2- table discussion
 

Hi,

Not sure if there's a better place to report this, but in case anyone is also working on M1069...

[code]GMP-ECM 6.4.3 [configured with GMP 5.1.0, --enable-asm-redc] [ECM]
Input number is 9655070422115390416186543323849937005579725788253857499949114359491746862455829812455134063238947823422491484330358385786011269349179010346254104507595134837566843566346678652406165295157034775190255938827595701944106198632228542317011888983903368867049017 (256 digits)
Using B1=260000000, B2=3178559884516, polynomial Dickson(30), sigma=3725672826
Step 1 took 1891828ms
Step 2 took 688319ms
********** Factor found in step 2: 5557036167944892502666285821951871600803581019193074182942021552512721
Found probable prime factor of 70 digits: 5557036167944892502666285821951871600803581019193074182942021552512721
Probable prime cofactor 1737449627880690251595720971509773434839416433962933088550256363151433564424184235036716533956588433095081958412139739283774089341095270403988232339045024984541371228013713117350142329577 has 187 digits
Report your potential champion to Richard Brent <champs@rpbrent.com>
(see http://wwwmaths.anu.edu.au/~brent/ftp/champs.txt)[/code]

congratultation again Ryanp; that's really nice.

[QUOTE=ryanp;348037]Hi,

Not sure if there's a better place to report this, but in case anyone is also working on M1069...

[code]GMP-ECM 6.4.3 [configured with GMP 5.1.0, --enable-asm-redc] [ECM]
Input number is 9655070422115390416186543323849937005579725788253857499949114359491746862455829812455134063238947823422491484330358385786011269349179010346254104507595134837566843566346678652406165295157034775190255938827595701944106198632228542317011888983903368867049017 (256 digits)
Using B1=260000000, B2=3178559884516, polynomial Dickson(30), sigma=3725672826
Step 1 took 1891828ms
Step 2 took 688319ms
********** Factor found in step 2: 5557036167944892502666285821951871600803581019193074182942021552512721
Found probable prime factor of 70 digits: 5557036167944892502666285821951871600803581019193074182942021552512721
Probable prime cofactor 1737449627880690251595720971509773434839416433962933088550256363151433564424184235036716533956588433095081958412139739283774089341095270403988232339045024984541371228013713117350142329577 has 187 digits
Report your potential champion to Richard Brent <champs@rpbrent.com>
(see http://wwwmaths.anu.edu.au/~brent/ftp/champs.txt)[/code][/QUOTE]

Very nice! This is one of the exponents not mentioned in the June version
of the EPFL report on the PS3 ECM work; but included in the subsequent
updated version: "for exponents M in [1000, 1125] not stated in Table ...
25000 curves were run"; which was a t65. No reason to consider this
as a miss; t70 was 110000 curves. Notice the p60 limits in Ryanp's
report. At C256 this wouldn't have been a likely candidate for the potential
snfs run mentioned as a possible use of the ECM pretests.

We need three more p7x's to clear the p6x's from the 2013 Top10; unless
there was another one that hasn't been reported yet?

-Bruce

One more tonight. This is M1051.

[code]GMP-ECM 6.4.3 [configured with GMP 5.1.0, --enable-asm-redc] [ECM]
Input number is 31279631061794613727861481049372603628390044054774072902880549940031778948907759788327766142243129226647466065137379540166638486318363027844304119015373713848508431069580503916147434049334568476694360380165638480251191661287591464489921182753644001 (248 digits)
Using special division for factor of 2^1051-1
Using B1=2900000000, B2=90975170616348, polynomial Dickson(30), sigma=856037322
dF=1048576, k=7, d=11741730, d2=19, i0=228
Expected number of curves to find a factor of n digits:
35 40 45 50 55 60 65 70 75 80
10 29 90 313 1191 4942 21919 103850 525670 2794661
Step 1 took 15371609ms
Using 30 small primes for NTT
Estimated memory usage: 5641M
Initializing tables of differences for F took 5846ms
Computing roots of F took 270388ms
Building F from its roots took 94229ms
Computing 1/F took 33866ms
Initializing table of differences for G took 2513ms
Computing roots of G took 270439ms
Building G from its roots took 97176ms
Computing roots of G took 276942ms
Building G from its roots took 97969ms
Computing G * H took 17910ms
Reducing G * H mod F took 18078ms
Computing roots of G took 277986ms
Building G from its roots took 97097ms
Computing G * H took 17564ms
Reducing G * H mod F took 17805ms
Computing roots of G took 275209ms
Building G from its roots took 97220ms
Computing G * H took 17799ms
Reducing G * H mod F took 18045ms
Computing roots of G took 275833ms
Building G from its roots took 96765ms
Computing G * H took 17655ms
Reducing G * H mod F took 18054ms
Computing roots of G took 275734ms
Building G from its roots took 96657ms
Computing G * H took 17512ms
Reducing G * H mod F took 17754ms
Computing roots of G took 274783ms
Building G from its roots took 97155ms
Computing G * H took 17488ms
Reducing G * H mod F took 17779ms
Computing polyeval(F,G) took 169647ms
Computing product of all F(g_i) took 654ms
Step 2 took 3401418ms
********** Factor found in step 2: 305017906063256842921494808558019733856326299412534951989303214657199
Found probable prime factor of 69 digits: 305017906063256842921494808558019733856326299412534951989303214657199
Probable prime cofactor 102550146860255526101297388116512688405087055342042127103734791437428154918641834551538564078084763246635542280778334443750922085574212074530157338374245476768304982405585875058799 has 180 digits[/code]

