mersenneforum.org  

Go Back   mersenneforum.org > Extra Stuff > Miscellaneous Math

Reply
 
Thread Tools
Old 2013-11-15, 16:30   #12
retina
Undefined
 
retina's Avatar
 
"The unspeakable one"
Jun 2006
My evil lair

5·1,223 Posts
Default

Quote:
Originally Posted by Mini-Geek View Post
It's far easier to trust a computer.
Sure. If the hackers/NSA/GCHQ haven't gotten their codez into it.
Quote:
Originally Posted by Mini-Geek View Post
Since you're already working from the prior knowledge that the factor exists, why not take both factors and multiply them together? That'd be a much less difficult feat.
The more difficult the task the better we become at doing the other tasks. Even so, this is more of a tedious task rather than a hard task.

I considered making him do it in hexadecimal just to show that he understood how hex works but somehow that seemed a little bit too cruel even for me.
retina is online now   Reply With Quote
Old 2013-11-15, 17:07   #13
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

11101001001002 Posts
Default

Quote:
Originally Posted by literka View Post
I don't want to ague with you about it, it is not a subject of this post. No statistics is made. I heard the story, but it is only a story, I am not sure that it is a true story. And the story is this:
Long time ago it was announced that a proof of Four Colors Theorem was found. It was presented in Finland and I talked to mathematicians, who were there. They did not accept this proof because of use of computers. They even told that the method was known long before. Presenters just wrote a program to verify what was known before.
This is everything I know about it. I heard this from a second hand so I cannot be sure that it is true.
A little knowledge is a dangerous thing. It is especially applicable in your case. You heard a second hand story, yet you presented the
"computer proof is not accepted by mathematicians" assertion as fact.

The story, as you heard is is half true. But you carefully omitted
relevant facts, probably out of ignorance rather than an attempt at
deceit. Despite this, you still made a bold, incorrect assertion.
R.D. Silverman is offline   Reply With Quote
Old 2013-11-15, 17:11   #14
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

1D2416 Posts
Default

Quote:
Originally Posted by Mini-Geek View Post
I'd be impressed if a "lesser minion" could do that longhand without making a mistake. Shoot, I'd be impressed if a mathematician could do that longhand without making a mistake (for that matter, just how did mathematicians, like Lucas with M127, do enormous calculations way-back-when?).
Hugh Williams did a thorough research of this very question.
Lucas used some 'templates' (which were basically analog mechanisms
for simulating binary arithmetic) to speed the arithmetic. Lucas did
the whole calculation in binary.

I heard Hugh give a talk on this. I believe he also put it in his book.
I would have to go look; I am in the office and the book is at home.
R.D. Silverman is offline   Reply With Quote
Old 2013-11-15, 18:14   #15
literka
 
literka's Avatar
 
Mar 2010

26×3 Posts
Default

Quote:
Originally Posted by R.D. Silverman View Post
A little knowledge is a dangerous thing. It is especially applicable in your case. You heard a second hand story, yet you presented the
"computer proof is not accepted by mathematicians" assertion as fact.

The story, as you heard is is half true. But you carefully omitted
relevant facts, probably out of ignorance rather than an attempt at
deceit. Despite this, you still made a bold, incorrect assertion.


At least I supported my statement with some facts. This what you wrote (your way of thinking) can be reversed for your case. You boldly write that my assertion is incorrect and you give no arguments. You presented a naked statement and you assume this as a fact.
Of course I am using a double standard comparing me with you (it is your double standard theory, which you applied before).
literka is offline   Reply With Quote
Old 2013-11-15, 18:57   #16
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

1D2416 Posts
Default

Quote:
Originally Posted by literka View Post
At least I supported my statement with some facts.
False. You presented no facts.
Quote:

This what you wrote (your way of thinking) can be reversed for your case.
This last sentence is gibberish. Your prose is on a level with your mathematics.

Quote:
You boldly write that my assertion is incorrect and you give no arguments.
You presented a naked statement and you assume this as a fact.
What I wrote is supported by an extensive publication history on the part
of the mathematical community in which computer proof systems have been
developed, used, and discussed. It is only YOUR ignorance of the
literature which leads you to erroneously believe that there is no evidence.


I am part of the mathematical community. I talk
with other mathematicians all the time. I attend conferences where
people talk. I am asked to referee papers where people use computers
as part of their work.

I took part in the active discussion regarding Haken & Appel's proof when
it was first presented. Extensive discussion followed in the 80's over the
Internet. There was never a question of the validity of the proof method.
The only question was about the accuracy of the code that was used and
of the reliability of the computation. Doubts were dispelled when others
performed an independent verification using other computers and source
code.

Last fiddled with by R.D. Silverman on 2013-11-15 at 18:58 Reason: typos
R.D. Silverman is offline   Reply With Quote
Old 2013-11-15, 19:09   #17
Batalov
 
Batalov's Avatar
 
"Serge"
Mar 2008
Phi(4,2^7658614+1)/2

938910 Posts
Default

literka,

Could you please present the number of operations needed to carry on all your calculation (1) - (5), etc, compared to a simple multiplication of two factors.

