Have a look at the partition numbers

[smh]

Reserving 14735 c104 (GNFS)

[smh]

Originally Posted by smh

Reserving 14735 c104 (GNFS)

N=12332341160200760366847722782949606825270758700381531837597837316250818326317511743110920063409242258717  ( 104 digits)
Divisors found:
 r1=1485143488073298293923414948921496082307837993 (pp46)
 r2=8303804487066576064650357194157798232044951246095567680469 (pp58)
Total time: 7.88 hours.

[R.D. Silverman]

Originally Posted by bdodson

As I recall, Paul Zimmermann and Arjen are still waiting for the
last quarter's prize. We quit when the prizes were no longer
being given. But my recollection is that there was another round
of factors waiting for the quarter to finish (to be eligible for the next
quarter's prize), and that some of the later ones that we did
submit never got included.

I was handling the verification of results and saw to it that the
marketing people mailed out the prizes.

Then I got laid off. My guess is that noone else took over and that
it fell through the cracks.......

[Andi47]

Originally Posted by Andi47

Reserving 28819 c102

Fri May 18 20:35:28 2007 prp46 factor: 2896980316642729353055002268196586942743872217
Fri May 18 20:35:28 2007 prp56 factor: 76503441441413011494575167031398404085894766827617190241

approx. 17 cpu-hours with msieve (QS)


Edit: Reserving 29130 c103

[fivemack]

Can I just check that people who find factors are sending them to kc2h-msm@asahi-net.or.jp directly, as well as posting them here?

[smh]

Originally Posted by fivemack

Can I just check that people who find factors are sending them to kc2h-msm@asahi-net.or.jp directly, as well as posting them here?

Not yet, but i will.

I would also like to reserve ALL remaining (the ones not reserved in this thread) 103 and 104 digit composites.

[xilman]

Originally Posted by R.D. Silverman

I was handling the verification of results and saw to it that the
marketing people mailed out the prizes.

Then I got laid off. My guess is that noone else took over and that
it fell through the cracks.......

Somewhere I still have a record of the partition factors from that computation.

I made a few hundred bucks over the course of several years. Not a way to get rich, especially not if the proceeds are donated to charity.

Eventually, I got tired of the gamesmanship implicit in the structure of the challenge.

Paul

[Andi47]

Originally Posted by Andi47

Reserving 14984 c104 for msieve (QS)

Sun May 20 18:37:19 2007 prp45 factor: 118510482144290848347459027738323943707271259
Sun May 20 18:37:19 2007 prp60 factor: 102892983755174194924983528561338993128741779676166226537677

[bdodson]

Originally Posted by fivemack

That data has already been merged into Mishima's tables; at least, I looked at http://www.ontko.com/~rayo/primes/ to get the RSA-challenge data, then grepped the Mishima tables for some of the largish factors from http://www.ontko.com/~rayo/primes/hr_p135.txt and found them all.

Not exactly. Ray Ontko's web page has "honor rolls" up to 165-169 digits;
but our factors went past that --- without looking too closely, I see
one at 181-digits

p(27767)=2222164031682059894256456152196952624981714302844666210807\
1623414759364252901862750351685028160708519432209308498756197660187\
55775489619728333950278651217220247620722925992858424104 (181 digits)
Factors: 2 * 2 * 2 * 3 * 7 * 7 * 143982495659 * 911432531209 * 275433214610326301 * 292795868758239438861071867242457 * 1785473110332531860394372415833097856861274839468200152025175114737\
01604587588088467419299970751228925337
Date: March 14, 2000
Method: trial division, ecm (p33, found December 1999)
Name: Bruce Dodson, Arjen K. Lenstra, Paul Zimmermann

Anyone have 170-174, 175-179 and maybe 180-? The contest
only covered maybe 40-digits? It's been a long time. 10-digits for
8-point numbers, 10-digits for 4-points, 10-digits for 2-points and
20-digits for "1-pointers"? For example, iirc, p(11087) was the smallest
unfactored prime index partition number when we factored it, and it
had 113-digits. That meant 113-digit to 122-digit partition numbers
of prime index were worth 8-points each, then 123- to 132-digits
4-points, 133- for 2-pts, then 143- to 162-digit numbers were worth
1-point. Nothing for larger numbers. So when Arjen sent in the factorization,
on the last day of that quarter's challenge, December 31, 1993 in this
case, that meant that there was a new smallest unfactored size ---
for eg, the 2nd largest unfactored might have been 115-digits, which
opened up two new digits 163- and 164-digits --- worth 0-points on
Dec 31, but 1-point on Jan 1st (point ranges only changing on the
Quarters). We knew that, but no one else did, and prepared easy new
1-point numbers before the end of the quarter (while they were worth 0),
and sent them in on Jan 1 (when they were worth 1-point). As you
can see, xilman's still annoyed; but Arjen took a large amount of pleasure
winning most Quarters. Someone at RSA Data's idea, not our fault. Really.
There was a Quarter when someone anticipated this, and sent in previous
0-point numbers, when we hadn't factored the smallest number, so the
effort was wasted, as they were still worth 0-points. Peter was a fan of
sending in new numbers before Arjen got to them; well ...

Ah, perhaps it's comming back to me, there may not have been an
honor roll for 180-digits? We'd factored enough smallest index numbers
so that 181-digit was supposed to count (RSADSI's rules), but I don't
recall whether we ran out of honor rolls, there at the end. But there
were honor rolls of 170-174 and for 175-179, which aren't on Ontko's
page --- looks like I ought to be able to dig out our reports from
170-181 digits, but the Honor rolls would have other ones aside from
the ones that we found -- it would be better if someone has them. -Bruce

PS -- Ah, that p33 in the RSA-report above is already listed; probably
most of the other easy ones as well. Oh. But Tom's right about the
conguences; looks like the theory's set now (although it wasn't when
Hendrik suggested partition numbers as challenge numbers). Seems that
Ono only gave one new example; then assigned the question for under-
grad research --- Weaver got 76065 new congruences, but the largest
prime divisor was 31 --- using modular forms of level 576. If we wanted
10-digit prime factors of partition numbers, the level would likely be hugh,
computing coef impossible, and the congurences wouldn't start until (very)
large index. I don't see 17303n+237 on her list though, is this more
recent? Lastly, my Mother's side of the family is quite proud of being
descended from someone that went west on the original 'Trail, some five
generations ago. Ought to be irrelevant, but UofO's office of financial
aid was quite annoyed to find someone that actually met the conditions
set by one of their donors; got me through the first two years of college.

[smh]

Originally Posted by smh

I would also like to reserve ALL remaining (the ones not reserved in this thread) 103 and 104 digit composites.

Done. I'll send them later.

I'll continue with the 105 digit composites.

[fivemack]

Originally Posted by smh

Done. I'll send them later.

I'll continue with the 105 digit composites.

That's quite impressive: 45 factorisations of about eight hours each in three days, so at least five full-time CPUs. Do you have a patch for the "lambda-comp" bug during polynomial selection that has thwarted my attempts at running completely automated ggnfs?

I suppose that with that level of automation there's little point in even running ECM first; I'm finding that about one in ten of the 20000..21000 cofactors produces a factor after 150 curves at 3e6, fairly independent of cofactor size, which means I get factors at about the same rate as if I were NFSing starting at the smallest ones. Mostly I'm NFSing the C100..120 numbers that appear when you break off a P3x from a composite (and MPQSing the occasional C8x, but that takes less than an hour); that's enough to fill up my compute resources nicely ... I still sometimes get an NFS factor smaller than the ECM factor.