10^263-1 sieving - Page 6

Batalov · 2009-02-04, 21:43

I've finished my A's, and the R's will be done by tomorrow.

Xyzzy · 2009-02-05, 04:39

Quote:

…and xyzzy has finished 15-20…

The 15-20 "A" side has been sent along with checksum file. We think we have done all we have committed to do. If not let us know.

fivemack · 2009-02-05, 08:45

The 15-20 file arrived in perfect shape; there are a few sections of 20-25 left for me to complete - pesky people running actually-relevant jobs on the computers I was using - but I expect to make a matrix over the weekend.

Thanks very much to all the contributors!

Andi47 · 2009-02-05, 11:42

Quote:

Originally Posted by fivemack

The 15-20 file arrived in perfect shape; there are a few sections of 20-25 left for me to complete - pesky people running actually-relevant jobs on the computers I was using - but I expect to make a matrix over the weekend.

Thanks very much to all the contributors!

Have my files have properly arrived?

fivemack: Yes, they're all here, I've merged them and cut out the small duplicated sections.

fivemack · 2009-02-06, 11:50

The computer continues to do its almost perfect rock imitation with a new PSU. Bother. I've got an RMA for the motherboard, but it'll be a week or so before I get to know if the new motherboard helps.

(the motherboard is the next bit to replace, isn't it?)

Batalov · 2009-02-06, 17:37

Bother! You don't suspect the memory? If only you had a twin (or a cousin) computer to swap parts, one by one...
_________

"Books and bother killed my mother. Books and bother killed my mother. And my father, too." (c) The Element of Crime

fivemack · 2009-02-07, 08:43

The duplicate rate is very high, so the 233M raw relations that have usually sufficed have turned into 144M unique relations, which wasn't enough.

So, we need to sieve 100M-115M, both sides. Any offers?

J.F. · 2009-02-07, 08:55

Sorry, I can't help for the next few days. As soon as the cluster opens up, I'll check back here if the ranges still need to be done.

Xyzzy · 2009-02-08, 02:54

We have several jobs running on our (puny) "cluster" but we could run some work in a few days.

However, it takes ~10 days for us to burn through 5M on both sides. What time frame are you wanting results in?

fivemack · 2009-02-08, 10:27

Ten days would be fine - I've only got four cores on 100-102, so that range will take about ten days. If a bigger cluster comes on line then we can just search further in the same time.

Xyzzy · 2009-02-09, 03:30

We will start up 102 to 106 on both sides later tomorrow.

2009-02-06, 11:50	#60
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 2³·11·73 Posts	A short delay The computer continues to do its almost perfect rock imitation with a new PSU. Bother. I've got an RMA for the motherboard, but it'll be a week or so before I get to know if the new motherboard helps. (the motherboard is the next bit to replace, isn't it?)

2009-02-07, 08:43	#62
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 6424₁₀ Posts	Not quite enough relations The duplicate rate is very high, so the 233M raw relations that have usually sufficed have turned into 144M unique relations, which wasn't enough. So, we need to sieve 100M-115M, both sides. Any offers?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
NFS sieving?	Dubslow	Factoring	8	2012-09-28 06:47
Line sieving vs. lattice sieving	JHansen	NFSNET Discussion	9	2010-06-09 19:25
10^420 + 1 sieving	juno1369	Factoring	20	2010-04-28 01:11
Sieving	OmbooHankvald	Prime Sierpinski Project	4	2005-06-30 07:51
Sieving	robert44444uk	Sierpinski/Riesel Base 5	8	2005-04-02 22:30

2009-02-04, 21:43	#56
Batalov "Serge" Mar 2008 Phi(4,2^7658614+1)/2 9,497 Posts	I've finished my A's, and the R's will be done by tomorrow.

2009-02-05, 08:45	#58
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 2³·11·73 Posts	The 15-20 file arrived in perfect shape; there are a few sections of 20-25 left for me to complete - pesky people running actually-relevant jobs on the computers I was using - but I expect to make a matrix over the weekend. Thanks very much to all the contributors!

2009-02-06, 17:37	#61
Batalov "Serge" Mar 2008 Phi(4,2^7658614+1)/2 9,497 Posts	Bother! You don't suspect the memory? If only you had a twin (or a cousin) computer to swap parts, one by one... _________ "Books and bother killed my mother. Books and bother killed my mother. And my father, too." (c) The Element of Crime

2009-02-07, 08:55	#63
J.F. Jun 2008 2³×3² Posts	Sorry, I can't help for the next few days. As soon as the cluster opens up, I'll check back here if the ranges still need to be done.

2009-02-08, 02:54	#64
Xyzzy "Mike" Aug 2002 2035₁₆ Posts	We have several jobs running on our (puny) "cluster" but we could run some work in a few days. However, it takes ~10 days for us to burn through 5M on both sides. What time frame are you wanting results in?

2009-02-08, 10:27	#65
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 14430₈ Posts	Ten days would be fine - I've only got four cores on 100-102, so that range will take about ten days. If a bigger cluster comes on line then we can just search further in the same time.

2009-02-09, 03:30	#66
Xyzzy "Mike" Aug 2002 8245₁₀ Posts	We will start up 102 to 106 on both sides later tomorrow.