|
Base 479 can be added at the next update.
|
I have completed testing of new base 479 all open sequences to >= 120 digits.
|
457^46 terminated
457^46 terminated after ~1580 more indices in p=601.
|
179^42 terminates
|
I added an index to 2^590. The C160 is ECM'd to t40.
When I released base 2, I had stated that all cofactors were C>=142. I had missed this one that was at C138. |
173^44 terminates
|
Base 487 can be added at the next update.
|
I have completed testing of new base 487 all open sequences to >= 120 digits.
|
I have completed testing of all bases > 170 all open sequences to >= 120 digits.
For this cycle, completed were bases in the 170 to 200 range. This included testing of bases 173, 179, 181, 191, 193, 197, and 199. Also completed were numerous new bases as previously reported in order to maintain the "all bases greater than XXX at >=120 digits" criteria. After going a long time without a termination/merge, this range had 2 of them. Work continues backwards by base for bases 170 to 150. There are 4 bases in this range with sequences at < 120 digits. Sequences disposed of with this effort so far: Terminations: 173^44 179^42 251^28 257^22 283^20 (cycle) 359^14 359^18 828^19 14536^19 Merges: 263^24 317^8 373^44 |
A lot of thanks to all for your work !
This weekend, there will be no update. Indeed, I sent the complete original file to Karsten who is doing a lot of work on it : corrections of errors, imperfections. He might also add some functions to sort the data, if it is possible. Finally, he will try to make the whole thing a bit more user-friendly. And if I do an update in the folder on my computer, there might be some old beugs in it. So I'll update in the new folder when Karsten has finished his work and sent me back this new, complete, corrected and improved folder. Thank you all for your understanding. And most of all, thanks to Karsten for the big work in progress ! |
[SIZE="3"][U]Big update of the page done.[/U][/SIZE]
Many thanks to all for your work. And a huge thanks to Karsten who did all the work this time, and much more, because the page now offers new possibilities. :smile: [B]Added bases : 479, 487. All the bases have been updated. New data sorting features have been added : lists sorted in different ways, open sequences, sequences ending in prime numbers, cycles and many others...[/B] Please report any problems you may encounter. |
I like the new look and operation! I did need to make a minor change to a script I use for one of the threads, but it is better handled now. :tu: to Karsten and Jean-Luc for all the work! Thanks Guys!
|
[QUOTE=EdH;624026]I like the new look and operation! I did need to make a minor change to a script I use for one of the threads, but it is better handled now. :tu: to Karsten and Jean-Luc for all the work! Thanks Guys![/QUOTE]
It takes some work to do that. Thank you for doing that. The movies are one thing. Doing it in the actual is something quite different. Thank you. |
The new look is great! I especially like the new data files section. I also really like the fact that the 1 is now on the far left and the 0 is on the far right for the exponents. I was often confused the other way. It always meant that there was the final straggling sequence at the end that was on its own line whenever the final exponent was divisible by 20. Also since there is no exponent 0, it always meant there was a blank on the beginning of the first line. Nice work!
One problem I see that may not be a problem: Initials for termination are not showing up on the new bases 479 and 487. Perhaps Jean-Luc needs to enter those manually. |
For me, the row labels should still end in zero. If I look in row 100 and column 2, I get exponent 102. Saying "row 101" seems less elegant, even though that number represents the first entry in that row.
|
[QUOTE=VBCurtis;624032]For me, the row labels should still end in zero. If I look in row 100 and column 2, I get exponent 102. Saying "row 101" seems less elegant, even though that number represents the first entry in that row.[/QUOTE]
Ha! I didn't even see that change. It took me a second to figure out what you meant. Yes, I agree. The row labels should be 0,20,40,etc. like they were before instead of 1,21,41,etc. In other words, column label plus row label should equal exponent. |
[QUOTE=gd_barnes;624029]
One problem I see that may not be a problem: Initials for termination are not showing up on the new bases 479 and 487. Perhaps Jean-Luc needs to enter those manually.[/QUOTE] [QUOTE=VBCurtis;624032]For me, the row labels should still end in zero. If I look in row 100 and column 2, I get exponent 102. Saying "row 101" seems less elegant, even though that number represents the first entry in that row.[/QUOTE] OK, we're working on that, but new bugs have popped up. We are working on it... [QUOTE=EdH;624026]I like the new look and operation! I did need to make a minor change to a script I use for one of the threads, but it is better handled now. :tu: to Karsten and Jean-Luc for all the work! Thanks Guys![/QUOTE] Honestly, it was Karsten who did all this work ! Thanks a lot to him, once again ! |
Bases 491 & 499 can be added at the next update.
