Queue management for 14e queue - Page 5

unconnected · 2021-09-03, 01:29

QUEUED AS C157_79_87

C157 from aliquot sequence 79^87 for 14e queue:

Code:

n: 4951817282321300125121563283973266890130432071999797605619959055323522627836598971946187541604723115152173600067798128300469087299960489125577620433072170779
# norm 3.275088e-15 alpha -8.292521 e 2.016e-12 rroots 5
skew: 49957707.18
c0: -208235189039389218619688718428466586619079
c1: 37781536379467125545223981130069230
c2: 489180382520633806645059772
c3: -38418454921101676246
c4: -239285831493
c5: 5016
Y0: -3970832215081885247847712953140
Y1: 114941225847724939
rlim: 33500000
alim: 33500000
lpbr: 29
lpba: 29
mfbr: 58
mfba: 58
rlambda: 2.6
alambda: 2.6
type: gnfs
lss: 0

Suggesting sieve range is 10M-50M, I'll take the LA.

RichD · 2021-09-04, 12:24

QUEUED AS 417__011_3m1

C153 from the OPN file t800.
417170460427416024594567400793767674921146783636256059892715875174336683894076800036991320011^3-1

Code:

n: 931716212649329827085826874632296805522376190752443111047033777427443503981234835282738781393247938253896381553901865467679801153870597073922973731405293
lss: 0
Y0: -607288295848199450676009139407
Y1: 204889130532497
c0: 17500403064728538394138027311449717866616
c1: 5584264633247370913975802344435818
c2: 292017629100948579858061139
c3: -25997407300505227152
c4: -203479838212
c5: 11280
skew: 23852827.67
# size 6.655e-15, alpha -8.500, combined = 3.359e-12 rroots = 5
# expecting poly E from 2.92e-12 to > 3.35e-12
type: gnfs
rlim: 33500000
alim: 33500000
lpbr: 29
lpba: 29
mfbr: 58
mfba: 84
rlambda: 2.5
alambda: 3.4

Trial sieving 5K blocks.

Code:

RichD · 2021-09-05, 01:18

This number is yielding below expectations. I've double checked the poly and my sieve results. Everything looks fine.

The one anomaly is the yield at 20M. Usually the curve is fairly flat for GNFS numbers in this range. Based on the surrounding yield numbers it should be about 13,4xx. This could explain the lower than expected.

Maybe keep an eye on the ongoing results because the Q range may need to be increased by 5M or 6M.

charybdis · 2021-09-05, 02:30

Quote:

Originally Posted by RichD

The one anomaly is the yield at 20M. Usually the curve is fairly flat for GNFS numbers in this range. Based on the surrounding yield numbers it should be about 13,4xx. This could explain the lower than expected.

I'm not sure it's just 20M that's an anomaly here. By complete coincidence, all four of the ranges that you picked contain more special-q than we would expect for a 5k range at those sizes; in total there are around 7% more than expected. I've explained before how to correct for this (the log is natural log):

Quote:

Originally Posted by charybdis

Calculate the yield per special-q using the number of special-q that the siever gives on the line

Code:

xxx Special q, yyy reduction iterations

Multiply this by C/log(Q) to get an estimate of yield in a range of length C around value Q.

RichD · 2021-09-05, 09:44

Could a couple percent be attributed to the poly being at the high end of the E-score range?

RichD · 2021-09-05, 10:00

Quote:

Originally Posted by charybdis

Calculate the yield per special-q using the number of special-q that the siever gives on the line

Code:

xxx Special q, yyy reduction iterations

Multiply this by C/log(Q) to get an estimate of yield in a range of length C around value Q.

I don't quite follow this. Here is the output from the siever for the 20M range of 5K block.

Code:

total yield: 15438, q=20005003 (0.24873 sec/rel) 
323 Special q, 477 reduction iterations

charybdis · 2021-09-05, 11:21

Yield per special-q = 15438/323 = 47.796
Expected number of special-q in 5k range at 20M = 5000/ln(20000000) = 297.42
Corrected yield estimate = 47.796*297.42 = 14215

RichD · 2021-09-05, 12:41

I went through the exercise to compute estimated from actual.

Code:

 Q   actual  estimated
===  ======  =========
 8M   13743    12423
10M   13780    13072
20M   15438    14215
30M   13296    13001
total  63M      59M

Even if I threw out the number for 20M it still doesn't explain the large difference from 46M which seems to be occurring.

