Success again?!?

ATH · 2011-09-21, 20:42

Just popped up on top producers page at #2:

http://www.mersenne.org/report_top_500_LL/

2 ANONYMOUS 459005.537 5028 1 |

Prime95 · 2011-09-21, 21:16

Very observant, but this is highly likely a false alarm. The error code is *real* ugly. This computer has never reported a result before. I have no email for ANONYMOUS, so I cannot request the last save file. If he emails us, I can request the save file to do the quick verification. Until then, I'm far from hopeful.

I think Jeff Gilchrist will run a low priority LL test. If not, I'll get a different volunteer. I'd like to avoid having several people run LL tests - a waste of resources.

Brain · 2011-09-21, 21:19

A lot of errors:

53593187 ANONYMOUS
00000000000000__ 0013174C 2011-09-21 18:45

Prime95 · 2011-09-21, 21:24

Nuts, one of curtisc's computers got the assigned the number. So Jeff is doing the first test on one of his fast PCs and curtisc's computer will be the double-check.

Please, no one else should test the number.

lycorn · 2011-09-21, 21:39

Is the residue all zeros, or like this?
http://www.mersenneforum.org/showpos...3&postcount=13
I guess it has to be all zeros, to be counted as a success.

ixfd64 · 2011-09-21, 21:44

George, do you know which version of Prime95 returned the result? If I remember correctly, versions prior to 23.5 could generate a false residue of zero.

Brain · 2011-09-21, 21:50

I will do some GPU TF on this exponent, just for fun. Up to 74 bits...

axn · 2011-09-21, 23:05

Quote:

Originally Posted by Prime95

I have no email for ANONYMOUS, so I cannot request the last save file. If he emails us, I can request the save file to do the quick verification.

Why not include the feature so that the client automatically reruns the last 5000 iterations before reporting the result?

Jeff Gilchrist · 2011-09-21, 23:48

Quote:

Originally Posted by Prime95

Nuts, one of curtisc's computers got the assigned the number. So Jeff is doing the first test on one of his fast PCs and curtisc's computer will be the double-check.

Please, no one else should test the number.

Should be interesting to see what is going on.

ixfd64: It is not 23.5, it is a fairly recent build.

Uncwilly · 2011-09-21, 23:54

Maybe one of the folks that CUDAlucas up and running can run a DC on this number.

Edit:
BTW: We are 5 days from surpassing the record for the longest gap between discoveries, during the GIMPS era.

davieddy · 2011-09-22, 00:10

Quote:

Originally Posted by Prime95

Very observant, but this is highly likely a false alarm. The error code is *real* ugly. This computer has never reported a result before. I have no email for ANONYMOUS, so I cannot request the last save file. If he emails us, I can request the save file to do the quick verification. Until then, I'm far from hopeful.

I think Jeff Gilchrist will run a low priority LL test. If not, I'll get a different volunteer. I'd like to avoid having several people run LL tests - a waste of resources.

Can't be another prime yet.
We're due for a big gap, and AFAIK you aren't on vacation

David

2011-09-21, 20:42	#1
ATH Einyen Dec 2003 Denmark 3,313 Posts	Success again?!? Just popped up on top producers page at #2: http://www.mersenne.org/report_top_500_LL/ 2 ANONYMOUS 459005.537 5028 1 \|

2011-09-21, 21:24	#4
Prime95 P90 years forever! Aug 2002 Yeehaw, FL 2⁴·17·29 Posts	Nuts, one of curtisc's computers got the assigned the number. So Jeff is doing the first test on one of his fast PCs and curtisc's computer will be the double-check. Please, no one else should test the number. Last fiddled with by Prime95 on 2011-09-21 at 21:25

2011-09-21, 21:39	#5
lycorn "GIMFS" Sep 2002 Oeiras, Portugal 1536₁₀ Posts	Is the residue all zeros, or like this? http://www.mersenneforum.org/showpos...3&postcount=13 I guess it has to be all zeros, to be counted as a success. Last fiddled with by lycorn on 2011-09-21 at 21:40

2011-09-21, 21:44	#6
ixfd64 Bemusing Prompter "Danny" Dec 2002 California 100110100010₂ Posts	George, do you know which version of Prime95 returned the result? If I remember correctly, versions prior to 23.5 could generate a false residue of zero. Last fiddled with by ixfd64 on 2011-09-21 at 21:44

2011-09-21, 21:50	#7
Brain Dec 2009 Peine, Germany 331 Posts	TF effort I will do some GPU TF on this exponent, just for fun. Up to 74 bits...

2011-09-21, 21:16	#2
Prime95 P90 years forever! Aug 2002 Yeehaw, FL 2⁴·17·29 Posts	Very observant, but this is highly likely a false alarm. The error code is real ugly. This computer has never reported a result before. I have no email for ANONYMOUS, so I cannot request the last save file. If he emails us, I can request the save file to do the quick verification. Until then, I'm far from hopeful. I think Jeff Gilchrist will run a low priority LL test. If not, I'll get a different volunteer. I'd like to avoid having several people run LL tests - a waste of resources.

2011-09-21, 21:19	#3
Brain Dec 2009 Peine, Germany 331₁₀ Posts	A lot of errors: 53593187 ANONYMOUS 00000000000000__ 0013174C 2011-09-21 18:45

2011-09-21, 23:54	#10
Uncwilly 6809 > 6502 """"""""""""""""""" Aug 2003 101×103 Posts 10535₁₀ Posts	Maybe one of the folks that CUDAlucas up and running can run a DC on this number. Edit: BTW: We are 5 days from surpassing the record for the longest gap between discoveries, during the GIMPS era. Last fiddled with by Uncwilly on 2011-09-21 at 23:58

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
NFS success probability in practice	paul0	Factoring	2	2015-02-23 03:55
Bad LL-D Success Rate	TheMawn	Data	14	2014-10-13 20:19
Another success (not)!	ATH	Data	133	2012-03-09 17:19
Success?... (M46 related)	lycorn	News	631	2010-05-26 10:33
Success	edorajh	Lounge	5	2004-08-15 16:51