Global Consciousness Project

[T'ai Chi]

Bumpety-bump.

Yeah... I'd say I already bumped it by posting to the thread before you did. Wouldn't you agree?

 

[CFLarsen]

Yeah... I'd say I already bumped it by posting to the thread before you did. Wouldn't you agree?

Look at the time stamp, Justin. We posted within the same minute. You simply pressed the button a few seconds before me.

 

[T'ai Chi]

Look at the time stamp, Justin. We posted within the same minute. You simply pressed the button a few seconds before me.

Glad you agree.

 

[Donks]

That is understood, but you are not understanding me. I am asking you for an official site that says that. An official DIEHARD site, for example, or some writing from the creator of the DIEHARD tests. You know, some official verification that that is indeed the case.[/B]

Just so the discussion can move on, from the file "diehard.doc" that comes with the DIEHARD CD:

2. Providing Input Files For DIEHARD.
The executable file diehard.exe will prompt you to name the file you want to be tested and ask you to select any or all of the 15 tests. You must provide the binary file that DIEHARD expects---a file of 10 to 11 megabytes, that is, a file of at least 80 million bits. DIEHARD will do the best it can with shorter binary files, then give up on the test under way, give an END OF FILE message and go to the next test.

 

[CFLarsen]

a file of 10 to 11 megabytes, that is, a file of at least 80 million bits

Thank you.

Justin,

Do you admit that the Orion RNG has not passed the DIEHARD test? Just yes or no.

 

[T'ai Chi]

Just so the discussion can move on, from the file "diehard.doc" that comes with the DIEHARD CD:

Cool, thanks Donks.

I've emailed someone in the GCP who might know more about the Orion.

"DIEHARD will do the best it can with shorter binary files"

might explain it, or there might be another explanation out there.

 

[CFLarsen]

Cool, thanks Donks.

I've emailed someone in the GCP who might know more about the Orion.

"DIEHARD will do the best it can with shorter binary files"

might explain it, or there might be another explanation out there.

But to pass the test, you need 10 Mb, yes or no?

 

[T'ai Chi]

Do you admit that the Orion RNG has not passed the DIEHARD test? Just yes or no.

Claus

Please don't attempt to limit my response to a question.

The Orion page says it has passed the tests. A text file that comes with the DIEHARD doc says that

"DIEHARD will do the best it can with shorter binary files"

so the Orion could have indeed passed the DIEHARD tests with a shorter sequence.

As mentioned, the excerpt at the bottom of the Orion page could have been referring to a different test as well. We (none of us) really know the story. Why don't you email them and ask?

 

[CFLarsen]

Claus

Please don't attempt to limit my response to a question.

The Orion page says it has passed the tests. A text file that comes with the DIEHARD doc says that

"DIEHARD will do the best it can with shorter binary files"

so the Orion could have indeed passed the DIEHARD tests with a shorter sequence.

The DIEHARD document is crystal clear:

You must provide the binary file that DIEHARD expects---a file of 10 to 11 megabytes, that is, a file of at least 80 million bits.

This is also supported by the quote from the website I linked to.

As mentioned, the excerpt at the bottom of the Orion page could have been referring to a different test as well. We (none of us) really know the story. Why don't you email them and ask?

You are the one who claims that the Orion RNGs passed the DIEHARD test. You do your own homework, buddy.

 

[T'ai Chi]

You are the one who claims that the Orion RNGs passed the DIEHARD test. You do your own homework, buddy.

I guess you don't really care to inquire, then.

 

[CFLarsen]

I guess you don't really care to inquire, then.

You are the one with the claims, Justin.

Until you get back with some evidence, let's sum up again:

You argued that the RNGs are calibrated. They are not.

You argued that the RNGs produce numbers that pass the DIEHARD test for non-randomness. They do not pass this test.

You argued that calibration means that they pass these tests. It does not.

 

[T'ai Chi]

You are the one with the claims, Justin.

