Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.physics > #528666 > unrolled thread

Better Randomness ? !

Back to article view | Back to sci.physics

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Better Randomness ? ! Jeff-Relf.Me <@.> - 2015-10-27 22:41 -0700
    Re: Better Randomness ? ! Julio Di Egidio <julio@diegidio.name> - 2015-10-27 22:46 -0700
    Re: Better Randomness ? ! Fabian Russell <root@localhost.localdomain> - 2015-10-28 06:50 +0000
      Re: The pseudo_random array is fully known, 100.0000% predictable. Fabian Russell <root@localhost.localdomain> - 2015-10-28 17:34 +0000
      Re: The pseudo_random array is fully known, 100.0000% predictable. benj <none@gmail.com> - 2015-10-30 05:48 -0400
        Re: The pseudo_random array is fully known, 100.0000% predictable. Fabian Russell <root@localhost.localdomain> - 2015-10-30 18:53 +0000
          Re: The pseudo_random array is fully known, 100.0000% predictable. noTthaTguY <abu.kuanysh05@gmail.com> - 2015-10-30 14:08 -0700
            Re: The pseudo_random array is fully known, 100.0000% predictable. Fabian Russell <root@localhost.localdomain> - 2015-10-30 22:36 +0000
              Re: The pseudo_random array is fully known, 100.0000% predictable. "reber g=emc^2" <herbertglazier0@gmail.com> - 2015-11-01 15:14 -0800
    Re: Better Randomness ? ! moroney@world.std.spaamtrap.com (Michael Moroney) - 2015-10-28 14:33 +0000
      Re: Better Randomness ? ! chrisv <chrisv@nospam.invalid> - 2015-10-28 09:36 -0500
        Re: Better Randomness ? ! benj <none@gmail.com> - 2015-10-30 05:56 -0400
      Re: Better Randomness ? ! "Ezekiel" <zeke@nosuchemail.com> - 2015-10-28 10:46 -0400
      QueryPerformanceCounter() -- Better Randomness. Jeff-Relf.Me <@.> - 2015-10-28 08:28 -0700
        Re: QueryPerformanceCounter() -- Better Randomness. "Ezekiel" <zeke@nosuchemail.com> - 2015-10-28 11:48 -0400
          RDRAND has numerous problems. Jeff-Relf.Me <@.> - 2015-10-28 09:35 -0700
        Re: QueryPerformanceCounter() -- Better Randomness. moroney@world.std.spaamtrap.com (Michael Moroney) - 2015-10-28 16:17 +0000
          QueryPerformanceCounter() -- Better Randomness. Jeff-Relf.Me <@.> - 2015-10-28 09:50 -0700
          Re: QueryPerformanceCounter() -- Better Randomness. Fabian Russell <root@localhost.localdomain> - 2015-10-28 17:44 +0000
            Re: QueryPerformanceCounter() -- Better Randomness. moroney@world.std.spaamtrap.com (Michael Moroney) - 2015-10-28 18:01 +0000
              Re: QueryPerformanceCounter() -- Better Randomness. Fabian Russell <root@localhost.localdomain> - 2015-10-28 18:39 +0000
      Re: Better Randomness ? ! Nomen Nescio <nobody@dizum.com> - 2015-10-29 19:38 +0100
        Re: Better Randomness ? ! Tim Streater <timstreater@greenbee.net> - 2015-10-29 19:14 +0000
          Re: Better Randomness ? ! Snit <usenet@gallopinginsanity.com> - 2015-10-29 13:10 -0700
        Re: Better Randomness ? ! Fabian Russell <root@localhost.localdomain> - 2015-10-29 21:36 +0000
          Re: Better Randomness ? ! noTthaTguY <abu.kuanysh05@gmail.com> - 2015-10-29 17:24 -0700
            Re: Better Randomness ? ! benj <nobody@gmail.com> - 2015-10-29 21:55 -0400
              Re: Better Randomness ? ! Fabian Russell <root@localhost.localdomain> - 2015-10-30 02:12 +0000
              Re: Better Randomness ? ! Fabian Russell <root@localhost.localdomain> - 2015-10-30 02:18 +0000
    Re: Better Randomness ? ! benj <none@gmail.com> - 2015-10-30 05:45 -0400

Page 1 of 2  [1] 2  Next page →

#528666 — Better Randomness ? !

