Fast Breeding (guru management) - Page 68

fivemack · 2016-10-19, 13:05

Queued those five - they're getting quite big for 14e but not completely intractable

swellman · 2016-10-19, 14:38

Quote:

Originally Posted by fivemack

Queued those five - they're getting quite big for 14e but not completely intractable

Thank you! Yes they were on the dotted line between 14/32 and 15/31. I suggested them for 14e to feed the grid.

Currently prepping a few more composites for 15e/32. Didn't figure any more candidates would be needed in that queue for weeks/months and I would just keep them in cold storage, but I did not account for the burst of activity in 15e during the recent challenge.

swellman · 2016-10-20, 19:10

Five more xyyx candidates for 15e. All have completed t55 courtesy of yoyo. All are best sieved on the -r side.

C249_128_105
C257_128_111
C262_129_110
C259_130_107
C260_126_125

Code:

n: 408142115848354487607374219172702590227716961541428941257817037760650713306584493536382972480050520569439818308139440871962367453098269986020320001294834662344715046476211444307521805621247072488839716576593016131849171553785626002993893150543743433
# 128^105+105^128, difficulty: 262.75, anorm: 3.80e+040, rnorm: 2.54e+048
# scaled difficulty: 264.06, suggest sieving rational side
# size = 2.756e-013, alpha = 0.000, combined = 4.771e-014, rroots = 0
type: snfs
size: 262
skew: 1.1991
c6: 11025
c0: 32768
Y1: -1329227995784915872903807060280344576
Y0: 2785962590401641140642702303409576416015625
rlim: 268000000
alim: 268000000
lpbr: 32
lpba: 32
mfbr: 64
mfba: 64
rlambda: 2.8
alambda: 2.8

Code:

n: 11352974339260713287419188709457754799656497083820769873019165370910574571671923124379100246282444027784935704982063498051829712028360238276792141076136022545176498050637445882144018930867458127822409717994900872184380943930334009622849224952428601394399491
# 128^111+111^128, difficulty: 265.89, anorm: 4.02e+040, rnorm: 8.25e+048
# scaled difficulty: 267.28, suggest sieving rational side
# size = 1.184e-013, alpha = 0.000, combined = 2.580e-014, rroots = 0
type: snfs
size: 265
skew: 1.1771
c6: 12321
c0: 32768
Y1: -170141183460469231731687303715884105728
Y0: 8949165805103346732088206385868884620523311
rlim: 268000000
alim: 268000000
lpbr: 32
lpba: 32
mfbr: 64
mfba: 64
rlambda: 2.8
alambda: 2.8

Code:

n: 3078917390898063299614641629280543677240017212615362692746867998413177465949601772851671747249762059104047760239686494407215870060815839794995325288362325681777034934142037072216502877577534036201683628130594824280195810533290320246828737451000297699054469479031
# 129^110+110^129, difficulty: 266.28, anorm: 5.93e+031, rnorm: 5.27e+058
# scaled difficulty: 270.78, suggest sieving rational side
# size = 3.292e-018, alpha = 0.000, combined = 2.208e-014, rroots = 1
type: snfs
size: 266
skew: 1.2801
c5: 16
c0: 55
Y1: -27100270549839440013580113370110596087308471041
Y0: 59590882688636047123006638050000000000000000000000000
rlim: 268000000
alim: 268000000
lpbr: 32
lpba: 32
mfbr: 64
mfba: 64
rlambda: 2.8
alambda: 2.8

Code:

n: 8738642855111442965233670480187098535413358350197329297402908429919633760898314244080407587740593977450148868248218595799425918955322438063879686614389706215923363057662004375507202190326186061970565820516555013553953850836122129079136859466750642802895878227
# 130^107+107^130, difficulty: 263.82, anorm: 2.60e+032, rnorm: 2.19e+058
# scaled difficulty: 268.14, suggest sieving rational side
# size = 2.857e-018, alpha = 0.000, combined = 2.119e-014, rroots = 1
type: snfs
size: 263
skew: 7.0077
c5: 1
c0: 16900
Y1: -247064529073450392704413000000000000000000000
Y0: 58073529249314919768851974028956890155855150099653849
rlim: 268000000
alim: 268000000
lpbr: 32
lpba: 32
mfbr: 64
mfba: 64
rlambda: 2.8
alambda: 2.8