This completes my initialization run. |
I have completed testing of new bases 491 and 499 all open sequences to >= 120 digits.
|
Page updated.
Thanks to all. [B]Added bases : 491, 499. Update of some bases, see below.[/B] [QUOTE=gd_barnes;624029] One problem I see that may not be a problem: Initials for termination are not showing up on the new bases 479 and 487. Perhaps Jean-Luc needs to enter those manually.[/QUOTE] Done. But I'm sorry, there may be attribution errors for these two bases, as I no longer had access to these attributions in Edwin's first thread post. [QUOTE=VBCurtis;624032]For me, the row labels should still end in zero. If I look in row 100 and column 2, I get exponent 102. Saying "row 101" seems less elegant, even though that number represents the first entry in that row.[/QUOTE] Done only for the updated bases : please indicate if the correction is appropriate (for example by looking at bases 2 to 10). A full update will be done in about a week. We are still in the verification phase. Please report any problems you find. |
Yes, looks good so far.
One word to he new datafile for merges: I first had an option to create all datafiles with or without any merged but thought it's easier to create an own file for them. So all datafiles are without merges. If any merge should be found there, this has to be marked by hand from Jean-Luc and the page for the base with the new merge and the datafiles has to be updated. The three files for merges also contain the merged aliquot seq (different from the other datafiles) to easily search for special merge-seqs and also to 3rd file is sorted by it. When I got the current data from Jean-Luc and after changing the scripts I also tried to update the new entries/terminations. But this was very timeconsuming. So my suggestion is: Create a new thead, where only change/terminations in short should be posted without any prosa like the reservation thread from the main alisqout project. Like [code] reserve 10^20 11^21 drop 10^21 11^22 terminate 10^22 update 12^12 new 999 [/code] So to keep it easier for JL (homage to Picard :-) only show the certain seqs you worked on or the new base to fill in. A new base has some more work to do, because also the main page has to be changed for automatic creation. Sure any change of a seq which will only be updated in FactorDB but not given in that thread wil only show up for the next big update by JL. For almost 20k seqs currently where 8.8k are open it's hard to keep an overview for JL. I remember his first page: there he had to update every change by hand. Now the great amount of seqs is a tremendous work to keep them current. Another problem is the new restriction for FactorDB: to update the whole 9k open+merge seqs needs about 7.8M database queries by FactorDB so he needs min. 5 hours for the whole update. Another hint: The datafile with "no 2" contains about 780 seqs which should (almost) easliy terminate: the lowest is given with a C131 so some low hanging fruits still there. Suggestions for the new thread? |
[QUOTE=garambois;624228]<snip>
Done. But I'm sorry, there may be attribution errors for these two bases, as I no longer had access to these attributions in Edwin's first thread post. Done only for the updated bases : please indicate if the correction is appropriate (for example by looking at bases 2 to 10). A full update will be done in about a week. We are still in the verification phase. Please report any problems you find.[/QUOTE] The attributions on bases 479, 487, 491, & 499 look correct for what I did. I see that base 177792 has question marks for exponents 30, 32, & 34. Those were terminated by me a couple of weeks ago per post 1108 in the other thread. Nice work on everything! |
[QUOTE=kar_bon;624230]<snip>
Another hint: The datafile with "no 2" contains about 780 seqs which should (almost) easily terminate: the lowest is given with a C131 so some low hanging fruits still there. Suggestions for the new thread?[/QUOTE] That one would have been in the works at the time you took your data. Previously between Ed and me, we quickly factored anything that popped up that was C<136. With Ed's recent batch of splits and my follow-up everything is now C>= 140. If you see anything lower, it's being worked on unless Yoyo or someone else factors one that we are not aware of. No low hanging fruit left for the trivial sequences. :-) |
The datafiles are taken from the current data Jean-Luc shows in his main page.
