mersenneforum.org > News Holy new Mersenne prime, Batman! (M47 related)
 Register FAQ Search Today's Posts Mark Forums Read

 2008-09-27, 13:01 #815 davieddy     "Lucan" Dec 2006 England 6,451 Posts Another analogy The "expected" ratio of successive exponents is conjectured to be ~1.47. The probability of having to wait that long is 1/e So there is a 50% probability of finding another Mersenne prime before the exponent has inceased by a ratio of ~1.3 (exact value is a not difficult exercise left to the reader) Since M40, GIMPS has enjoyed a run of 6 "heads" in a row. I think that puts an accurate complexion on RDS' "very very very very lucky streak" comment. David
 2008-09-27, 17:51 #816 drake2   Aug 2005 24 Posts Press writers mathematical prowess. Anyone notice the Associated Press and Fox News changed the number of digits in the 45th known Mersenne prime to 13,000,000 from 12,978,189 digits. They also called it the "46th known Mersenne prime" and wrote 2P-1 instead of (2^P)-1 . Also, United Press International has the headline "46th Prime Number". I believe Mr. Silverman may be failing the world at his crusade on mathematical preciseness. Oh well, keep it up and maybe math education will improve some time in the future.
2008-09-27, 18:01   #817
Brian-E

"Brian"
Jul 2007
The Netherlands

5·653 Posts

Quote:
 Originally Posted by davieddy The "expected" ratio of successive exponents is conjectured to be ~1.47. The probability of having to wait that long is 1/e So there is a 50% probability of finding another Mersenne prime before the exponent has inceased by a ratio of ~1.3 (exact value is a not difficult exercise left to the reader) Since M40, GIMPS has enjoyed a run of 6 "heads" in a row. I think that puts an accurate complexion on RDS' "very very very very lucky streak" comment. David
Perhaps you should say 6 "heads" in a row using a slightly weighted coin which rather more frequently lands as "tails". It's definitely fairly surprising if we believe the conjectured distribution of the Mersenne primes and perhaps the recent lucky streak gives the GIMPS project some further genuine value in that any further finds in the coming years might start raising some serious questions about the truth of the conjectured distribution.

 2008-09-27, 18:05 #818 ixfd64 Bemusing Prompter     "Danny" Dec 2002 California 7×331 Posts It took some time, but we finally got mentioned by the AP! http://ap.google.com/article/ALeqM5j...LCJdwD93F5DT01
2008-09-27, 19:18   #819
davieddy

"Lucan"
Dec 2006
England

6,451 Posts

Quote:
 Originally Posted by Brian-E Perhaps you should say 6 "heads" in a row using a slightly weighted coin which rather more frequently lands as "tails".
I left the precise "50% tipping ratio" as an exercise for the reader
with the sole intention of using an unbiassed coin

Last fiddled with by davieddy on 2008-09-27 at 19:18

 2008-09-27, 21:59 #820 philmoore     "Phil" Sep 2002 Tracktown, U.S.A. 22·32·31 Posts Our local paper, The Register-Guard of Eugene, Oregon picked up the L.A. Times article on A.P: http://hosted.ap.org/dynamic/stories...MPLATE=DEFAULT although the summary in the print version of the paper was condensed a bit from this version.
2008-09-27, 23:11   #821

"Richard B. Woods"
Aug 2002
Wisconsin USA

22·3·599 Posts

Quote:
 Originally Posted by philmoore Our local paper, The Register-Guard of Eugene, Oregon picked up the L.A. Times article on A.P: http://hosted.ap.org/dynamic/stories...MPLATE=DEFAULT
Yep, listed in the science section right under CERN's melting magnets: http://hosted.ap.org/dynamic/fronts/...G&SECTION=HOME

Check out the next article, too: a new distributed computing project ... but only for laptops that have built-in accelerometers. Take a guess before you look!