You're a little confused here. I'm not claiming anything. I'm saying what the Orion page has on it.

You argued that the RNGs are calibrated. They are not.

If they pass the tests, they certainly are.

You argued that the RNGs produce numbers that pass the DIEHARD test for non-randomness. They do not pass this test.

Ahh, now we see a claim by you. Where is your actual evidence? Can you produce the output file? Have you emailed the Orion people to have them explain? NO!?? Why not? You'd have to do some of these, it seems, to be able to claim that "They do not pass this test". For example, they could have had a typo on their webpage, been talking about a different test(s), you could have misunderstood (impossible as that may seem!), or any number of possible explanations.

You argued that calibration means
that they pass these tests. It does not.

Yes, in RNG-land, to the best of my knowledge, it does.

 

[CFLarsen]

You're a little confused here. I'm not claiming anything. I'm saying what the Orion page has on it.

You are using the Orion page to claim that the RNGs are calibrated.

You are using the RNGs to claim that something is happening in GCP.

You are the one who is confused, Justin.

If they pass the tests, they certainly are.

They don't pass the tests, Justin. And they are not calibrated, because calibration does not mean that they pass a test. You are wrong on this.

Ahh, now we see a claim by you. Where is your actual evidence? Can you produce the output file? Have you emailed the Orion people to have them explain? NO!?? Why not? You'd have to do some of these, it seems, to be able to claim that "They do not pass this test". For example, they could have had a typo on their webpage, been talking about a different test(s), you could have misunderstood (impossible as that may seem!), or any number of possible explanations.

We have two separate sources that clearly states that there must be 10 MB. Orion only uses 2 Mb. You admit that they do, Justin.

Ergo, the Orion RNG has not passed the DIEHARD test.

Yes, in RNG-land, to the best of my knowledge, it does.

 

[T'ai Chi]

You are using the Orion page to claim that the RNGs are calibrated.

Then you are confused. That is not me making a claim then. It is them. I am just repeating what is on their page. I didn't make the claim.

You are using the RNGs to claim that something is happening in GCP.

Statistical significance, yes. Anything other "something" that you believe I am speaking about is your invention.

They don't pass the tests, Justin.

Got the actual test resuts to make such a claim? I'm all for you actually showing them to me, instead of asking me to take your words on faith.

Show me where "they" (ie. the RNG's used in the GCP, that is, the Orion, PEAR, and MicroREG) don't pass the DIEHARD tests.

We have two separate sources that clearly states that there must be 10 MB. Orion only uses 2 Mb. You admit that they do, Justin.

Claus, READ what the webpage actually says. INQUIRE about what it says. At the top of the page, it says the RNG passed the DIEHARD tests. OK. Further down the page, it talks about another test and using 2Mb. Here, however, it does not specifically mention that this was again the DIEHARD tests. You assume that. Moreover, without actually inquiring, you have no idea if there was a typo, if they were talking about another test or the DIEHARD tests here, or if you misunderstood something.

But don't let that stop your 'skeptical inquiry'..

 

[CFLarsen]

Justin,

Look, if you can't present a coherent argument on your own, then don't. But please don't expect me to do it for you, complete with sources, evidence and facts. You have to do that yourself.

If you can't do that, then don't enter debates at all. You lose one debate after another, Justin. You are simply not up to this.

You are free to dismiss the facts people here have unearthed about the Orion RNG. The Orion RNG has not passed the DIEHARD test, no matter how much you try to shift the burden of evidence onto others.

You are also free to continue to claim that the RNGs are calibrated. Each time you refer to GCP for whatever clouded reasons you may or may not have (it seems very hard for you to come out and state clearly what you mean about anything - do you have problems with the notion of being held accountable for your actions?), you immediately lose the argument.

Yet another one.

 

[T'ai Chi]

Claus,

Look, you just avoided the issues again.

1) you claimed the RNGs don't pass the tests.
a) show us the test resuts
b) for each RNG (Orion, MicroREG, and PEAR)
c) show us a statement from the makers

Choose one or more, just don't expect anyone to believe you without actual evidence.

