Fast Breeding (guru management) - Page 66

RichD · 2016-09-29, 14:52

Another possible option is to convert it to a team sieve (which I am willing to setup) and finish it that way. But we still have the problem of finding a post-processor with sufficient resources,

fivemack · 2016-09-29, 16:53

I have sufficient resources (even with the two quite large post-processings I'm currently committed to) and am happy to post-process

VBCurtis · 2016-09-29, 17:40

I'll join a team sieve with a couple desktop cores.

Pascal Ochem · 2016-09-30, 08:50

FYI, 70841 comes from sigma(19^6) = 701 * 70841.
With no known factors for sigma(70841^52), we have to branch on 701, which creates a large subtree.

unconnected · 2016-09-30, 14:24

Several polys from near-repdigit project reserved by Lionel.

93×10^247-1

Code:

n: 929999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999 
m: 100000000000000000000000000000000000000000
deg: 6
c6: 930
c0: -1
skew: 0.32
# Murphy_E = 1.606e-13
type: snfs
lss: 1
rlim: 99000000
alim: 99000000
lpbr: 30
lpba: 30
mfbr: 62
mfba: 62
rlambda: 2.7
alambda: 2.7

(55×10^247+539)/9

Code:

n: 61111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111117 
m: 100000000000000000000000000000000000000000
deg: 6
c6: 550
c0: 53
skew: 0.68
# Murphy_E = 1.659e-13
type: snfs
lss: 1
rlim: 98000000
alim: 98000000
lpbr: 30
lpba: 30
mfbr: 62
mfba: 62
rlambda: 2.7
alambda: 2.7

(61×10^244-79)/9

Code:

n: 67777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777777 
m: 50000000000000000000000000000000000000000 
deg: 6 
c6: 976 
c0: -175 
skew: 0.75 
# Murphy_E = 2.513e-13 
type: snfs 
lss: 1 
rlim: 92000000 
alim: 92000000 
lpbr: 30 
lpba: 30 
mfbr: 61 
mfba: 61 
rlambda: 2.7 
alambda: 2.7

VBCurtis · 2016-09-30, 16:04

How is it useful to have mfbr just over twice lpbr? 30/62, from what I understand, seems a bad choice. I've seen 31/62 and 31/61 and 30/60, all with the idea that a number that breaks into two large primes has a decent chance of both large primes being less than the lpbr bound. Could someone explain?

jyb · 2016-10-01, 05:33

Quote:

Originally Posted by unconnected

Several polys from near-repdigit project reserved by Lionel.

93×10^247-1

(55×10^247+539)/9

(61×10^244-79)/9

Are these being proposed for the 14e queue? They're a little on the high side for difficulty, but the queue is running out, so that's fine. Some corrections, though:

- Looks like it should be (55x10^247+53)/9

- Looks like it should be (61*10^244-7)/9

- I too am curious about the 30-bit large primes for such a big job (as well as the 62-bit cracking limit). Is that right?

swellman · 2016-10-01, 15:22

Some new xyyx candidates for 14e, all sieved on the -r side:

C237_141_52
C234_127_96
C219_140_57

Code:

n: 257482409096276689356291439166666994823361454871347010452026020329330357973299282576629913210540032482684441884217004588410556444182735802213387904572856631057443909281177245363559811768245618086365377269085032563924873733356342129373609
# 141^52+52^141, difficulty: 243.67, anorm: 2.03e+033, rnorm: 6.16e+053
# scaled difficulty: 247.09, suggest sieving rational side
# size = 5.223e-017, alpha = 0.000, combined = 1.623e-013, rroots = 1
type: snfs
size: 243
skew: 3.2846
c5: 52
c0: 19881
Y1: -3105926159393528563401
Y0: 1117104038291523308698708754461764526705953734656
rlim: 134000000
alim: 134000000
lpbr: 31
lpba: 31
mfbr: 62
mfba: 62
rlambda: 2.7
alambda: 2.7

Code:

n: 179035128462164539855029131827055986941460988970878694760144277503349404481103511146504689819728028379456994567930244211342663605502442253724493225871467815271273056704531187114629065580967241689402470749914398666231704046718820032971
# 127^96+96^127, difficulty: 253.55, anorm: 4.90e+036, rnorm: -8.78e+047
# scaled difficulty: 255.43, suggest sieving rational side
# size = 2.064e-012, alpha = 0.000, combined = 1.993e-013, rroots = 0
type: snfs
size: 253
skew: 1.0699
c6: 2
c0: 3
Y1: -848644673048462844861747991570893338836992
Y0: 4579937329576774398276408998492161
rlim: 134000000
alim: 134000000
lpbr: 31
lpba: 31
mfbr: 62
mfba: 62
rlambda: 2.7
alambda: 2.7

