Linear algebra reservations, progress and results - Page 162

RichD · 2017-01-25, 16:40

Got it to work - I think.

http://pastebin.com/mdS0wmbw

VBCurtis · 2017-01-27, 03:25

13*2^905-1 produced filtering message "matrix not dense enough, retrying" a few times, finally building a matrix of density 40 (target density was set to 140). After ~70 hrs of LA, only the trivial dependencies were recovered.

Restarting postprocessing with filter_maxrels set to 310M.

Next time, is "matrix not dense enough, retrying" the sort of nearly-fatal message that should incite me to restart filtering with fewer relations?

RichD · 2017-01-27, 03:43

Quote:

Originally Posted by VBCurtis

Next time, is "matrix not dense enough, retrying" the sort of nearly-fatal message that should incite me to restart filtering with fewer relations?

I saw that message once but didn't think much about it because it eventually completed. But it was a 10 day solve.

http://pastebin.com/SvXfm9Wh

jyb · 2017-01-27, 17:43

Quote:

Originally Posted by VBCurtis

13*2^905-1 produced filtering message "matrix not dense enough, retrying" a few times, finally building a matrix of density 40 (target density was set to 140). After ~70 hrs of LA, only the trivial dependencies were recovered.

Restarting postprocessing with filter_maxrels set to 310M.

Next time, is "matrix not dense enough, retrying" the sort of nearly-fatal message that should incite me to restart filtering with fewer relations?

Strangely enough, that message seems to appear in two different situations. One is that there aren't enough relations, and the other is that there are too many. In practice I've only seen it in the latter situation.

In this case, the number was way oversieved. For a 31-bit job, 250M should have been plenty. Some extra can produce a smaller matrix, but 350M is firmly in the territory where one sees this problem. I would probably have just started with maxrels set to 300M.

swellman · 2017-01-28, 13:58

Reserving 9+5_295 from the 15e queue.

RichD · 2017-01-28, 16:19

I'll take 32668561^31-1 next.

swellman · 2017-01-29, 15:23

Quote:

Originally Posted by swellman

Reserving 19218301^31 - 1 if it is available.

Code:

prp76 factor: 2758005348426399256471975117475109030251493697251061390611599493168874121569
prp103 factor: 7812280223399775417889125540463887990280512149934969036205872019610582450050573879946788206077444320093

swellman · 2017-01-29, 15:25

Reserving C222_125_92.

VBCurtis · 2017-01-29, 23:45

Code:

prp62 factor: 46205165266133662482948385141378840157279684813622824070309817
prp110 factor: 29742352095538657419873460130787243208882192833790233431382969850597187601711543963363200375101906633497733877

Log attached. http://pastebin.com/y7CjDbUa

fivemack · 2017-01-31, 15:31

Code:

p61 factor: 2828552482149628110952085410657566731619129765282892686071101
p126 factor: 129999454062486614642509451137992207528880979362609412660254837839413317446553082608089399055348960433331952479568500186490943

22.26M matrix at density 120 took around 255 hours on 7 threads E5-2650v2.

Log attached and at http://pastebin.com/mjjhwWvC

swellman · 2017-02-02, 10:28

Quote:

Originally Posted by swellman

Reserving 9468940004449^17 - 1 for post processing.

Code:

prp71 factor: 49319570324911942576065847490549577181702305054656307368513056057736237
prp117 factor: 236408236208426095318413372774730508091627951849214226624392009594648988074456868955277552955382718640478169517494601

2017-01-25, 16:40	#1772
RichD Sep 2008 Kansas 7·463 Posts	P55 (19269...)^5-1 Log Got it to work - I think. http://pastebin.com/mdS0wmbw

2017-01-28, 13:58	#1776
swellman Jun 2012 101101111111₂ Posts	9+5_295 (15e) Reserving 9+5_295 from the 15e queue.

2017-01-29, 15:25	#1779
swellman Jun 2012 101101111111₂ Posts	C222_125_92 Reserving C222_125_92.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
restarting nfs linear algebra	cubaq	YAFU	2	2017-04-02 11:35
Linear algebra at 600%	CRGreathouse	Msieve	8	2009-08-05 07:25
Linear algebra crashes	10metreh	Msieve	3	2009-02-02 08:34
Linear algebra proof	Damian	Math	8	2007-02-12 22:25
Linear algebra in MPQS	R1zZ1	Factoring	2	2007-02-02 06:45

2017-01-27, 03:25	#1773
VBCurtis "Curtis" Feb 2005 Riverside, CA 4618₁₀ Posts	13*2^905-1 produced filtering message "matrix not dense enough, retrying" a few times, finally building a matrix of density 40 (target density was set to 140). After ~70 hrs of LA, only the trivial dependencies were recovered. Restarting postprocessing with filter_maxrels set to 310M. Next time, is "matrix not dense enough, retrying" the sort of nearly-fatal message that should incite me to restart filtering with fewer relations?

2017-01-28, 16:19	#1777
RichD Sep 2008 Kansas 7·463 Posts	I'll take 32668561^31-1 next.