Professor William_Unruh wrote:
> a program to efficiently find the minimum of an arbitrary function
> is incredibly powerful.
> 
> Now the question of course is whether or not the DWave system can
> actually find such minima faster than a classical computer can. 
> 
> That is still somewhat up in the air. 
> And whether their computer actually uses quantum mechanics
> in the process (rather than thermal monti-carlo say).

Better Randomness ? ! I don't think so.

Seriously, _How_ could thermo/quantum randomness be
better than what I've got, on my i7_4790s/Windows_8 PC ? !
How ? !

I used genetic algorithms back in 1992;
it's no big deal, nothing new.

"Randomness" is just ignorance, nothing more.
Intrinsically, the (4D) timescape is static, immutable.
"Life" is virtual, not real.

We're robots, programmed to want food, water, air, etc.

Like the sun, stars and everything else...
we consume "exergy" (energy that can do work);
i.e. we ratchet entropy.

[toc] | [next] | [standalone]

#528667

On Wednesday, October 28, 2015 at 5:41:35 AM UTC, Jeff-Relf.Me wrote:

> We're robots, programmed to want food, water, air, etc.

Feel free to be clueless, but speak for yourself.

Julio

[toc] | [prev] | [next] | [standalone]

#528672

On Tue, 27 Oct 2015 22:41:33 -0700, Jeff-Relf.Me wrote:

> 
> "Randomness" is just ignorance, nothing more.
>

Totally incorrect.

The currently accepted definition of randomness is Martin-Loef
randomness, which is a further elaboration of the Kolmogorov
idea of incompressibility.

A Martin-Loef random sequence is not-computable by any algorithmic
means.  Thus, Martin-Loef randomness is not attainable in practice.

In practice, it is sufficient for a sequence to "act randomly,"
and that means that the sequence is able to pass statistical
tests.

[toc] | [prev] | [next] | [standalone]

#528735 — Re: The pseudo_random array is fully known, 100.0000% predictable.

On Wed, 28 Oct 2015 00:22:46 -0700, Jeff-Relf.Me wrote:

> 
> The pseudo_random array is fully known, 100.0000% predictable.
>

Try your 100% certainty at the Las Vegas casinos.

[toc] | [prev] | [next] | [standalone]

#529102 — Re: The pseudo_random array is fully known, 100.0000% predictable.

On 10/28/2015 03:22 AM, Jeff-Relf.Me wrote:
>
> Replying to me, you (Fabian_Russell) wrote:
>> > "Randomness" is just ignorance, nothing more.
>>
>> Totally incorrect.
>>
>> The currently accepted definition of randomness is Martin-Loef
>> randomness, which is a further elaboration of the Kolmogorov
>> idea of incompressibility.
>>
>> A Martin-Loef random sequence is not-computable by any algorithmic
>> means.  Thus, Martin-Loef randomness is not attainable in practice.
>>
>> In practice, it is sufficient for a sequence to "act randomly,"
>> and that means that the sequence is able to pass statistical tests.
>
> The pseudo_random array is fully known, 100.0000% predictable.

But can be made to pass any statistical "random" test. Hence the 
definition of "passing statistical tests" to tell if something is 
"random" is obviously inadequate.



-- 

       ___           ___           ___            ___
      /\  \         /\  \         /\__\          /\  \
     /::\  \       /::\  \       /::|  |         \:\  \
    /:/\:\  \     /:/\:\  \     /:|:|  |     ___ /::\__\
   /::\~\:\__\   /::\~\:\  \   /:/|:|  |__  /\  /:/\/__/
  /:/\:\ \:|__| /:/\:\ \:\__\ /:/ |:| /\__\ \:\/:/  /
  \:\~\:\/:/  / \:\~\:\ \/__/ \/__|:|/:/  /  \::/  /
   \:\ \::/  /   \:\ \:\__\       |:/:/  /    \/__/
    \:\/:/  /     \:\ \/__/       |::/  /
     \::/__/       \:\__\         /:/  /
      ~~            \/__/         \/__/

[toc] | [prev] | [next] | [standalone]

#529183 — Re: The pseudo_random array is fully known, 100.0000% predictable.

