Fast Breeding (guru management) - Page 20

swellman · 2015-09-08, 15:36

There are three 30-bit LP candidates in the xyyx project that have been ECM'd to t50 by yoyo@home but test sieving shows all three to be much slower to sieve than their SNFS difficulty would suggest.

C207_131_58
C196_129_62
C186_124_73

fivemack · 2015-09-08, 15:51

Quote:

Originally Posted by swellman

There are three 30-bit LP candidates in the xyyx project that have been ECM'd to t50 by yoyo@home but test sieving shows all three to be much slower to sieve than their SNFS difficulty would suggest.

C207_131_58
C196_129_62
C186_124_73

Slow-sieving candidates seem just the right things to put on NFS@home to keep the queues full - check sieving speeds with 31-bit LP and then enqueue them with appropriate large primes.

swellman · 2015-09-08, 16:25

I found two more slow-to-sieve xyyx numbers with (notionally) 30-bit LPs.

C175_142_39
C189_139_46

Both have been ECM'd to t50 - the first by yoyo@Home, the second by me.

wombatman · 2015-09-08, 17:32

I'll have a C184 from HP2(4496) soon as well. Running ECM up to 0.5t55.

debrouxl · 2015-09-08, 17:54

C207_131_58 is FF in FactorDB: http://factordb.com/index.php?id=1000000000044677155 .

I'll queue C196_129_62 as a 31-bit LPs task. t50 ECM work for SNFS difficulty 231 is a bit low, but shouldn't yield severe ECM misses.

Ben: GNFS 184 is clearly 15e territory, and we'll probably want at least t55 for that. Tom ?

The near-repdigit numbers are nearly untapped, and there shall be a XYYXF number followed by two Aliquot numbers, so we don't need a GCW number right now.

wombatman · 2015-09-08, 18:31

I based that 1/2t55 off of 2/7ths of 184 (52.5). I can do a full t55 if need be, but that'll take probably another week or so.

VictordeHolland · 2015-09-10, 12:41

We might need to think about adding some harder numbers (SNFS 240-250) to the 14e queue. The currently queued SNFS(220-230) don't keep the crunchers fed for long.

swellman · 2015-09-10, 13:25

C172_122_101 will be ECM'd to t55 NLT Sep 17. At SNFS of 245, it will help feed the sievers. For a few days anyway ;-)

C178_147_44 will be reach t55 within a week after that.

I believe C186_124_73, C175_142_39 and C189_139_46 are on deck to be queued, but they are only SNFS 23x. Carbs, no meat.

Then there are these.

wblipp · 2015-09-11, 22:23

(12491^61-1)/12490 is ready for SNFS thanks to ECM by Ryan Propper.

debrouxl · 2015-09-12, 08:01

I tried to sum up the (potential) reservations for NFS@Home's 14e in the latest several dozens of posts:

Will queue today:
12491_61_minus1 (ECM to 2/9 of SNFS difficulty by Ryan Propper)
C175_142_39 (ECM to t50 should be enough, SNFS difficulty ~227)

t55 in progress by Sean Wellman:
C172_122_101 (ETA September 17th)
C178_147_44 (ETA by September 24th)

Need another t50 because of SNFS difficulty 232: (I'll contribute some curves)
C196_129_62
C186_124_73
C189_139_46

Need t55, in slow progress mostly by Jon (jyb):
11^236+4^236 (100 curves at B1=26e7 by Lionel Debroux)
8^271-5^271
12^227-11^227

Additionally, there's a GNFS difficulty 184 from Home Primes sequence 4496, which has received 0.5t55. It's clearly food for 15e, but I don't manage that part of the queue. Tom, how much ECM work do you want on a GNFS 184 task ?

debrouxl · 2015-09-13, 09:07

Hmm. On my computer, trial sieving of the SNFS difficulty 226 C175_142_39 with 31-bit LPs shows the same sieving speed as SNFS difficulty 250 (!) 12491_61_minus1 with a sextic polynomial, whether a quintic or a sextic polynomial (they produce similar yields, unsurprisingly) is used for C175_142_39.
Therefore, it needs quite a bit more than the t50 ECM work indicated by the rule of thumb, in order to reduce the likelihood of an ECM miss. So far, I've done another 600 curves at B1=43e6, but t50 + t45 is a far cry from 3t50 or 4t50...

