Fast Breeding (guru management) - Page 61

RichD · 2016-08-31, 23:07

Quote:

Originally Posted by pinhodecarlos

What's up with 70841^53-1 (15e)?

The poly file contains lss: 0.
Does that mean the GNFS side for this SNFS poly?

jyb · 2016-08-31, 23:54

Quote:

Originally Posted by RichD

The poly file contains lss: 0.
Does that mean the GNFS side for this SNFS poly?

It means that it should sieve special-q on the algebraic side, rather than the rational side. It's most common to use the algebraic side for GNFS, and somewhat more common to use the rational side for SNFS, but that doesn't mean one has to use the rational side for SNFS.

I.e. there's nothing inherently wrong with having "lss: 0" in the poly file for an SNFS job. In fact, there are some SNFS jobs in the 14e queue right now with lss: 0, per the recommendation of the person who submitted them.

pinhodecarlos · 2016-09-01, 00:42

For the current HCN in sieve progress can more 20M special Q be added please. Thank you.

jyb · 2016-09-01, 02:38

Quote:

Originally Posted by pinhodecarlos

For the current HCN in sieve progress can more 20M special Q be added please. Thank you.

Which numbers are you talking about, specifically?

pinhodecarlos · 2016-09-01, 04:57

Quote:

Originally Posted by jyb

Which numbers are you talking about, specifically?

The one reserved by myself (9+5_254) and the next three. Basically all should have at least 260M raw relations or more to have a lower LA time when matrix is built.

pinhodecarlos · 2016-09-01, 05:12

If the sieve CPU power is there available we should use it to over-sieve the numbers because the clients only care about the points, the rest is BS for them. My case the faster the LA is processed the better so I can go on to the next one. The one I'm running now had a ETA for LA of 80 and something hours when previous ones over-sieved were around 40-50 hours. This is when I only have an old laptop with 4 cores and not 6/8/12 as our co-workers.
Thank you Jon.

pinhodecarlos · 2016-09-02, 08:08

As Greg mention we are very close to reach 1000 factorizations on 14e side. I was wondering if we have more integers to queue so I can request some teams support to quickly achieve this milestone. This means we are 62 short to reach 1000. What are you all thoughts?

Carlos

swellman · 2016-09-02, 14:11

The xyyxf project has finally achieved t50 on all composites, with the bulk of the credit going to yoyo@Home. Now the project has started to enqueue numbers with yoyo to partial/full t55. Most of these numbers will be ready for SNFS once yoyo completes its work @B1=11e7. A few will likely complete ECM as soon as this weekend, with a steady stream of NFS ready composites expected thereafter.

I have polys for several hundred of the xyyxf composites. Willing to post in this thread preprocessed numbers/polys as yoyo does its work and as the queues/gatekeepers require them.

XYYXF · 2016-09-03, 00:51

C284_135_127 survived 20k+ curves at B1=300M. Maybe that's enough for 15e?

pinhodecarlos · 2016-09-04, 19:24

Jon,

By looking at the relations gathered C202_123_83, C182_133_62, C178_140_114, 5+4_347 and 3+2_508 need to be pushed further, do you agree (more special Q needed)?
Also please aware that we are back to daily 20k (14e) wus so maybe queue up more 14e integers.

Thank you.

Carlos

unconnected · 2016-09-05, 14:15

There are several near-repdigit composites reserved by Lionel a while ago. Two of them were cracked by ECM (18883_291 and 54441_248) and others received adequate ECM work and could be loaded into the 14e queue.

Also 90007_248 already done.

2016-09-01, 05:12	#666
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 4947₁₀ Posts	If the sieve CPU power is there available we should use it to over-sieve the numbers because the clients only care about the points, the rest is BS for them. My case the faster the LA is processed the better so I can go on to the next one. The one I'm running now had a ETA for LA of 80 and something hours when previous ones over-sieved were around 40-50 hours. This is when I only have an old laptop with 4 cores and not 6/8/12 as our co-workers. Thank you Jon. Last fiddled with by pinhodecarlos on 2016-09-01 at 05:15

2016-09-02, 08:08	#667
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 3×17×97 Posts	62 As Greg mention we are very close to reach 1000 factorizations on 14e side. I was wondering if we have more integers to queue so I can request some teams support to quickly achieve this milestone. This means we are 62 short to reach 1000. What are you all thoughts? Carlos

2016-09-04, 19:24	#670
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 3·17·97 Posts	Jon, By looking at the relations gathered C202_123_83, C182_133_62, C178_140_114, 5+4_347 and 3+2_508 need to be pushed further, do you agree (more special Q needed)? Also please aware that we are back to daily 20k (14e) wus so maybe queue up more 14e integers. Thank you. Carlos Last fiddled with by pinhodecarlos on 2016-09-04 at 19:28

2016-09-05, 14:15	#671
unconnected May 2009 Russia, Moscow 2,593 Posts	There are several near-repdigit composites reserved by Lionel a while ago. Two of them were cracked by ECM (18883_291 and 54441_248) and others received adequate ECM work and could be loaded into the 14e queue. Also 90007_248 already done. Last fiddled with by unconnected on 2016-09-05 at 14:16

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

2016-09-01, 00:42	#663
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 1353₁₆ Posts	For the current HCN in sieve progress can more 20M special Q be added please. Thank you.

2016-09-02, 14:11	#668
swellman Jun 2012 3,089 Posts	The xyyxf project has finally achieved t50 on all composites, with the bulk of the credit going to yoyo@Home. Now the project has started to enqueue numbers with yoyo to partial/full t55. Most of these numbers will be ready for SNFS once yoyo completes its work @B1=11e7. A few will likely complete ECM as soon as this weekend, with a steady stream of NFS ready composites expected thereafter. I have polys for several hundred of the xyyxf composites. Willing to post in this thread preprocessed numbers/polys as yoyo does its work and as the queues/gatekeepers require them.

2016-09-03, 00:51	#669
XYYXF Jan 2005 Minsk, Belarus 620₈ Posts	C284_135_127 survived 20k+ curves at B1=300M. Maybe that's enough for 15e?