Some Somewhat Easier n^i Sequences Available for Termination - Page 78

EdH · 2022-09-17, 14:17

I'm doing an update run, although I know you haven't entered all the new data. When I try an update, it overwrites the old, which causes me to lose most of the lists I work from if I don't catch the trouble. Actually, I normally find the trouble when all my lists are already trashed. But I work elsewhere. (The fix is simple, but I haven't made it, yet - perhaps today. . .)

EdH · 2022-09-17, 14:22

Quote:

Originally Posted by garambois

I won't be updating this weekend anyway.
Let's wait until the following week for the update.

Although my update gathers data from your tables, the only time I'm hindered is if a table is in work when I access it, and it still shouldn't cause any real issue.

gd_barnes · 2022-09-17, 21:25

Quote:

Originally Posted by garambois

I won't be updating this weekend anyway.
Let's wait until the following week for the update.

Sounds good. I'll have even more stuff to enter by then.

gd_barnes · 2022-09-18, 13:15

109^95 terminates

Quote:

Originally Posted by EdH

I'm going to be tied up tomorrow pretty much all day. I've set the following to run one term: 385^65, 1184^54, 14288^40, 200560490130^14, 7420738134810^14. I'll see how long it takes to get through them. Two should be near a day each. The rest, not too far ahead of that.

It looks like these were all eventually cracked. I have now done my usual work on all of them. I had previously reported 385^65 terminated. 14288^40 is the only somewhat reasonable one and the only one without a factor 3. It has been ECM'd to t40. The other 3 are both large size and have a factor 3. They are ECM'd to t35.

EdH · 2022-09-18, 14:50

I went ahead and queued 14288^40. I'm playing with a bunch of different stuff, but the farm was rather idle, other than a few machines doing isolated things. I estimate ~10 hours for the c155.

gd_barnes · 2022-09-19, 01:27

233^79 terminates

That mostly completes the work on the same-parity > 185 digit effort. I still have a lot to enter into the DB. Getting 5 terminations out of it was unexpected.

Ed, with this effort I found a very challenging one for you that has no factor ECM'd to t45:
113^91 184/184

That will be your next big challenge! :-)

**********

New base 317:

317^51, 53, 57, and 61 terminate

This base was tested a lot prior to Nov. 2018 for both parities. All same-parity exponents < 50 plus 55 and 59 were already terminated long ago. It was very easy to terminate the above.

Only 317^63 and 65 remain. I'm still working on those.

EdH · 2022-09-19, 02:25

I'm confident I could factor a c184 by NFS, but I estimate taking one month to do so. Although I have been part of longer projects, my attention span would be taxed if I ran it with GNFS. It still needs to survive a bunch of ECM prior to any NFS.

I'll catch the updates tomorrow.

What do you think about knocking down some of the index 1s that are referenced in the other thread? We've run this thread for a while now.

gd_barnes · 2022-09-19, 02:34

Quote:

Originally Posted by EdH

What do you think about knocking down some of the index 1s that are referenced in the other thread? We've run this thread for a while now.

Sure. I will be happy to do that. I responded more in the other thread.

After your latest split on 14288^40 and some subsequent work by me, it is now looking quite interesting at 155/121/7. I'm still working on it. I'll terminate it from here if it doesn't hit a C>~140. If not, I'm sure between us we will get it terminated. That will be a nice base to complete.

VBCurtis · 2022-09-19, 02:43

I just finished this morning a GNFS-182. Sieving took 100M cpu-seconds on 2.5Ghz Ivy Bridge cores, and yielded a 21.6M matrix using msieve TD =110 (18.6M matrix using CADO filtering, but the matrix crashed in lingen and was not recoverable). The matrix took 4.5 days to solve after ten weeks of sieving, so I likely oversieved.

I've since learned how to test-sieve in CADO, and I guess we could find 5-10% more speed along with fewer relations, so a C184 would take ~120M cpu-seconds to sieve on that old CPU arch. IMO, too much time by far for this project.

EdH · 2022-09-19, 02:44

I generated a list similar to Karsten's, but mine is a bit out of date as well, since I ran it against my local set. I plan to rewrite it tomorrow with a script that harvests its data from the db. Once we get an initial set, we can simply remove those that pass index 1 and add any new bases as they are added. to the tables.