2015-09-08, 17:54	#214
debrouxl Sep 2009 977 Posts	C207_131_58 is FF in FactorDB: http://factordb.com/index.php?id=1000000000044677155 . I'll queue C196_129_62 as a 31-bit LPs task. t50 ECM work for SNFS difficulty 231 is a bit low, but shouldn't yield severe ECM misses. Ben: GNFS 184 is clearly 15e territory, and we'll probably want at least t55 for that. Tom ? The near-repdigit numbers are nearly untapped, and there shall be a XYYXF number followed by two Aliquot numbers, so we don't need a GCW number right now. Last fiddled with by debrouxl on 2015-09-08 at 17:57

2015-09-12, 08:01	#219
debrouxl Sep 2009 977 Posts	I tried to sum up the (potential) reservations for NFS@Home's 14e in the latest several dozens of posts: Will queue today: 12491_61_minus1 (ECM to 2/9 of SNFS difficulty by Ryan Propper) C175_142_39 (ECM to t50 should be enough, SNFS difficulty ~227) t55 in progress by Sean Wellman: C172_122_101 (ETA September 17th) C178_147_44 (ETA by September 24th) Need another t50 because of SNFS difficulty 232: (I'll contribute some curves) C196_129_62 C186_124_73 C189_139_46 Need t55, in slow progress mostly by Jon (jyb): 11^236+4^236 (100 curves at B1=26e7 by Lionel Debroux) 8^271-5^271 12^227-11^227 Additionally, there's a GNFS difficulty 184 from Home Primes sequence 4496, which has received 0.5t55. It's clearly food for 15e, but I don't manage that part of the queue. Tom, how much ECM work do you want on a GNFS 184 task ? Last fiddled with by debrouxl on 2015-09-12 at 08:04

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
System management notes	kriesel	kriesel	7	2020-10-21 18:52
Improving the queue management.	debrouxl	NFS@Home	10	2018-05-06 21:05
Script-based Primenet assignment management	ewmayer	Software	3	2017-05-25 04:02
Do normal adults give themselves an allowance? (...to fast or not to fast - there is no question!)	jasong	jasong	35	2016-12-11 00:57
Power Management settings	PrimeCroat	Hardware	3	2004-02-17 19:11

2015-09-08, 15:36	#210
swellman Jun 2012 11×281 Posts	There are three 30-bit LP candidates in the xyyx project that have been ECM'd to t50 by yoyo@home but test sieving shows all three to be much slower to sieve than their SNFS difficulty would suggest. C207_131_58 C196_129_62 C186_124_73

2015-09-08, 16:25	#212
swellman Jun 2012 11×281 Posts	I found two more slow-to-sieve xyyx numbers with (notionally) 30-bit LPs. C175_142_39 C189_139_46 Both have been ECM'd to t50 - the first by yoyo@Home, the second by me.

2015-09-08, 17:32	#213
wombatman I moo ablest echo power! May 2013 13·137 Posts	I'll have a C184 from HP2(4496) soon as well. Running ECM up to 0.5t55.

2015-09-08, 18:31	#215
wombatman I moo ablest echo power! May 2013 13·137 Posts	I based that 1/2t55 off of 2/7ths of 184 (52.5). I can do a full t55 if need be, but that'll take probably another week or so.

2015-09-10, 12:41	#216
VictordeHolland "Victor de Hollander" Aug 2011 the Netherlands 2³×3×7² Posts	We might need to think about adding some harder numbers (SNFS 240-250) to the 14e queue. The currently queued SNFS(220-230) don't keep the crunchers fed for long.

2015-09-10, 13:25	#217
swellman Jun 2012 3091₁₀ Posts	C172_122_101 will be ECM'd to t55 NLT Sep 17. At SNFS of 245, it will help feed the sievers. For a few days anyway ;-) C178_147_44 will be reach t55 within a week after that. I believe C186_124_73, C175_142_39 and C189_139_46 are on deck to be queued, but they are only SNFS 23x. Carbs, no meat. Then there are these.

2015-09-11, 22:23	#218
wblipp "William" May 2003 New Haven 2×7×13² Posts	(12491^61-1)/12490 is ready for SNFS thanks to ECM by Ryan Propper.

2015-09-13, 09:07	#220
debrouxl Sep 2009 977 Posts	Hmm. On my computer, trial sieving of the SNFS difficulty 226 C175_142_39 with 31-bit LPs shows the same sieving speed as SNFS difficulty 250 (!) 12491_61_minus1 with a sextic polynomial, whether a quintic or a sextic polynomial (they produce similar yields, unsurprisingly) is used for C175_142_39. Therefore, it needs quite a bit more than the t50 ECM work indicated by the rule of thumb, in order to reduce the likelihood of an ECM miss. So far, I've done another 600 curves at B1=43e6, but t50 + t45 is a far cry from 3t50 or 4t50...