On Fri, 30 Oct 2015 05:48:40 -0400, benj wrote:

> 
> Hence the 
> definition of "passing statistical tests" to tell if something is 
> "random" is obviously inadequate.
>

Inadequate?  Obviously???!!!

Well, it may be obvious to you, but to the informed world the "passing
of statistical test" forms (in part) the VERY DEFINITION of randomness
(the other parts being algorithmic incompressibility and unpredictability). 

As I mentioned earlier, the "pseudo" in a pseudo-random sequence arises
because it is generated by a computer algorithm, and this violates
another part of the current mathematical definition of randomness (i.e.
algorithmic incompressibility).  But a pseudo-random sequence passes all
statistical tests and is therefore, for all intents and purposes, random
to the core.

The posters in this thread, myself excepted, are more mystically rather
than scientifically oriented.  They seek some "philosopher's stone" of
randomness, or at least claim that such a "philosopher's stone" does
not exist.

But daily practice around the globe proves otherwise.

[toc] | [prev] | [next] | [standalone]

#529232 — Re: The pseudo_random array is fully known, 100.0000% predictable.

how do you call one of two players,
tossing three coins at once.  easier:
tossing two coins at once

[toc] | [prev] | [next] | [standalone]

#529275 — Re: The pseudo_random array is fully known, 100.0000% predictable.

On Fri, 30 Oct 2015 14:08:41 -0700, noTthaTguY wrote:

> how do you call one of two players,
> tossing three coins at once.  easier:
> tossing two coins at once

Two heads are better than one.

But three is perfection.

[toc] | [prev] | [next] | [standalone]

#529793 — Re: The pseudo_random array is fully known, 100.0000% predictable.

On Friday, October 30, 2015 at 3:37:33 PM UTC-7, Fabian Russell wrote:
> On Fri, 30 Oct 2015 14:08:41 -0700, noTthaTguY wrote:
> 
> > how do you call one of two players,
> > tossing three coins at once.  easier:
> > tossing two coins at once
> 
> Two heads are better than one.
> 
> But three is perfection.

Random can go with chaos. Relate to uncertainty principle Going in in diamensions 

[toc] | [prev] | [next] | [standalone]

#528705

Jeff-Relf.Me <@.> writes:

>Professor William_Unruh wrote:
>> a program to efficiently find the minimum of an arbitrary function
>> is incredibly powerful.
>> 
>> Now the question of course is whether or not the DWave system can
>> actually find such minima faster than a classical computer can. 
>> 
>> That is still somewhat up in the air. 
>> And whether their computer actually uses quantum mechanics
>> in the process (rather than thermal monti-carlo say).

>Better Randomness ? ! I don't think so.

>Seriously, _How_ could thermo/quantum randomness be
>better than what I've got, on my i7_4790s/Windows_8 PC ? !
>How ? !

Any computer algorithm is just pseudorandomness, "ignorance" as you put 
it.  Dig deep enough into the algorithm and you can _predict_ the next
value, although doing so can be extremely difficult.

On the other hand, something based on truly random events, such as 
radioactive decay, is, well, truly random, if implemented properly.

<snip crap>

[toc] | [prev] | [next] | [standalone]

#528706

Michael Moroney wrote:

>Jeff-Relf.Me <@.> writes:

*plonk*

[toc] | [prev] | [next] | [standalone]

#529104

On 10/28/2015 10:36 AM, chrisv wrote:
> Michael Moroney wrote:
>
>> Jeff-Relf.Me <@.> writes:
>
> *plonk*
>
excellent choice! Rolf gets it all the time.

-- 

       ___           ___           ___            ___
      /\  \         /\  \         /\__\          /\  \
     /::\  \       /::\  \       /::|  |         \:\  \
    /:/\:\  \     /:/\:\  \     /:|:|  |     ___ /::\__\
   /::\~\:\__\   /::\~\:\  \   /:/|:|  |__  /\  /:/\/__/
  /:/\:\ \:|__| /:/\:\ \:\__\ /:/ |:| /\__\ \:\/:/  /
  \:\~\:\/:/  / \:\~\:\ \/__/ \/__|:|/:/  /  \::/  /
   \:\ \::/  /   \:\ \:\__\       |:/:/  /    \/__/
    \:\/:/  /     \:\ \/__/       |::/  /
     \::/__/       \:\__\         /:/  /
      ~~            \/__/         \/__/