Code:

n: 32913476719275760478620052474445896037571073259082234164549670589435206687781704152547706415948673433646638256771784239841524020079688752126713049342153591001183874020272771880143518130094539761529679374401573509105293719442246670446666803340579189948686150179
# 126^125+125^126, difficulty: 266.03, anorm: 6.73e+037, rnorm: 4.42e+049
# scaled difficulty: 268.00, suggest sieving rational side
# size = 3.845e-013, alpha = 0.000, combined = 5.700e-014, rroots = 0
type: snfs
size: 266
skew: 1.3399
c6: 14
c0: 81
Y1: -42722847678148575866619908287591521863073792
Y0: 108420217248550443400745280086994171142578125
rlim: 268000000
alim: 268000000
lpbr: 32
lpba: 32
mfbr: 64
mfba: 64
rlambda: 2.8
alambda: 2.8

RichD · 2016-10-22, 20:39

William hasn’t been seen around these parts in a while. I hope I’m not stepping out of bounds.

Going through some yoyo logs I found the following which appear ready for SNFS. Obviously these are quartics. Only the last run is listed.

Code:

 7500 @ 11e7 P55.15147_5M.C187
1514760402726090616648095033964880550337576601674963123^5-1
 7400 @ 11e7 P55.16635_5M.C162
1663521218376980133890596434622640837401485234568910473^5-1
 7400 @ 11e7 P55.16522_5M.C165
1652248266838673398348076441450265064693280306501309263^5-1
 7400 @ 11e7 P55.16058_5M.C170
1605871655610123748163599722104985996429838132153382107^5-1
 6200 @ 11e7 P54.29826_5M.C178
298260921928932067643011924483514064806042134514253973^5-1
 6100 @ 11e7 P54.24466_5M.C183
244664873957085470551973266612254452474958115042072789^5-1
 6100 @ 11e7 P54.29985_5M.C155
299851446221895648182226058564272663414018071384153431^5-1
 6100 @ 11e7 P54.26386_5M.C168
263864774073502583589315918259949271735733361405058571^5-1
 6000 @ 11e7 P54.23709_5M.C177
237098814036965303695624894383088172934311371478170807^5-1
 4900 @ 11e7 P53.56074_5M.C167
56074737781646790508395132462325980692442366265534933^5-1

jyb · 2016-10-22, 21:57

Quote:

Originally Posted by RichD

William hasn’t been seen around these parts in a while. I hope I’m not stepping out of bounds.

Going through some yoyo logs I found the following which appear ready for SNFS. Obviously these are quartics. Only the last run is listed.

Code:

 7500 @ 11e7 P55.15147_5M.C187
1514760402726090616648095033964880550337576601674963123^5-1
 7400 @ 11e7 P55.16635_5M.C162
1663521218376980133890596434622640837401485234568910473^5-1
 7400 @ 11e7 P55.16522_5M.C165
1652248266838673398348076441450265064693280306501309263^5-1
 7400 @ 11e7 P55.16058_5M.C170
1605871655610123748163599722104985996429838132153382107^5-1
 6200 @ 11e7 P54.29826_5M.C178
298260921928932067643011924483514064806042134514253973^5-1
 6100 @ 11e7 P54.24466_5M.C183
244664873957085470551973266612254452474958115042072789^5-1
 6100 @ 11e7 P54.29985_5M.C155
299851446221895648182226058564272663414018071384153431^5-1
 6100 @ 11e7 P54.26386_5M.C168
263864774073502583589315918259949271735733361405058571^5-1
 6000 @ 11e7 P54.23709_5M.C177
237098814036965303695624894383088172934311371478170807^5-1
 4900 @ 11e7 P53.56074_5M.C167
56074737781646790508395132462325980692442366265534933^5-1

I think all of these could make reasonably good 14e candidates except for P54.29985_5M.C155. That one could be more easily done by GNFS, and IMO at 155 digits it's probably too small for NFS@Home. Contrary opinions welcome.