But please don't expect me to do it for you, complete with sources, evidence and facts. You have to do that yourself.

You made a claim, you have the burden. You are not magically immune because you label yourself a Skeptic.

You are also free to continue to claim that the RNGs are calibrated.

All calibrations means is that they passed a RNG test. If you continue to claim they aren't, then you are basically just incorrect.

(it seems very hard for you to come out and state clearly what you mean about anything

You'll have to be more specific than that, Claus. I can't help it if you have issues with me saying there is statistical significance in the GCP. If you want me to say there is a global concsciousness field affecting everything, etc., you will be disappointed.

 

[CFLarsen]

 

[T'ai Chi]

More of the same evasions.

Claus,

1) you claimed the RNGs don't pass the tests.

a) show us the test resuts
b) for each RNG (Orion, MicroREG, and PEAR)
c) show us a statement from the makers

Choose one or more, just don't expect anyone to believe you without actual evidence.

You made a claim, you have the burden. You are not magically immune because you label yourself a Skeptic.

Are you now claiming that you didn't make the claim??

 

[T'ai Chi]

Claus, READ what the webpage actually says. INQUIRE about what it says. At the top of the page, it says the RNG passed the DIEHARD tests. OK. Further down the page, it talks about another test and using 2Mb. Here, however, it does not specifically mention that this was again the DIEHARD tests. You assume that. Moreover, without actually inquiring, you have no idea if there was a typo, if they were talking about another test or the DIEHARD tests here, or if you misunderstood something.

But don't let that stop your 'skeptical inquiry'.. [/B]

So, I got an email back from the contact on the Orion page.

Basically, Claus did jump to a conclusion without inquiry, in exactly the manner I suggested above. I should apply for the million.

I am awaiting approval to post the full email contents.

 

[T'ai Chi]

Here is is. Start from the bottom of the post and work up. I emailed the email address on http://www.randomnumbergenerator.nl/rng/home.html

----


And then I emailed him the specific thread (this one) where his additional information would be posted.


-----Original Message-----
Date: Mon, 14 Mar 2005 17:06:49 +0100
Subject: RE: FW: Orion RNG and DIEHARD test


Of course you may.
Which forum is this?

kind regards, Michiel Klonhammer


-----Original Message-----
Sent: maandag 14 maart 2005 17:03
Subject: Re: FW: Orion RNG and DIEHARD test

Hi again,

Additionally, could I have your permission to post the full contents of your email on an internet bulletin board where a discussion on RNG's is taking place?

Sincerely,

Justin


-----Original Message-----
Date: Mon, 14 Mar 2005 12:46:15 +0100
Subject: FW: Orion RNG and DIEHARD test


Dear Sir,
The design of the RNG has been tested according to the DIEHARD test. But EACH RNG is actually tested according tot the 8192 x 8 bit samples test. This is to ensure that each RNG is in perfect working order. It is not necessary to test eacht shipped RNG with the diehard test.

kind regards, Michiel Klonhammer


-----Original Message-----
Sent: zaterdag 12 maart 2005 22:16
Subject: Orion RNG and DIEHARD test

Hello,

I have a question I was hoping I could get some help with.

On http://www.randomnumbergenerator.nl/rng/home.html

it says that

"The Random Number Generator parallel port dongle(Mac/DOS/Win) is the first true RNG to pass Marsaglia's famous DIEHARD randomness test."

and

"Each RNG passes a 256 run random ness test before being shipped. Each run consists of 8192 8-bit samples."

I've heard, but don't know for sure, that the DIEHARD tests require 10,000,000 bytes (I can't confirm that from any official DIEHARD source however). The issue that is confusing me, is that the above info from the Orion page only yields 256*8192 = 2,097,152 bytes, which wouldn't be adequate for testing.

Any info is appreciated,
Sincerely,


Justin Smith