mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Factoring (https://www.mersenneforum.org/forumdisplay.php?f=19)
-   -   Gratuitous factors thread (https://www.mersenneforum.org/showthread.php?t=10029)

jcrombie 2012-03-17 23:19

In the curiosity department,

48[SUP]131[/SUP]-1 c188 = p84.p105

[CODE]
p84= 666597940127567478926314250053215593484751546665734633414090507253770559719213333999
p105= 102043430077755943507387622730021855516367729839780613874018363499064959278025856799035207732985020302479
[/CODE]

The largest triple quote/unquote?

unconnected 2012-03-19 09:18

Lucky hit on seq 11040:i9388 c149:

[CODE]Using B1=43000000, B2=240490660426, polynomial Dickson(12), sigma=360494343
Step 1 took 237569ms
Step 2 took 72704ms
********** Factor found in step 2: 58188790498537295496280120496146152069818685197607499
Found probable prime factor of 53 digits: 58188790498537295496280120496146152069818685197607499
Composite cofactor 581029012516063493844714271258439220958080054298271434349094661167856372895231814321798430235827 has 96 digits[/CODE]

Remaining c96 splits as p46*p51 by gnfs. So, the largest factor was found by ECM :smile:

chris2be8 2012-03-19 16:57

[QUOTE=jcrombie;293328]In the curiosity department,

48[SUP]131[/SUP]-1 c188 = p84.p105

[CODE]
p84= 666597940127567478926314250053215593484751546665734633414090507253770559719213333999
p105= 102043430077755943507387622730021855516367729839780613874018363499064959278025856799035207732985020302479
[/CODE]The largest triple quote/unquote?[/QUOTE]

Are you working on the Brent tables? I am and would like to avoid duplicated effort.

Chris K

jcrombie 2012-03-19 19:41

[QUOTE=chris2be8;293508]Are you working on the Brent tables? I am and would like to avoid duplicated effort.

Chris K[/QUOTE]

Hi Chris,

I'm currently working on first holes from Brent's holes.txt file from about 3 years ago. I've made it up to 48 131- as you can see. Those are the only concentrated numbers
that I work on (and an occasional cunningham). (I also do ecm sweeps such as t20
for base 1001 - 9999 exponents up to 150). I send my updates to Prof Brent, but since
he has yet to absorb my last giant dump from Dec 5, I've been holding off on new updates to him). If you are working on first holes too, please update the factordb.com
as I always check there first before starting a new number and I send my factors there
as well. (My ecm sweep factors should be also available on my home web server
at [URL]http://myfactors.mooo.com[/URL] which is semi-regularly updated). You can PM me any
numbers that you are working on so that I can avoid them.

Cheers,

Jonathan

unconnected 2012-03-26 11:34

ECM strikes on sequence 842592:i8008 c161:
[CODE]Using B1=110000000, B2=776278396540, polynomial Dickson(30), sigma=625751920
Step 1 took 746426ms
Step 2 took 197356ms
********** Factor found in step 2: 8364584338015904739838273511475443354525819955383893373
Found probable prime factor of 55 digits: 8364584338015904739838273511475443354525819955383893373
Probable prime cofactor has 106 digits[/CODE]

fivemack 2012-03-30 14:29

Unusually good luck
 
In precisely one out of sixteen parallel ECM jobs, I got a hit. Note the run number :smile:

[code]
Run 224 out of 225:
Using B1=100000000, B2=776268975310, polynomial Dickson(30), sigma=2132373404
Step 1 took 964564ms
Step 2 took 273725ms
********** Factor found in step 2: 72884415673453771044511224741154120152573825983051
Found probable prime factor of 50 digits: 72884415673453771044511224741154120152573825983051
Probable prime cofactor 266087932759807516412782791665301365513151543357271524278499799994851545256483405953991561813263218272738563059 has 111 digits
[/code]

Dubslow 2012-04-12 21:55

C125 splits as P63*P63.
[url]http://factordb.com/index.php?id=1100000000218611640[/url]
The larger divided by the smaller is approximately 1.36.

10metreh 2012-04-13 08:57

[QUOTE=Dubslow;296278]C125 splits as P63*P63.
[url]http://factordb.com/index.php?id=1100000000218611640[/url]
The larger divided by the smaller is approximately 1.36.[/QUOTE]

Here's a properly nice split:
[code]04/12/12 16:30:02 v1.31.1 @ ________, prp47 = [color=red]42[/color]396433854107117200014151850098395717424864963
04/12/12 16:30:02 v1.31.1 @ ________, prp47 = [color=red]42[/color]636915368713553443901733545681353700650075997[/code]

unconnected 2012-04-21 08:36

From sequence 842592:i8010
[CODE]Using B1=110000000, B2=776278396540, polynomial Dickson(30), sigma=3697488491
Step 1 took 974123ms
Step 2 took 256964ms
********** Factor found in step 2: 1815624236101952532630664834371236936165244754014489599
Found probable prime factor of 55 digits: 1815624236101952532630664834371236936165244754014489599[/CODE]

fivemack 2012-04-30 09:03

slightly surprising result
 
I just ran a C162 using 14e and 31-bit rather than 30-bit large primes, and am quite surprised how quickly it finished; more than 20% faster than an adjacent number done with 30-bit LP, and with only 162M relations of which 140M unique. The matrix is a little large (9.3M) but I can cope with ugly matrices; maybe I'm unusual in having similar resources for sieving and matrix, but the extra time running the matrix is less than half what's been saved in the sieving.

Will try a C160 with the larger large primes and see if I can find the crossover point.

lorgix 2012-05-13 20:24

ECM stage1 hit: 229487392729284619870165192547258703516391

p42 of (7549^43-1)/((7549-1)*1979*160649*564586759413619*38929084316387597)

[CODE]Group order:
[ <2, 4>, <3, 3>, <7, 1>, <283, 1>, <2063, 1>, <2213, 1>, <4211, 1>, <5531, 1>,
<8461, 1>, <15269, 1>, <18077, 1>, <18859, 1>, <[COLOR=DarkRed][B]57259[/B][/COLOR], 1> ][/CODE]


All times are UTC. The time now is 21:53.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.