mersenneforum.org Current status
 Register FAQ Search Today's Posts Mark Forums Read

2008-05-29, 15:11   #34
R.D. Silverman

Nov 2003

22·5·373 Posts

Quote:
 Originally Posted by frmky The factorization of 7,271- is complete. The sieving for this one went faster than we expected, and therefore it was very oversieved. I initially filtered only the relations I had here in California to see the size of the matrix created. Here's that filtering log: Greg

Nice result......

I am about half way through the LA for 2,1101+, half-way through the
sieving for 2,1104+ and will then do 2,1538M. Sam Wagstaff does not show
anyone doing 2,799+. Is anyone doing it or may I reserve it?

2008-05-29, 15:17   #35
bsquared

"Ben"
Feb 2007

357810 Posts

Quote:
 Originally Posted by R.D. Silverman Nice result...... I am about half way through the LA for 2,1101+, half-way through the sieving for 2,1104+ and will then do 2,1538M. Sam Wagstaff does not show anyone doing 2,799+. Is anyone doing it or may I reserve it?
I'm doing it. Sieving is 95% done. Sorry, I never reserved it with Sam Wagstaff.

Last fiddled with by bsquared on 2008-05-29 at 15:18

2008-05-29, 17:36   #36
R.D. Silverman

Nov 2003

1D2416 Posts

Quote:
 Originally Posted by bsquared I'm doing it. Sieving is 95% done. Sorry, I never reserved it with Sam Wagstaff.

Thanks for the info.

2008-05-30, 14:52   #37
R.D. Silverman

Nov 2003

22·5·373 Posts

Quote:
 Originally Posted by bsquared I'm doing it. Sieving is 95% done. Sorry, I never reserved it with Sam Wagstaff.

The student, of course, is 100% correct. The question did not
say "compute the value of x". The teacher needs to be more careful......

2008-05-30, 15:09   #38
bsquared

"Ben"
Feb 2007

1101111110102 Posts

Quote:
 Originally Posted by R.D. Silverman I like your Avatar....
Thanks! I ran across that and several others like it in a mass e-mail awhile back (attached).

Quote:
 Originally Posted by R.D. Silverman I like your Avatar.... The student, of course, is 100% correct. The question did not say "compute the value of x". The teacher needs to be more careful......
Agreed. Nonetheless, something tells me that the teacher did not award any creativity points to the student in this case...

- ben.
Attached Files
 solutions.zip (168.0 KB, 269 views)

2008-05-30, 15:10   #39
bsquared

"Ben"
Feb 2007

2·1,789 Posts

More that didn't fit in previous post...
Attached Files
 solutions2.zip (222.6 KB, 259 views)

 2008-05-30, 19:54 #40 bsquared     "Ben" Feb 2007 67728 Posts I'd like to reserve 2,1598L for SNFS. SNFS seems to be the way to go, especially since I now have experience with SNFS on this size number (namely, 2,799+), while I have none with GNFS on a C171. I'm still playing around with parameterization, using this poly Code:  n: 344682277495113483576737243789457319844973285487058684789549245491555989770819433641425904625087602042261741581803359854653090524174719942048090727106451863564668263059909 skew: 1 c6: 2 c3: -2 c0: 1 Y1: -1 Y0: 10889035741470030830827987437816582766592 - ben. p.s. I've sent email to Sam Wagstaff about both this and 2,799+. p.p.s Is this the best place to post such a reservation, or should I start a new thread?
2008-05-31, 00:37   #41
frmky

Jul 2003
So Cal

42148 Posts

Quote:
 Originally Posted by bsquared I'd like to reserve 2,1598L for SNFS.
Numbers like this always frustrate me. I keep thinking there surely must be a way to use an algebraic factor like, for this one,

65536x^736+65536x^713 + 32768x^690 - 16384x^644 - 16384x^621 - 8192x^598 + 4096x^552 + 4096x^529 + 2048x^506 - 1024x^460 - 1024x^437 - 512x^414 + 256x^368 - 128x^322 - 128x^299 - 64x^276 + 32x^230 + 32x^207 + 16x^184 - 8x^138 - 8x^115 - 4x^92 + 2x^46 + 2x^23 + 1

Degree 32 (or 16) polys are useless, though.

Quote:
 Originally Posted by bsquared p.p.s Is this the best place to post such a reservation, or should I start a new thread?
They don't need to be reserved here at all, just with Sam.

Greg

2008-05-31, 02:26   #42
bsquared

"Ben"
Feb 2007

1101111110102 Posts

Quote:
 Originally Posted by frmky Numbers like this always frustrate me. I keep thinking there surely must be a way to use an algebraic factor... Greg
Yeah, I know how you feel. I had pari print out the first 90 some splittings, and stared at that same one for a while.

something like
for(n=2,99,print(n,",",1598\n,",",factor(2^(1598%n)*x^n+1)))

2008-06-04, 17:04   #43
R.D. Silverman

Nov 2003

22×5×373 Posts

Quote:
 Originally Posted by bdodson The NFSNET resources are mostly Richard's and Greg's; the two 10,257's (minus and plus) at difficulty 257.
Did you do ECM pre-testing on 10,257-? While the C241 had 3 factors,
and would have required NFS to finish in any case, should the smaller
factor (a p54) be considered an ECM miss?

2008-06-04, 17:42   #44
frmky

Jul 2003
So Cal

22·547 Posts

Quote:
 Originally Posted by R.D. Silverman Did you do ECM pre-testing on 10,257-? While the C241 had 3 factors, and would have required NFS to finish in any case, should the smaller factor (a p54) be considered an ECM miss?
Probaby so. From Bruce,

Quote:
 I have the pre-test recorded as 7*t50 (which refers to the effort, not the limits; almost all of the curves were either B1=110M (p55-optimal) or B1=260M (p60-optimal)). That's supposed to be somewhere in the range between t55 and 2*t55; which means that the chances of finding a known p55 are between 63.2% and 86.4%, midrange might be 1-exp(-1.5) = 77.68%, so I spent an effort sufficient to find 3 of 4 known p55's, leaving 1 of 4 to be found by sieving. There've certainly been lots of p54-p56's found; and among recent sieving candidates I found a p53 (on-schedule) and a p57 (early).
Greg

 Similar Threads Thread Thread Starter Forum Replies Last Post petrw1 PrimeNet 20 2007-05-24 18:10 fivemack NFSNET Discussion 90 2006-11-13 13:37 moo LMH > 100M 0 2006-09-02 01:15 HiddenWarrior Operation Billion Digits 1 2005-08-19 21:42 rogue Cunningham Tables 4 2005-06-10 18:28

All times are UTC. The time now is 17:33.

Sat Oct 16 17:33:59 UTC 2021 up 85 days, 12:02, 1 user, load averages: 1.03, 1.16, 1.30