http://hosted.ap.org/dynamic/stories...09-25-17-46-38

Last fiddled with by cheesehead on 2008-09-27 at 23:18

2008-09-27, 23:24   #822
jinydu

Dec 2003
Hopefully Near M48

2×3×293 Posts

Front page of yahoo.com

Quote:
 Originally Posted by http://news.yahoo.com/s/ap/20080927/ap_on_fe_st/prime_number_prize LOS ANGELES - Mathematicians at UCLA have discovered a 13-million-digit prime number, a long-sought milestone that makes them eligible for a $100,000 prize. PAID POLITICAL ADVERTISEMENT The group found the 46th known Mersenne prime last month on a network of 75 computers running Windows XP. The number was verified by a different computer system running a different algorithm. "We're delighted," said UCLA's Edson Smith, the leader of the effort. "Now we're looking for the next one, despite the odds." It's the eighth Mersenne prime discovered at UCLA. Primes are numbers like three, seven and 11 that are divisible by only two whole positive numbers: themselves and one. Mersenne primes — named for their discoverer, 17th century French mathematician Marin Mersenne — are expressed as 2P-1, or two to the power of "P" minus one. P is itself a prime number. For the new prime, P is 43,112,609. Thousands of people around the world have been participating in the Great Internet Mersenne Prime Search, or GIMPS, a cooperative system in which underused computing power is harnessed to perform the calculations needed to find and verify Mersenne primes. The$100,000 prize is being offered by the Electronic Frontier Foundation for finding the first Mersenne prime with more than 10 million digits. The foundation supports individual rights on the Internet and set up the prime number prize to promote cooperative computing using the Web. The prize could be awarded when the new prime is published, probably next year. ___ Information from: Los Angeles Times, http://www.latimes.com/
Attached Thumbnails

Last fiddled with by jinydu on 2008-09-27 at 23:36

2008-09-27, 23:29   #823
jinydu

Dec 2003
Hopefully Near M48

2×3×293 Posts

and msn.com

Quote:
 Originally Posted by http://www.msnbc.msn.com/id/26914730/?GT1=43001 LOS ANGELES - Mathematicians at UCLA have discovered a 13-million-digit prime number, a long-sought milestone that makes them eligible for a $100,000 prize. The group found the 46th known Mersenne prime last month on a network of 75 computers running Windows XP. The number was verified by a different computer system running a different algorithm. "We're delighted," said UCLA's Edson Smith, the leader of the effort. "Now we're looking for the next one, despite the odds." Story continues below ↓advertisement It's the eighth Mersenne prime discovered at UCLA. Primes are numbers like three, seven and 11 that are divisible by only two whole positive numbers: themselves and one. Mersenne primes — named for their discoverer, 17th-century French mathematician Marin Mersenne — are expressed as 2P-1, or two to the power of "P" minus one. P is itself a prime number. For the new prime, P is 43,112,609. Thousands of people around the world have been participating in the Great Internet Mersenne Prime Search, or GIMPS, a cooperative system in which underused computing power is harnessed to perform the calculations needed to find and verify Mersenne primes. Click for related content Math wizards map 248-dimensional object Cosmic Log: Math classic jumps off the page Russian refuses to accept math prize The$100,000 prize is being offered by the Electronic Frontier Foundation for finding the first Mersenne prime with more than 10 million digits. The foundation supports individual rights on the Internet and set up the prime number prize to promote cooperative computing using the Web. Large prime numbers serve as the foundation for the cryptological techniques used to keep online transactions and coded communications secure — although the current generation of crypto codes doesn't require numbers quite as big as the one discovered by UCLA. The Electronic Frontier Foundation's prize could be awarded to the researchers when the new prime is published, probably next year.
Attached Thumbnails

Last fiddled with by jinydu on 2008-09-27 at 23:34

 2008-09-28, 00:14 #824 jrk     May 2008 21058 Posts "2P-1" They likely copy&pasted it from some text that used super script and lost the formatting. That doesn't excuse the mistake, though. You ever wonder just how often the news is wrong about things you aren't familiar with (so wouldn't know it is wrong), because you see how often they are wrong on things that you do know something about?
 2008-09-28, 01:22 #825 ADBjester   Aug 2002 111102 Posts Foxnews.com http://www.foxnews.com/story/0,2933,429201,00.html (AP regurgitation, with apologies to Ernst)

 Similar Threads Thread Thread Starter Forum Replies Last Post T.Rex Wagstaff PRP Search 6 2019-11-23 22:46 R.D. Silverman NFSNET Discussion 4 2008-10-02 01:28 ewmayer Science & Technology 4 2008-03-14 19:19 ixfd64 News 265 2006-01-04 09:47 adpowers Lounge 40 2004-08-12 22:05

All times are UTC. The time now is 13:16.

Tue Nov 24 13:16:47 UTC 2020 up 75 days, 10:27, 4 users, load averages: 1.14, 1.38, 1.66