[big snip]
Derek, thank you for saying this clearly!
Yes, suggesting alternates is normal when you think in a problem
oriented manner. This way of discussion means a spirit of freedom of
thinking, of willingness to share knowledge, a spirit of evolution and
so on. I always learn a lot when reading or participating in a good
discussion. Who wants to prevent me (or someone else) from learning?
However, if someone thinks more in a vanity oriented manner (e.g.
thinking that he is the most intelligent person on this planet), it's
likely that he mistakes "discussion" for "bickering".
Regards,
Juergen
8. Re: Let's try this ONE more time :-)
Hello again,
I hope i can respond to some posts before the list 'scrolls' out
of view. That's going to be a problem sometimes. I cant always get
back exactly the same (or next) day.
By the way, anyone know the period of the current Euphoria random
number generator?
Pete:
Maybe some people need more coffee
Derek:
Using 'cant' is an experiment of sorts, to find out if
the apostrophe is really needed for good communication.
No one has EVER said they couldnt understand what 'cant'
meant, over YEARS of emails and message boards. True
this isnt the final word on 'cant' but it's good enough
for me
I'm fully aware about the distribution problems that can
come up, and i was the first one to mention this some
years ago here, which is why i suggested again
rnd=(rand(n)-1)*n+rand(n)-1
or the alternate form
rnd=(rand(n)-1)*n+rand(n-1)
depending on what range you need.
And what i meant by 'add' was of the form
a*n+b
not simply add two rand numbers.
Also, right after posting that proposed solution somebody else
posted the same formula he he.
BTW i didnt think yours was bad or anything, except for the rand(n) of
the first part, which of course doesnt work.
Igor:
My second language is Spanish.
As i was saying, when i said 'add' it was in the context where i
assumed a*n+b not simply a+b. As you noted a*b doesnt work.
If you have flat distrib with rand(n), then you have flat distrib with
(rand(n)-1)*n+rand(n)-1 as well. This is true because rand(n) is
assumed truely random. If not, then going to 32 bits isnt going
to change it that much unless it's real bad to start with.
Yes, as someone else pointed out, in the prng sequence x(n+1)=f(x(n))
but f(x(n)) is probably remote enough from x(n) to be still considered
random, for all but the most demanding tasks, which brings us to the
second post, which mentioned a file in the archives. In other words,
if you can get by with rand(n) then you can probably get by with the
additive (concatenated) form.
Lastly, when i said
"note you cant use rand(n) to generate 'a' above",
'a' is the first random part of the whole 32 bit number, and 'b' is the
second part. You cant use a=rand(n) because Euphoria's
rand(n) function starts at one, not zero, so you would never get
a random 32 bit number less then hex #10000 (funny).
Juergen:
You'll have to clear up your point. You're deduction sounds incorrect.
If you're talking about me, im lost without my spell checker
Christian:
I dont see how you get 2^16 possibility from two concatenated 2^16 random
sequences. Is this what you meant? Please explain how you arrived at
far less then 2^32 ...
All:
I'll post this again, so if anyone thinks it doesnt work, speak now...
n=#1000
r32=(rand(n)-1)*n+rand(n)-1
I could be wrong, so show me
Take care,
Al
And, good luck with your random number programming,
sounds like you're gonna need it
My bumper sticker: "I brake for LED's"
9. Re: Let's try this ONE more time :-)
Al Getz wrote:
> By the way, anyone know the period of the current Euphoria random
> number generator?
I believe it's something like power(2, 61).
If you had been computing one rand() per second
since the "Big Bang", you might be getting close
to repeating by now.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
10. Re: Let's try this ONE more time :-)
Robert Craig wrote:
>
> Al Getz wrote:
> > By the way, anyone know the period of the current Euphoria random
> > number generator?
>
> I believe it's something like power(2, 61).
>
> If you had been computing one rand() per second
> since the "Big Bang", you might be getting close
> to repeating by now.
>
> Regards,
> Rob Craig
> Rapid Deployment Software
> <a href="http://www.RapidEuphoria.com">http://www.RapidEuphoria.com</a>
>
Wow, that sounds a little too good
Any way you know of to test that...
Take care,
Al
And, good luck with your Euphoria programming!
My bumper sticker: "I brake for LED's"
11. Re: Let's try this ONE more time :-)
Al Getz wrote:
>
> Robert Craig wrote:
> >
> > Al Getz wrote:
> > > By the way, anyone know the period of the current Euphoria random
> > > number generator?
> >
> > I believe it's something like power(2, 61).
> >
> > If you had been computing one rand() per second
> > since the "Big Bang", you might be getting close
> > to repeating by now.
>
> Wow, that sounds a little too good
> Any way you know of to test that...
No.
Even at a billion/sec it should take years to repeat.
I got that statistic from some notes I wrote several years
ago when I implemented the algorithm. The algorithm
was on a university web site somewhere.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
12. Re: Let's try this ONE more time :-)
Rob,
Ok thanks.
Take care,
Al
And, good luck with your Euphoria programming!
My bumper sticker:
"I brake for LED's", but not for Euphoria programmers"
hee hee, kidding