Edit: I guess I must have randomly hit a few hot spots to skew the numbers.

charybdis · 2021-09-05, 13:15

Normally a hot spot is just a range with more special-q than expected. The variation in rels/q smooths out pretty well over a 5k range. I tried some more ranges and got the following:

Code:

Q    Actual Estimated
13M  13427  13800
16M  12182  13856
23M  14168  13977
26M  13255  13525

These are very much in agreement with your figures (and note how much smoother the corrected values are). I think something must have gone wrong at the NFS@home end.

unconnected · 2021-09-09, 22:30

QUEUED AS C162_159978_10752

C162 from 159978:i10752 for 14e queue:

Code:

n: 307971724803606446462872819609832224962609961269161295963764484634572235683426075635658509577753868636513892744089422683812968625603917337113577082298516439966433
# norm 1.157002e-15 alpha -7.090889 e 1.110e-12 rroots 3
skew: 11451205.90
c0: -3944555032855439079562882441928924509762
c1: -202025289949929521507548847820369
c2: 191219311077986300620049528
c3: -14361015183390153921
c4: 951667925579
c5: 68820
Y0: -21387460552177921418441173809027
Y1: 247235112212624731
rlim: 67000000
alim: 67000000
lpbr: 30
lpba: 30
mfbr: 60
mfba: 60
rlambda: 2.6
alambda: 2.6
type: gnfs
lss: 0

Suggesting sieve range is 10M-90M, I'll take the LA.

swellman · 2021-09-11, 14:25

Quote:

Originally Posted by RichD

QUEUED AS 417__011_3m1

C153 from the OPN file t800.
417170460427416024594567400793767674921146783636256059892715875174336683894076800036991320011^3-1

Code:

n: 931716212649329827085826874632296805522376190752443111047033777427443503981234835282738781393247938253896381553901865467679801153870597073922973731405293
lss: 0
Y0: -607288295848199450676009139407
Y1: 204889130532497
c0: 17500403064728538394138027311449717866616
c1: 5584264633247370913975802344435818
c2: 292017629100948579858061139
c3: -25997407300505227152
c4: -203479838212
c5: 11280
skew: 23852827.67
# size 6.655e-15, alpha -8.500, combined = 3.359e-12 rroots = 5
# expecting poly E from 2.92e-12 to > 3.35e-12
type: gnfs
rlim: 33500000
alim: 33500000
lpbr: 29
lpba: 29
mfbr: 58
mfba: 84
rlambda: 2.5
alambda: 3.4

Trial sieving 5K blocks.

Code:

Resubmitting job as 417__011_3m1a.

2021-09-05, 01:18	#47
RichD Sep 2008 Kansas 3³×149 Posts	417__011_3m1 This number is yielding below expectations. I've double checked the poly and my sieve results. Everything looks fine. The one anomaly is the yield at 20M. Usually the curve is fairly flat for GNFS numbers in this range. Based on the surrounding yield numbers it should be about 13,4xx. This could explain the lower than expected. Maybe keep an eye on the ongoing results because the Q range may need to be increased by 5M or 6M.

2021-09-05, 12:41	#52
RichD Sep 2008 Kansas 3³×149 Posts	I went through the exercise to compute estimated from actual. Code: Q actual estimated === ====== ========= 8M 13743 12423 10M 13780 13072 20M 15438 14215 30M 13296 13001 total 63M 59M Even if I threw out the number for 20M it still doesn't explain the large difference from 46M which seems to be occurring. Edit: I guess I must have randomly hit a few hot spots to skew the numbers. Last fiddled with by RichD on 2021-09-05 at 12:44

2021-09-05, 13:15	#53
charybdis Apr 2020 7·163 Posts	Normally a hot spot is just a range with more special-q than expected. The variation in rels/q smooths out pretty well over a 5k range. I tried some more ranges and got the following: Code: Q Actual Estimated 13M 13427 13800 16M 12182 13856 23M 14168 13977 26M 13255 13525 These are very much in agreement with your figures (and note how much smoother the corrected values are). I think something must have gone wrong at the NFS@home end.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Queue management for e_small and 15e queues	VBCurtis	NFS@Home	254	2022-01-02 01:59
System management notes	kriesel	kriesel	10	2021-09-01 15:56
Run down the queue on MPRIME without quitting GIMPS	Rodrigo	Software	7	2018-05-25 13:26
Improving the queue management.	debrouxl	NFS@Home	10	2018-05-06 21:05
split a prime95 queue & client installation	joblack	Information & Answers	1	2009-01-06 08:45

2021-09-05, 09:44	#49
RichD Sep 2008 Kansas 3³·149 Posts	Could a couple percent be attributed to the poly being at the high end of the E-score range?

2021-09-05, 11:21	#51
charybdis Apr 2020 7·163 Posts	Yield per special-q = 15438/323 = 47.796 Expected number of special-q in 5k range at 20M = 5000/ln(20000000) = 297.42 Corrected yield estimate = 47.796*297.42 = 14215