7- table
 

[code]Size Base Index Mod Diff Ratio
320 7 379 - 320.2 1
274 7 383 - 323.6 0.846
299 7 389 - 328.7 0.909
220 7 395 - 267.0 0.823 /5q
281 7 419 - 354 0.792
305 7 421 - 355.7 0.855
221 7 425 - 287.3 0.767 /5q
315 7 431 - 364.2 0.863
342 7 433 - 365.9 0.933
341 7 439 - 370.9 0.917 [/code]

[code]
7,259- C175 = P69 * P106
P69 = 357708888729481429441295130414301389734167320382217282028474835603757
P106 = 4105565335483009892405469384537186162336640263688626513538197586407434630567675699756450562264314334763133
[/code]
This was done using ggnfs.

2300 curves at B1=11M on 7,263- 7,281- 7,387- 7,361- 7,373- 7,389- 7,391-. Adds 0.50108 to p45 and 0.06662 to p50 for each.

Alex

8394 curves at B1=44M on 7,379- c320. Adds 6.47185 to p45, 1.09425 to p50 and 0.16726 to p55.

Alex

[QUOTE=akruppa]8394 curves at B1=44M on 7,379- c320. Adds 6.47185 to p45, 1.09425 to p50 and 0.16726 to p55.

Alex[/QUOTE]


Hi,

What criteria do you use to select your ECM candidates?

Mostly base 3, where 3,487- just started the matrix job. I'll do others soon. Besides base 3, p,q- with p,q prime and no/few known primitive factors are high on my list, too. Currently ECM is running on 11,239- c246.

Alex

263 = 50 + 213, completes 7, 347-
 

[QUOTE=garo;48284][CODE]Base Index Size 11M(45digits) 43M(50digits) 110M(55digits) 260M(60digits) Decimal
...
7 347- C263 0(0.267423) 0(0.0522979) 165(0.00921839) 0(0.00148122)
33026113620772114208766604063467661003723471054258542654638480163235395066667816968649454626706486286757449262870499223580340566358988528768748899525538521074418576135045990723823283809539231151245039475832681824086824878451971862643338135772425108809320687434713
... [/CODE][/QUOTE]

p50 = 89823582672830825295829211905814036272016809423171

nearing half-way through the last run on the last large incomplete range for
testing to p50, 2100 curves/7830 left to go on the generic c251-c299 list
(not 2- or 2+ with n < 1200). The only other, smaller range left having
been updated on the 2- discussion thread.

I'm assuming this one's complete, but Paul notes that he has
the 213-digit cofactor listed as "prp213", with the comment

[QUOTE]PS: note that I write 'prp213' and not 'p213', since one of the major computer
algebra systems (which I am using) says (courtesy Richard Pinch):

> ifactor(381707*293621*29363);

(3290922560713061)

> isprime(2432870015941);

true

> ifactor(2432870015941);

(1213) (551461) (3637)
[/QUOTE]

-bd (factor reported May 14th)

New 7-repunit pseudoprime
 

(7^35201-1)/6 is a base-3 pseudoprime.

At 29749 digits it's much too big for ECPP (20562 digits took six K8/2400-years, 15071 digits took two Xeon/2600-years); 35200 looks splendidly composite, so I suppose I should start hunting cyclotomic factors.

[QUOTE=fivemack;110607](7^35201-1)/6 is a base-3 pseudoprime.

At 29749 digits it's much too big for ...[/QUOTE]

Not to mention being a bit beyond the limit for the Cunningham
Tables. -bd

[QUOTE=fivemack;110607](7^35201-1)/6 is a base-3 pseudoprime.

At 29749 digits it's much too big for ECPP (20562 digits took six K8/2400-years, 15071 digits took two Xeon/2600-years); 35200 looks splendidly composite, so I suppose I should start hunting cyclotomic factors.[/QUOTE]

Indeed. The cyclotomic factors will not help ECPP, but they can greatly help
APR-CL.

[QUOTE=garo;48284][CODE]Base Index Size 11M(45digits) 43M(50digits) 110M(55digits) 260M(60digits) Decimal
...
7 317- C202 5215(1.40359) 0(0.203361) 165(0.00921839) 0(0.00148122)
4817387775478910748337470675295461658412305210077087712511098562369825581753381999542716517188369317201504120140218066670989302797199195269475798377529788582602272660294198048152636381680533617071470639
... [/CODE][/QUOTE]

This short table's now one shorter, after 7,317- c202 = p51*p151, with

p51 = 399676211948778347375576150888294069325575626336229

The 3rd cure2 duo factor (from the B1 = 260M run; on the part of
c190-c209 above difficulty 250). PaulZ's c120-355 file lists this as
entering ecmnet files as a c249, with what looks like a p47 found by
Philip McLaughlin 20.02.05. -Bruce