Recommended bases and efforts - Page 6

gd_barnes · 2012-10-14, 06:05

As a general rule, bases of the form 2^q-1 are more prime than usual and have a very high conjecture so I thought it might be interesting to pick out the lowest conjectured such base, test it to n=2500, and recommend it. Therefore I have added R1023 with CK=68394 and 724 k's remaining at n=2500 to the recommended list. As usual the k's remaining are in the file attached to the first post.

gd_barnes · 2012-10-22, 06:40

I have added new bases R145 and R171 tested to n=2500 and recommended to n=25K to the list. Completion of these would put all bases < 190 with a CK < 1M at n=25K. They are not nearly as difficult as many of the bases already recommended. R171 has the fewest k's remaining on the new base list and R145 has a smaller conjecture than S145, which has already been tested to n=25K.

As usual, the k's remaining at n=2500 are attached to the first post.

I also removed S241 from the list. With 4000+ k's remaining at n=2500, it's way too difficult and doesn't accomplish any range completion. It still has a sieve file on the reservations page.

firejuggler · 2012-11-23, 20:02

hi! I have one Idle core. do you think S316 will take less than one week? or should I try R383?

gd_barnes · 2012-11-23, 21:50

Quote:

Originally Posted by firejuggler

hi! I have one Idle core. do you think S316 will take less than one week? or should I try R383?

No. S316 would take several CPU months. R383 has 9 k's remaining at n=25K. It would take several CPU weeks to test it to n=50K.

Let me give an example: I have been working on testing many bases with 5 k's remaining from n=25K to 50K. Running 40+ cores, it takes ~7-8 hours for each base...not including sieving. That means ~300 CPU hours for each base or just a little under 2 CPU weeks. Some of my CPU's are a little slow. Including sieving and testing on R383 with almost twice as many k's on a very modern machine is still likely to take you ~3 weeks assuming that you find the expected number of primes...about 2.

gd_barnes · 2012-11-24, 03:00

Quote:

Originally Posted by gd_barnes

No. S316 would take several CPU months. R383 has 9 k's remaining at n=25K. It would take several CPU weeks to test it to n=50K.

I stand corrected. I see that R383 is the lowest difficulty Riesel base and so its k's are likely half the average weight of the average base. You could likely do it in 1-1/2 to 2 weeks on one core. Don't quote me on that. That's a rough extrapolation based on my experience with average-weight bases with 5 k's remaining.

gd_barnes · 2013-01-26, 00:29

I've added S283 to the list for n=10K-25K. It is the only unreserved base at n>=10K that has not been searched to n>=25K.

firejuggler · 2013-01-26, 00:43

if nobody take it before I end my current reservation (S488, @ 93000=>100000, 2 k removed for now, a third is unlikely, 7 k left) i'll take it

gd_barnes · 2013-01-26, 08:00

Quote:

Originally Posted by firejuggler

if nobody take it before I end my current reservation (S488, @ 93000=>100000, 2 k removed for now, a third is unlikely, 7 k left) i'll take it

I'll mark it down for you.

gd_barnes · 2013-04-13, 02:51

C10ck3r searched R331 to n=20K and released so I have left it recommended for n=20K-25K.

gd_barnes · 2013-06-24, 07:08

I have added 8 new recommended bases in the 1st post with conjected k's of 10000-20000 tested to n=2500 and sieved for n=2500-25K by Ian. Links to k's remaining are in the 1st post and links to sieve files are on the reservations pages.

gd_barnes · 2013-07-09, 04:35

Added S642 recommended for n=10K-25K searched to n=10K by firejuggler.

2012-10-22, 06:40	#57
gd_barnes May 2007 Kansas; USA 101×103 Posts	I have added new bases R145 and R171 tested to n=2500 and recommended to n=25K to the list. Completion of these would put all bases < 190 with a CK < 1M at n=25K. They are not nearly as difficult as many of the bases already recommended. R171 has the fewest k's remaining on the new base list and R145 has a smaller conjecture than S145, which has already been tested to n=25K. As usual, the k's remaining at n=2500 are attached to the first post. I also removed S241 from the list. With 4000+ k's remaining at n=2500, it's way too difficult and doesn't accomplish any range completion. It still has a sieve file on the reservations page. Last fiddled with by gd_barnes on 2012-10-22 at 06:41

2012-11-23, 20:02	#58
firejuggler Apr 2010 Over the rainbow 19·137 Posts	hi! I have one Idle core. do you think S316 will take less than one week? or should I try R383? Last fiddled with by firejuggler on 2012-11-23 at 20:04

2013-01-26, 00:43	#62
firejuggler Apr 2010 Over the rainbow 19·137 Posts	if nobody take it before I end my current reservation (S488, @ 93000=>100000, 2 k removed for now, a third is unlikely, 7 k left) i'll take it Last fiddled with by firejuggler on 2013-01-26 at 00:44

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Cunningham ECM efforts	pinhodecarlos	Cunningham Tables	7	2017-12-21 13:29
Doublecheck efforts; S66/S79 to start with	gd_barnes	Conjectures 'R Us	16	2014-08-07 02:11
ECM efforts mistake?	10metreh	mersennewiki	1	2008-12-28 13:31
Recommended TF bit levels for M(>10^8)	NBtarheel_33	Math	19	2008-11-03 17:19
ECM Efforts	R.D. Silverman	Factoring	63	2005-06-24 13:41

2012-10-14, 06:05	#56
gd_barnes May 2007 Kansas; USA 101·103 Posts	As a general rule, bases of the form 2^q-1 are more prime than usual and have a very high conjecture so I thought it might be interesting to pick out the lowest conjectured such base, test it to n=2500, and recommend it. Therefore I have added R1023 with CK=68394 and 724 k's remaining at n=2500 to the recommended list. As usual the k's remaining are in the file attached to the first post.

2013-01-26, 00:29	#61
gd_barnes May 2007 Kansas; USA 101·103 Posts	I've added S283 to the list for n=10K-25K. It is the only unreserved base at n>=10K that has not been searched to n>=25K.

2013-04-13, 02:51	#64
gd_barnes May 2007 Kansas; USA 101·103 Posts	C10ck3r searched R331 to n=20K and released so I have left it recommended for n=20K-25K.

2013-06-24, 07:08	#65
gd_barnes May 2007 Kansas; USA 101×103 Posts	I have added 8 new recommended bases in the 1st post with conjected k's of 10000-20000 tested to n=2500 and sieved for n=2500-25K by Ian. Links to k's remaining are in the 1st post and links to sieve files are on the reservations pages.

2013-07-09, 04:35	#66
gd_barnes May 2007 Kansas; USA 101×103 Posts	Added S642 recommended for n=10K-25K searched to n=10K by firejuggler.