Very impressive. Kudos!

EDIT: someone cracked 2,1067-, too. (And also a hole-in-one. Lucky, hmm...) Sweet mother of Jesus!

[QUOTE=Batalov;348830]Very impressive. Kudos!

EDIT: someone cracked 2,1067-, too. (And also a hole-in-one. Lucky, hmm...) Sweet mother of Jesus![/QUOTE]

someone?? Who did it ???

Were these numbers not sufficiently pre-tested perhaps?

[QUOTE=R.D. Silverman;348852]someone?? Who did it ???

Were these numbers not sufficiently pre-tested perhaps?[/QUOTE]

The same comments apply to this p69 from M1051 as to the
p70 from M1069; epfl ran a t65, 25K 3e9 curves; t70 would have
been 110K 3e9 curves.

The 2,1067- C260 factor hasn't been reported on either Sam's or
Paul's page. Google doesn't help ...

-Bruce

(Meanwhile, Ryanp is also tearing up the SNFS list - 16 factorizations
on page 126, already; another four reservations. In other news,
the first two NFS@Home gnfs numbers both had a p62 factor, after
a t60.)

[QUOTE=bdodson;348870]The same comments apply to this p69 from M1051 as to the
p70 from M1069; epfl ran a t65, 25K 3e9 curves; t70 would have
been 110K 3e9 curves.
[/QUOTE]

Certainly. Plus the terrific work you did, plus the (very small, I admit) work I
did, plus the work everyone else did that was reported on the GIMPS ECM
web page, plus the work that was never reported, etc. etc.

[QUOTE=R.D. Silverman;348872]Certainly. Plus the terrific work you did, plus the (very small, I admit) work I
did, plus the work everyone else did that was reported on the GIMPS ECM
web page, plus the work that was never reported, etc. etc.[/QUOTE]

I would imagine that Bos & Co. did some pre-testing as well.......

[QUOTE=bdodson;348870]The same comments apply to this p69 from M1051 as to the
p70 from M1069; epfl ran a t65, 25K 3e9 curves; t70 would have
been 110K 3e9 curves.

The 2,1067- C260 factor hasn't been reported on either Sam's or
Paul's page. Google doesn't help ...

-Bruce

(Meanwhile, Ryanp is also tearing up the SNFS list - 16 factorizations
on page 126, already; another four reservations. In other news,
the first two NFS@Home gnfs numbers both had a p62 factor, after
a t60.)[/QUOTE]
See [url]http://factordb.com/index.php?query=2%5E1067-1[/url]
(and if you click on the p68 and the on "More information", it will be reported that it was added sometime on the 4th of Aug., but not "who done it.")

By the MOA (and size), this factor seems to be also Ryan's.

Ryan, the Paul Zimmermann's [URL="http://www.loria.fr/~zimmerma/ecmnet/facform.html"]report form[/URL] is easy to use and gets instantly seen by other interested parties. Your last p70 seems to have been inserted by someone else, and that someone forgot to report the sigma value - see [URL="http://www.loria.fr/~zimmerma/cgi-bin/last.cgi?date"]bottom of this list[/URL].

[QUOTE=Batalov;348884]See [url]http://factordb.com/index.php?query=2%5E1067-1[/url]
(and if you click on the p68 and the on "More information", it will be reported that it was added sometime on the 4th of Aug., but not "who done it.")

By the MOA (and size), this factor seems to be also Ryan's.

Ryan, the Paul Zimmermann's [URL="http://www.loria.fr/~zimmerma/ecmnet/facform.html"]report form[/URL] is easy to use and gets instantly seen by other interested parties. Your last p70 seems to have been inserted by someone else, and that someone forgot to report the sigma value - see [URL="http://www.loria.fr/~zimmerma/cgi-bin/last.cgi?date"]bottom of this list[/URL].[/QUOTE]

Sam seems not to have received either factor. Someone should let Bos et.al.
know so they can discontinue their NFS work on M1051.