[toc] | [prev] | [next] | [standalone]

#528709

"Michael Moroney" <moroney@world.std.spaamtrap.com> wrote in message 
news:n0qmbc$8pt$1@pcls7.std.com...
> Jeff-Relf.Me <@.> writes:
>
>>Professor William_Unruh wrote:
>>> a program to efficiently find the minimum of an arbitrary function
>>> is incredibly powerful.
>>>
>>> Now the question of course is whether or not the DWave system can
>>> actually find such minima faster than a classical computer can.
>>>
>>> That is still somewhat up in the air.
>>> And whether their computer actually uses quantum mechanics
>>> in the process (rather than thermal monti-carlo say).
>
>>Better Randomness ? ! I don't think so.
>
>>Seriously, _How_ could thermo/quantum randomness be
>>better than what I've got, on my i7_4790s/Windows_8 PC ? !
>>How ? !
>
> Any computer algorithm is just pseudorandomness, "ignorance" as you put
> it.  Dig deep enough into the algorithm and you can _predict_ the next
> value, although doing so can be extremely difficult.
>
> On the other hand, something based on truly random events, such as
> radioactive decay, is, well, truly random, if implemented properly.
>

Agreed. There isn't really a software substitute for external "random" 
entropy.

-- 
"Coding in C#, or any other object oriented language, is *not* programming. 
It is only arranging predetermined classes like a child will arrange toy 
blocks."

Fabian Russell
5 Aug 2015 <pan.2015.08.05.14.06.13@localhost.localdomain>

[toc] | [prev] | [next] | [standalone]

#528716 — QueryPerformanceCounter() -- Better Randomness.

Replying to me, you (Michael_Moroney) wrote:
> > Seriously, _How_ could thermo/quantum randomness be
> > better than what I've got, on my i7_4790s/Windows_8 PC ? !
> > How ? !
> 
> Any computer algorithm is just pseudorandomness, 
> "ignorance" as you put it.  
> 
> Dig deep enough into the algorithm and you can _predict_ the next
> value, although doing so can be extremely difficult.
> 
> On the other hand, something based on truly random events, such as 
> radioactive decay, is, well, truly random, if implemented properly.

Every second, a 64 bit counter on my 4 GigaHz PC,
QueryPerformanceCounter(), goes up by another 4 billion.

Use the low bits of that counter to seed srand(),
and you have a nice array of PSEUDO_random numbers;
no one could predict what's in the array.

How, pray tell, would that not be good enough ?

[toc] | [prev] | [next] | [standalone]

#528718 — Re: QueryPerformanceCounter() -- Better Randomness.

"Jeff-Relf.Me" <@.> wrote in message 
news:Jeff-Relf.Me@Oct.28{8.28A.Seattle.2015}...
> Replying to me, you (Michael_Moroney) wrote:
>> > Seriously, _How_ could thermo/quantum randomness be
>> > better than what I've got, on my i7_4790s/Windows_8 PC ? !
>> > How ? !
>>
>> Any computer algorithm is just pseudorandomness,
>> "ignorance" as you put it.
>>
>> Dig deep enough into the algorithm and you can _predict_ the next
>> value, although doing so can be extremely difficult.
>>
>> On the other hand, something based on truly random events, such as
>> radioactive decay, is, well, truly random, if implemented properly.
>
> Every second, a 64 bit counter on my 4 GigaHz PC,
> QueryPerformanceCounter(), goes up by another 4 billion.
>
> Use the low bits of that counter to seed srand(),
> and you have a nice array of PSEUDO_random numbers;
> no one could predict what's in the array.
>
> How, pray tell, would that not be good enough ?

You don't even need QueryPerformanceCounter - Intel added hardware support 
to make this even better.

<quote>
RDRAND (also RDRAND; previously known as Bull Mountain) is an instruction 
for returning random numbers from an Intel on-chip hardware random number 
generator. RDRAND is available in Ivy Bridge processors[a] and is part of 
the Intel 64 and IA-32 instruction set architectures. AMD added support for 
the instruction in June 2015.