It kind of looks like we may take on this "extra" project. I'll look into initiating it sometime soon. I'll want some scripts in hand prior to launch.

gd_barnes · 2022-09-19, 02:47

Quote:

Originally Posted by EdH

I generated a list similar to Karsten's, but mine is a bit out of date as well, since I ran it against my local set. I plan to rewrite it tomorrow with a script that harvests its data from the db. Once we get an initial set, we can simply remove those that pass index 1 and add any new bases as they are added. to the tables.

It kind of looks like we may take on this "extra" project. I'll look into initiating it sometime soon. I'll want some scripts in hand prior to launch.

I updated my post to remove the para about the out-of-date list. It looks like Jean-Luc posted an updated one.

2022-09-19, 01:27	#853
gd_barnes "Gary" May 2007 Overland Park, KS 2E3F₁₆ Posts	233^79 terminates That mostly completes the work on the same-parity > 185 digit effort. I still have a lot to enter into the DB. Getting 5 terminations out of it was unexpected. Ed, with this effort I found a very challenging one for you that has no factor ECM'd to t45: 113^91 184/184 That will be your next big challenge! :-) ********** New base 317: 317^51, 53, 57, and 61 terminate This base was tested a lot prior to Nov. 2018 for both parities. All same-parity exponents < 50 plus 55 and 59 were already terminated long ago. It was very easy to terminate the above. Only 317^63 and 65 remain. I'm still working on those. Last fiddled with by gd_barnes on 2022-09-19 at 01:33

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Unexpected termination of PM-1	Miszka	Software	22	2021-11-19 21:36
Easier pi(x) approximation	mathPuzzles	Math	8	2017-05-04 10:58
Would finding a definate Pi value easier if...	xtreme2k	Math	34	2013-09-09 23:54
Aliquot Termination Question - Largest Prime?	EdH	Aliquot Sequences	6	2010-04-06 00:12
A new termination below 100k	10metreh	Aliquot Sequences	0	2010-03-11 18:24

2022-09-17, 14:17	#848
EdH "Ed Hall" Dec 2009 Adirondack Mtns 5×1,051 Posts	I'm doing an update run, although I know you haven't entered all the new data. When I try an update, it overwrites the old, which causes me to lose most of the lists I work from if I don't catch the trouble. Actually, I normally find the trouble when all my lists are already trashed. But I work elsewhere. (The fix is simple, but I haven't made it, yet - perhaps today. . .)

2022-09-18, 14:50	#852
EdH "Ed Hall" Dec 2009 Adirondack Mtns 5·1,051 Posts	I went ahead and queued 14288^40. I'm playing with a bunch of different stuff, but the farm was rather idle, other than a few machines doing isolated things. I estimate ~10 hours for the c155.

2022-09-19, 02:25	#854
EdH "Ed Hall" Dec 2009 Adirondack Mtns 5·1,051 Posts	I'm confident I could factor a c184 by NFS, but I estimate taking one month to do so. Although I have been part of longer projects, my attention span would be taxed if I ran it with GNFS. It still needs to survive a bunch of ECM prior to any NFS. I'll catch the updates tomorrow. What do you think about knocking down some of the index 1s that are referenced in the other thread? We've run this thread for a while now.

2022-09-19, 02:43	#856
VBCurtis "Curtis" Feb 2005 Riverside, CA 2³·3·5·47 Posts	I just finished this morning a GNFS-182. Sieving took 100M cpu-seconds on 2.5Ghz Ivy Bridge cores, and yielded a 21.6M matrix using msieve TD =110 (18.6M matrix using CADO filtering, but the matrix crashed in lingen and was not recoverable). The matrix took 4.5 days to solve after ten weeks of sieving, so I likely oversieved. I've since learned how to test-sieve in CADO, and I guess we could find 5-10% more speed along with fewer relations, so a C184 would take ~120M cpu-seconds to sieve on that old CPU arch. IMO, too much time by far for this project.

2022-09-19, 02:44	#857
EdH "Ed Hall" Dec 2009 Adirondack Mtns 5·1,051 Posts	I generated a list similar to Karsten's, but mine is a bit out of date as well, since I ran it against my local set. I plan to rewrite it tomorrow with a script that harvests its data from the db. Once we get an initial set, we can simply remove those that pass index 1 and add any new bases as they are added. to the tables. It kind of looks like we may take on this "extra" project. I'll look into initiating it sometime soon. I'll want some scripts in hand prior to launch.