BOINC NFS sieving - NFS@Home - Page 172

Batalov · 2021-12-25, 15:21

S.S.W. suggested

Quote:

Maybe you could do 2,2622L c229 as well.
It is just beyond my computer capability.
A good snfs polynomial is
x^4 - 2x^3 + 2x^2 - 4x + 4
with m = 2^219,

I replied that quartics (this one is diff.265-ish) are quite hard but maybe NFS @ Home or forum would be interested.
I think that in his internal table this candidate shows up as the 'easiest' but it is a quartic.

swellman · 2021-12-25, 15:57

Cownoise scores this quartic at SNFS difficulty 263.1 but that’s quibbling. It will be a hard factorization to complete but 16e_small may be able to complete it as a 34/33 or 34/34 job.

Or as you suggest, run it as a team effort. Doesn’t seem to require Greg’s intervention.

For comparison, I just posted a SNFS 242 quartic from the HCN project. It’s a 33/32 job.

jyb · 2021-12-29, 21:47

Quote:

Originally Posted by Batalov

I replied that quartics (this one is diff.265-ish) are quite hard but maybe NFS @ Home or forum would be interested.
I think that in his internal table this candidate shows up as the 'easiest' but it is a quartic.

Quote:

Originally Posted by swellman

Cownoise scores this quartic at SNFS difficulty 263.1 but that’s quibbling. It will be a hard factorization to complete but 16e_small may be able to complete it as a 34/33 or 34/34 job.

Or as you suggest, run it as a team effort. Doesn’t seem to require Greg’s intervention.

For comparison, I just posted a SNFS 242 quartic from the HCN project. It’s a 33/32 job.

This might be interesting to look at as an octic. We're getting closer to the point where quartics and octics are equally painful, and it would be interesting to know how this one fares. BTW, note that the SNFS difficulty of 263.1 is for a polynomial slightly different from the one Sam suggested (which has a difficulty of 263.7). The octic would have a difficulty of 263.1 as well.

charybdis · 2021-12-29, 23:05

Quote:

Originally Posted by jyb

This might be interesting to look at as an octic. We're getting closer to the point where quartics and octics are equally painful, and it would be interesting to know how this one fares. BTW, note that the SNFS difficulty of 263.1 is for a polynomial slightly different from the one Sam suggested (which has a difficulty of 263.7). The octic would have a difficulty of 263.1 as well.

Haven't test-sieved this but I agree that the octic should be looked at too. From a small amount of test-sieving on 7,395+/- (diff. 268, the smallest remaining Cunningham quartics before the base-2 extension), the quartic appears to have a slight edge over the octic, but this is partly because the octic has larger coefficients, the largest being 7^4 = 2401. For 2,2622L, the quartic and octic have the same non-zero coefficients.

By the time you get to 2,1180+ (diff. 285), the octic is ~3 times faster than the quartic.

pinhodecarlos · 2021-12-30, 15:08

Note on the upcoming challenge, please do not download any post processing jobs for the first 24 hours, I get the feeling the server will struggle to process the bunkering wus. Keep an eye on the "in progress" wus for 16eV5.https://www.boincstats.com/stats/cha...team/chat/1098

jyb · 2021-12-30, 16:23

Quote:

Originally Posted by pinhodecarlos

Note on the upcoming challenge, please do not download any post processing jobs for the first 24 hours, I get the feeling the server will struggle to process the bunkering wus. Keep an eye on the "in progress" wus for 16eV5.https://www.boincstats.com/stats/cha...team/chat/1098

Interestingly it would appear that bunkering is only being done for the 16eV5 queue, at least so far. None of the other queues appears to be affected.

pinhodecarlos · 2021-12-30, 17:00

Quote:

Originally Posted by jyb

Interestingly it would appear that bunkering is only being done for the 16eV5 queue, at least so far. None of the other queues appears to be affected.

Highest paid ones.

jyb · 2021-12-30, 17:08

Quote:

Originally Posted by pinhodecarlos

Highest paid ones.

Yeah, I figured that was it, but it has interesting implications for the other queues, which we've been trying to fill up in anticipation. If the challenge passes them by, then there will be quite a backlog of work for a while.

swellman · 2021-12-30, 17:29

But don’t the challenge participants automatically shift to the other queues when the first runs dry? Last challenge I saw BOINC folks complain they couldn’t get anything from the 14d queue (it was always empty or the work was instantly grabbed whenever a new number appeared).

I think we’ve preloaded the various queues as best we can. Even Greg’s big queue has a lot of work in it. Just a matter of WU flow rate, which I believe Greg can open up if needed.

pinhodecarlos · 2021-12-30, 17:55

Only a few teams know how to bunker so I would expect other clients to run the other apps as soon as the challenge starts. Also don't forget that due to memory limitation you will also see more clients running them. In parallel the new Formula Boinc season starts. You have done enough...now enjoy the show.

pinhodecarlos · 2022-01-01, 01:13

Can't reach the server, Greg?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Boinc Statistics for NFS@Home borked ?	thomasn	NFS@Home	1	2013-10-02 15:31
BOINC NFS sieving - RSALS	debrouxl	NFS@Home	621	2012-12-14 23:44
BOINC?	masser	Sierpinski/Riesel Base 5	1	2009-02-09 01:10
BOINC?	KEP	Twin Prime Search	212	2007-04-25 10:29
BOINC	bebarce	Software	3	2005-12-15 18:35

2021-12-25, 15:57	#1883
swellman Jun 2012 7412₈ Posts	Cownoise scores this quartic at SNFS difficulty 263.1 but that’s quibbling. It will be a hard factorization to complete but 16e_small may be able to complete it as a 34/33 or 34/34 job. Or as you suggest, run it as a team effort. Doesn’t seem to require Greg’s intervention. For comparison, I just posted a SNFS 242 quartic from the HCN project. It’s a 33/32 job.

2021-12-30, 15:08	#1886
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 12013₈ Posts	Note on the upcoming challenge, please do not download any post processing jobs for the first 24 hours, I get the feeling the server will struggle to process the bunkering wus. Keep an eye on the "in progress" wus for 16eV5.https://www.boincstats.com/stats/cha...team/chat/1098

2021-12-30, 17:29	#1890
swellman Jun 2012 2·5²·7·11 Posts	But don’t the challenge participants automatically shift to the other queues when the first runs dry? Last challenge I saw BOINC folks complain they couldn’t get anything from the 14d queue (it was always empty or the work was instantly grabbed whenever a new number appeared). I think we’ve preloaded the various queues as best we can. Even Greg’s big queue has a lot of work in it. Just a matter of WU flow rate, which I believe Greg can open up if needed.

2021-12-30, 17:55	#1891
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 5131₁₀ Posts	Only a few teams know how to bunker so I would expect other clients to run the other apps as soon as the challenge starts. Also don't forget that due to memory limitation you will also see more clients running them. In parallel the new Formula Boinc season starts. You have done enough...now enjoy the show.

2022-01-01, 01:13	#1892
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 12013₈ Posts	Can't reach the server, Greg?