The random number generator is compliant with security and cryptographic 
standards such as NIST SP 800-90A, FIPS 140-2, and ANSI X9.82. Intel also 
requested Cryptography Research Inc. to review the random number generator 
in 1999 and 2012, which resulted in two published papers: The Intel Random 
Number Generator in 1999, and Analysis of Intel's Ivy Bridge Digital Random 
Number Generator in 2012.
</quote>
https://en.wikipedia.org/wiki/RdRand


For just about everything (banking, ssh to work, etc) this is sufficient. 
But for extremely high security applications (NSA, military, etc) they want 
something much more random than this.

[toc] | [prev] | [next] | [standalone]

#528725 — RDRAND has numerous problems.

Ezekiel,  RDRAND has numerous problems;
if it worked properly, rand() would use it.

1. It doesn't work on most systems, including mine.
2. You can't guarantee that it'll work as advertized.

I don't see how passing the low bits of 
QueryPerformanceCounter() to srand() wouldn't
be good enough for anyone and everyone.

[toc] | [prev] | [next] | [standalone]

#528720 — Re: QueryPerformanceCounter() -- Better Randomness.

Jeff-Relf.Me <@.> writes:

>Replying to me, you (Michael_Moroney) wrote:
>> > Seriously, _How_ could thermo/quantum randomness be
>> > better than what I've got, on my i7_4790s/Windows_8 PC ? !
>> > How ? !
>> 
>> Any computer algorithm is just pseudorandomness, 
>> "ignorance" as you put it.  
>> 
>> Dig deep enough into the algorithm and you can _predict_ the next
>> value, although doing so can be extremely difficult.
>> 
>> On the other hand, something based on truly random events, such as 
>> radioactive decay, is, well, truly random, if implemented properly.

>Every second, a 64 bit counter on my 4 GigaHz PC,
>QueryPerformanceCounter(), goes up by another 4 billion.

>Use the low bits of that counter to seed srand(),
>and you have a nice array of PSEUDO_random numbers;

As I said, only pseudorandom, not genuinely random.

>no one could predict what's in the array.

>How, pray tell, would that not be good enough ?

"Good enough" depends on what you are up to.  Are you trying to generate
statistical noise for a test, or to encrypt something just so a casual
observer can't see a file/packet of not very significant importance?
Or are you trying to keep the NSA or some foreign government from ever
seeing the contents of a file?  Because if the NSA really wants to, 
they'll figure out your pseudorandom sequence and decrypt your file
just like that.

Remember, computers are deterministic.  I was just talking about someone
about using the computer's cycle counter to test memory latency in 
different configurations.  For a given configuration, a particular test
of his sees the cycle counter increment by the same amount, as long as 
interrupts, timer etc. are all locked out.

[toc] | [prev] | [next] | [standalone]

#528727 — QueryPerformanceCounter() -- Better Randomness.

Replying to me, you (Michael_Moroney) wrote:
> > Every second, a 64 bit counter on my 4 GigaHz PC,
> > QueryPerformanceCounter(), goes up by another 4 billion.
> > 
> > Use the low bits of that counter to seed srand(),
> > and you have a nice array of PSEUDO_random numbers;
> > no one could predict what's in the array.
> 
> if the NSA really wants to, they'll figure out your 
> pseudorandom sequence and decrypt your file just like that.

Proof ?  are you just making it up, on the fly ?

[toc] | [prev] | [next] | [standalone]

#528736 — Re: QueryPerformanceCounter() -- Better Randomness.

On Wed, 28 Oct 2015 16:17:26 +0000, Michael Moroney wrote:

> 
> As I said, only pseudorandom, not genuinely random.
> 

Fscking idiot.

They are pseudo-random because they are generated by an algorithm
(which therefore violates the mathematical definition of randomness).

But their statistical properties are genuinely random.

[toc] | [prev] | [next] | [standalone]

#528744 — Re: QueryPerformanceCounter() -- Better Randomness.

Fabian Russell <root@localhost.localdomain> writes:

>On Wed, 28 Oct 2015 16:17:26 +0000, Michael Moroney wrote:

>> As I said, only pseudorandom, not genuinely random.

>They are pseudo-random because they are generated by an algorithm
>(which therefore violates the mathematical definition of randomness).

>But their statistical properties are genuinely random.

The world of cryptography and related mathematics is much more 
sophisticated than that.

Someone from the NSA could explain it to you, but then they'd have to
kill you.

As I said, who are you trying to protect your data from? A casual packet
sniffer?  Or Mossad or MI6?

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | sci.physics

csiph-web