RichD · 2016-10-22, 23:06

Quote:

Originally Posted by jyb

I think all of these could make reasonably good 14e candidates except for P54.29985_5M.C155. That one could be more easily done by GNFS, and IMO at 155 digits it's probably too small for NFS@Home. Contrary opinions welcome.

Let’s see if my thoughts are correct. The above has a polynomial difficulty of SNFS-214. Because it is one degree from the optimal (quintic vs. quartic) there is a 15-digit penalty. Therefore, the number has a realistic SNFS difficulty of 229. The GNFS crossover is around 0.7 so this equates to GNFS-160. Since the remaining composite is only 155, you are correct, GNFS is the preferable approach.

jyb · 2016-10-22, 23:38

Quote:

Originally Posted by RichD

Let’s see if my thoughts are correct. The above has a polynomial difficulty of SNFS-214. Because it is one degree from the optimal (quintic vs. quartic) there is a 15-digit penalty. Therefore, the number has a realistic SNFS difficulty of 229. The GNFS crossover is around 0.7 so this equates to GNFS-160. Since the remaining composite is only 155, you are correct, GNFS is the preferable approach.

I'm not sure that it's as rigidly defined as an exact 15-digit penalty, but the basic reasoning seems about right. I was actually comparing to the recent P53 (374249...)^5-1 from the 14e queue, which was a quartic with difficulty 210. It was relatively easy, but still plausible for 14e, and this one will be a bit harder. Whereas I know that a GNFS-155 is easy enough to do on personal hardware.

VBCurtis · 2016-10-23, 01:23

A more accurate GNFS crossover is 0.55 * snfs + 30 (per Batalov). GNFS still quicker, I think.

RichD · 2016-10-23, 02:55

As the composite size increases I believe that is a more appropriate calculation.

jyb · 2016-10-23, 04:22

Quote:

Originally Posted by VBCurtis

A more accurate GNFS crossover is 0.55 * snfs + 30 (per Batalov). GNFS still quicker, I think.

That formula would suggest that SNFS is better for this number. But it doesn't account for it being a quartic.

Or is that what you were saying?

VBCurtis · 2016-10-23, 04:54

I have no experience with quartics, so I gave no opinion. I've no reason to doubt the "add 15 digits as an estimate" (for what size is this correct? It can't be a fixed penalty for any size), though doing so makes GNFS appear only a little faster.
Perhaps a bit of test-sieving is in order?

2016-10-19, 13:05	#738
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 3×2,141 Posts	Queued those five - they're getting quite big for 14e but not completely intractable Last fiddled with by fivemack on 2016-10-19 at 14:09

2016-10-23, 04:54	#748
VBCurtis "Curtis" Feb 2005 Riverside, CA 2²×1,217 Posts	I have no experience with quartics, so I gave no opinion. I've no reason to doubt the "add 15 digits as an estimate" (for what size is this correct? It can't be a fixed penalty for any size), though doing so makes GNFS appear only a little faster. Perhaps a bit of test-sieving is in order? Last fiddled with by VBCurtis on 2016-10-23 at 04:55

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

2016-10-23, 01:23	#745
VBCurtis "Curtis" Feb 2005 Riverside, CA 2²·1,217 Posts	A more accurate GNFS crossover is 0.55 * snfs + 30 (per Batalov). GNFS still quicker, I think.

2016-10-23, 02:55	#746
RichD Sep 2008 Kansas 3387₁₀ Posts	As the composite size increases I believe that is a more appropriate calculation.