Code:

n: 178958278876837895159288917533782786546212909068135589183617391347231631177646550764785930690392025492373124179356643619372915078488611203941893722516472931433820139850583803382720687857658916002876126901044687332853579
# 140^57+57^140, difficulty: 245.82, anorm: 2.80e+032, rnorm: 5.44e+054
# scaled difficulty: 249.54, suggest sieving rational side
# siever: 15
# size = 4.435e-017, alpha = 0.000, combined = 1.369e-013, rroots = 1
type: snfs
size: 245
skew: 1.537
c6: 3249
c0: 42875
Y1: 41322093568000000000
Y0: -24272900770553981941874687268486966725193
rlim: 134000000
alim: 134000000
lpbr: 31
lpba: 31
mfbr: 62
mfba: 62
rlambda: 2.7
alambda: 2.7

And one more, first posted here for C232_126_107:

Code:

n: 2887884444212159417484451416563223463940590134126168004586647514085637539224478341689850930050221157525576138337988534178438744496207163444059886895143961485721141270709867989065788570209629350576520930852028248609276141003325224407
# 126^107+107^126, difficulty: 256.85, anorm: 6.73e+037, rnorm: 3.58e+048
# scaled difficulty: 258.64, suggest sieving rational side
# size = 9.897e-013, alpha = 0.000, combined = 1.169e-013, rroots = 0
type: snfs
size: 256
skew: 2.239
c6: 1
c0: 126
Y1: 4140562374860211619063098135818149338786107
Y0: -64072225938746379480587511979135205376
rlim: 134000000
alim: 134000000
lpbr: 31
lpba: 31
mfbr: 62
mfba: 62
rlambda: 2.7
alambda: 2.7

pinhodecarlos · 2016-10-06, 19:17

Do we have more candidates for the 14e queue? Just wandering due to the challenge upcoming in less than 5 hours.

jyb · 2016-10-07, 18:49

Quote:

Originally Posted by pinhodecarlos

Do we have more candidates for the 14e queue? Just wandering due to the challenge upcoming in less than 5 hours.

I haven't seen any evidence so far of increased work. Still, it's a reasonable question.

- There's GC(4,411), a GNFS(166) referenced in post #452.
- There are some OddPerfect numbers referenced in posts #554 and #569, subsequently summarized in post #597. The upshot is that there are 6 composites left there that seem appropriate (the second through seventh in fivemack's list, all quartics).
- There's an OddPerfect number referenced in post #567.
- There are unconnected's near-repdigit numbers from post #720. I'd still like clarification on the large prime bound for those.
- There are Fibonacci and Lucas numbers, but AFAICT they all have difficulty a little too high for 14e. Please correct me if I'm wrong on that.

unconnected · 2016-10-07, 23:43

Quote:

Originally Posted by jyb

- There are unconnected's near-repdigit numbers from post #720. I'd still like clarification on the large prime bound for those.

Sorry, I've missed your post here.
These params are auto-adjusted by project, for large prime bound suggested formulae is:

Quote:

lpbr: floor(d/25+21) where d=[digits]

Of course we can choose another params or even another poly.

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

2016-09-29, 14:52	#716
RichD Sep 2008 Kansas 3×1,129 Posts	Another possible option is to convert it to a team sieve (which I am willing to setup) and finish it that way. But we still have the problem of finding a post-processor with sufficient resources,

2016-09-29, 16:53	#717
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 6423₁₀ Posts	I have sufficient resources (even with the two quite large post-processings I'm currently committed to) and am happy to post-process

2016-09-29, 17:40	#718
VBCurtis "Curtis" Feb 2005 Riverside, CA 2²×1,217 Posts	I'll join a team sieve with a couple desktop cores.

2016-09-30, 08:50	#719
Pascal Ochem Apr 2006 103 Posts	FYI, 70841 comes from sigma(19^6) = 701 * 70841. With no known factors for sigma(70841^52), we have to branch on 701, which creates a large subtree.

2016-09-30, 16:04	#721
VBCurtis "Curtis" Feb 2005 Riverside, CA 2²·1,217 Posts	How is it useful to have mfbr just over twice lpbr? 30/62, from what I understand, seems a bad choice. I've seen 31/62 and 31/61 and 30/60, all with the idea that a number that breaks into two large primes has a decent chance of both large primes being less than the lpbr bound. Could someone explain?

2016-10-06, 19:17	#724
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 4947₁₀ Posts	Do we have more candidates for the 14e queue? Just wandering due to the challenge upcoming in less than 5 hours.