It appears that to assert these:
Code:
Several equalities will be needed:

(1)                          p=208648999^2+126945596^2 = 512 * q+1

(2)                          s = 208648999*52542249 + 126945596*31967597

(3)                          126945596*52542249 - 208648999*31967597 = 1

(4)                         309*q + r = 2^55

(5)                         r = (s div 512)+1
one will have to do much more than to simply assert
Code:
59649589127497217 *
5704689200685129054721
= 340282366920938463463374607431768211457 
= 2^128+1
And even less work in octal (with an octal multiplication table at your side)
Code:
2^128+1 = 4000000000000000000000000000000000000000001
(note: this equality you will have for free!) 
and then

3237257607274243001
*
1152401672664431414535001 
= 
4000000000000000000000000000000000000000001

Last fiddled with by Batalov on 2013-11-16 at 02:49 Reason: 309*q + r = 2^55. Very important and extremely easy to verify! ;-)
Batalov is offline   Reply With Quote
Old 2013-11-15, 19:20   #18
literka
 
literka's Avatar
 
Mar 2010

19210 Posts
Default

Quote:
Originally Posted by R.D. Silverman View Post
This last sentence is gibberish. Your prose is on a level with your mathematics.



What I wrote is supported by an extensive publication history on the part
of the mathematical community in which computer proof systems have been
developed, used, and discussed. .

I did not see anything you did in mathematics. At least I did not see any announcement of your achievements in this forum, while I presented new results from time to time. I read your book, but only to check the level of your mathematics.
I am not surprised that people of a community, in which computer proof systems have been developed, used, and discussed, think that computer based proofs must be accepted in mathematics.
Frankly saying I have enough to see your insults. Take care of yourself and don't write in my threads.
literka is offline   Reply With Quote
Old 2013-11-15, 19:27   #19
wblipp
 
wblipp's Avatar
 
"William"
May 2003
New Haven

93916 Posts
Default

Quote:
Originally Posted by literka View Post
I did not see anything you did in mathematics
Google is your friend.

Bob has an extensive publication history, much of it relevant to the factoring topics commonly discussed here. Bob is even mentioned in the Cunningham Book.

Moved to Misc Math.
wblipp is offline   Reply With Quote
Old 2013-11-15, 19:33   #20
literka
 
literka's Avatar
 
Mar 2010

26×3 Posts
Default

Quote:
Originally Posted by Batalov View Post
literka,

Could you please present the number of operations needed to carry on all your calculation (1) - (5), etc, compared to a simple multiplication of two factors.

It appears that to assert these:
Code:
Several equalities will be needed:

(1)                          p=2086489992+1269455962 = 512 * q+1

(2)                          s = 208648999*52542249 + 126945596*31967597

(3)                          126945596*52542249 - 208648999*31967597 = 1

(4)                         309*q + r = 255

(5)                         r = (s div 512)+1
one will have to do much more than to simply assert
Code:
59649589127497217 *
5704689200685129054721
= 340282366920938463463374607431768211457 
= 2^128+1
And even less work in octal (with an octal multiplication table at your side)
Code:
2^128+1 = 4000000000000000000000000000000000000000001
(note: this equality you will have for free!) 
and then

3237257607274243001
*
1152401672664431414535001 
= 
4000000000000000000000000000000000000000001



Go to the page of Wikipedia, where the proof that 641 is a factor of F5. Compute the number of operations in this proof and number number of operations to multiply 641 by 6700417. Page of Wikipedia is not about F5 but about all Fermat numbers. Still, they thought that it is worthy to include proof about the number 641. So, if you have complains, write to Wikipedia first.
And yes, I have new ideas to find proofs corresponding to Fermat numbers with larger indexes.
literka is offline   Reply With Quote
Old 2013-11-15, 19:40   #21
literka
 
literka's Avatar
 
Mar 2010

3008 Posts
Default

Quote:
Originally Posted by wblipp View Post
Google is your friend.

Bob has an extensive publication history, much of it relevant to the factoring topics commonly discussed here. Bob is even mentioned in the Cunningham Book.

Moved to Misc Math.


I understand that you care that nobody should write insults. Silverman wrote that I am crank after I wrote my first post here. I never exceed level of insulting Silverman used with respect to me.
literka is offline   Reply With Quote
Old 2013-11-15, 20:07   #22
Batalov
 
Batalov's Avatar
 
"Serge"
Mar 2008
Phi(4,2^7658614+1)/2

41·229 Posts
Default

Quote:
Originally Posted by literka View Post
Page of Wikipedia is not about F5 but about all Fermat numbers.
Exactly.

Quote:
Originally Posted by literka View Post
Still, they thought that it is worthy to include proof about the number 641.
What proof? What are you talking about?
Batalov is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
On Fermat's Last Number c10ck3r Miscellaneous Math 14 2012-11-29 20:36
Fermat number F6=18446744073709551617 is a composite number. Proof. literka Factoring 5 2012-01-30 12:28
Fermat number and Modulo for searching divisors CyD Factoring 4 2011-05-31 11:24
Fermat number factors Citrix Math 35 2007-01-23 23:17
New Fermat number divisor! ET_ Factoring 1 2004-10-08 03:34

All times are UTC. The time now is 08:18.

Tue Apr 13 08:18:23 UTC 2021 up 5 days, 2:59, 1 user, load averages: 2.14, 1.65, 1.52

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.