Generating a random noise file, a Linux example

"Anyone who considers arithmetical methods of producing random digits is, of course, in a state of sin." - John von Neumann

On a sunny day (Sun, 25 Sep 2011 19:14:54 GMT) it happened Chiron613 wrote in :

Interesting statement. I would think that ultimately most if not all systems can be mathematically described, so I think he was a bit dogmatic there. hehe :-)

described,

processing.

If you really want random noise, insult a couple dozen Mac users. ;-)

described,

If you don't know anything about electronics, you can try hacking bad dogrel.

John

data processing.

Ours are 2U rack mount devices which have NIST certifications, and are not in any way cheap.

Hey, this is electronics design, right? Has anyone ever hacked into an ionization smoke detector and picked off the noise from the ion detector thingie and injected it into a PSRG somewhere?

Supposedly, radioactive decay is virtually _guaranteed_ random, maybe even the standard for "randomness."

BTW, that damn cat must be dead by now!

Cheers! Rich

On a sunny day (Tue, 27 Sep 2011 14:59:46 -0700) it happened Rich Grise wrote in :

Well you gave the answer I did expect. You also show you do not follow physics.

I would have expected somebody to claim radioactive decay could be used to make a "pure random", note the quotes, stream. Bad luck for you, it was found radioactive decay depends (likely) on the neutrino flux from the sun, and shows a periodic change in sync with the earth position / rotation. Earlier papers [Russian] found a dependency on the orientation of the earth in the galactic plane. To translate that into your language: There is an AM modulation, so sidebands, that are very precisely known for radioactive decay. So much for random. This is what I referred to when replying to the other poster: Ultimately most processes can be mathematically described.

data processing.

The rng in NetHack works just fine. :-)

processing.

You mean PSEUDOrandom..

On a sunny day (Wed, 28 Sep 2011 20:30:31 -0800) it happened Robert Baer wrote in :

data processing.

Yea, OK.

data processing.

SSL random numbers are proved to be of cryptographic quality, though, so they're pretty much indistinguishable from real random numbers.

Also, of course, any finite-length string of true random numbers will have nonzero correlations, just because it's finite.

Cheers

Phil Hobbs

On a sunny day (Thu, 29 Sep 2011 10:19:43 -0400) it happened Phil Hobbs wrote in :

data processing.

No problem, just gzip an infinte string ;-)

So, are you saying that you've found a method for predicting when the next alpha particle is going to be emitted from the americium in the smoke detector?

Thanks, Rich

When will the next ball of tar drip out of the viscosimeter the professor placed it into back in 1927?

On a sunny day (Thu, 29 Sep 2011 12:53:24 -0700) it happened Rich Grise wrote in :

Statistically yes. :-)

Cheers! Rich

There's *good* doggerel?

mike

No, only great doggerel:

-- Les Cargill