So again it's more helpful for him to give him the bases/index which worked on instead of giving "all C<140" are done. A complete update is a tremendous work for one person and only doable in several hours. Please support him with more clear and short information. |
@kar_bon: If [C]terminate n^i[/C] is posted, does that automatically credit the poster? Are the user ids associated with credit ids?
If Gary terminates an unreserved matched parity sequence, which I track in the other thread, should he post it in this thread to update the sequence automatically? Instead of keeping track in the other threads, as I've been doing, should I let terminations be done in this thread now, by the "Terminator?" |
[QUOTE=kar_bon;624241]The datafiles are taken from the current data Jean-Luc shows in his main page.
So again it's more helpful for him to give him the bases/index which worked on instead of giving "all C<140" are done. A complete update is a tremendous work for one person and only doable in several hours. Please support him with more clear and short information.[/QUOTE] Karsten, Support him with more clear and short information? What are you talking about? You are lecturing and completely missed my point. The "all C<140" are done was meant for your ears, not his. The statement was meant in response to you saying there was low-lying fruit out there (meaning a C131) when in fact there was not. I would not give that as a status to Jean-Luc. It is far too cryptic. Even when we do give status updates in the other thread, he cannot update every base every time. Many times there are multiple sequences that are only moved 1-2 indexes. Hence why it looks like there's a C131 when there really is not. As an example, see the large group of sequences that were at C<140 in the other thread that were recently worked on but not terminated. Is he expected to update all of those bases each time we have an update? If so, he would be updating 10s of bases every time. What about Yoyo's bases? They constantly have updates that are not reflected for several weeks on the pages because they are working on ~15-20 bases at a time. It is only when he does the "all bases" update that we know that it catches everything up to date. We know that it takes a lot of time and effort to do that so we don't expect it often. Please make sure you understand the context of a comment before implying that someone is not supporting the project leader. |
Gentlemen,
Before continuing and making more "mathematical" comments, I would like to tell you a few things. I have been working with all of you now for quite some time and I think I know you well enough to know that none of you meant to offend the other. When I do an update, it is very rarely complete, because as Karsten says, it takes hours. The project page is not meant to be updated every day or even week, but you already know that. It is intended to provide data categorized, arranged in different ways so that anyone can cross-reference the data as he or she sees fit and try to spot "regularities" perhaps to formulate conjectures. Some may have other motivations: calculating, creating pages to make data sorting more efficient, examining the data... And it is our individual work that once put together makes sense. And each gives ideas based on the particular job they are doing and tries to find the best solution based on their particular needs. And naturally, as soon as a new function is added, on the other hand, the constraints will increase for some, but the result is that the tool will become more powerful. Karsten is right when he says that it is laborious. And each of us is doing hard work. It's hard to read all the posts and not forget anything when you update. I regularly make mistakes and you point them out to me. We're just trying to think of ways to reduce the number of errors, because there must be a lot of errors on the page that we haven't noticed, and one of the sources of error is that update requests from participants can sometimes get "buried" in the middle of a lot of other discussions, which makes them less visible, hence Karsten's idea of creating a thread that would only receive update requests according to the calculations made. But the informations that are given in general are very interesting and it would be a shame not to give them anymore. Feel free to speak up if you have any suggestions regarding Karsten's post #2252. [QUOTE=gd_barnes;624232]I see that base 177792 has question marks for exponents 30, 32, & 34. Those were terminated by me a couple of weeks ago per post 1108 in the other thread. [/QUOTE] OK, I've figured out what the problem is and it will be fixed in the next update. [QUOTE=EdH;624244]@kar_bon: If [C]terminate n^i[/C] is posted, does that automatically credit the poster? Are the user ids associated with credit ids? [/QUOTE] I'll let Karsten answer but as far as I know, there is nothing automatic, it's just to have only the information (and only these) to take into account for the update. [QUOTE=EdH;624244] If Gary terminates an unreserved matched parity sequence, which I track in the other thread, should he post it in this thread to update the sequence automatically? [/QUOTE] The other thread is more than enough for the matched parity sequences, because all the data needed for the updates are in the first post and that's fine. There is no need to put the data needed for the update in another place a second time. This would not change the current way we have been working with matched parity sequences. [QUOTE=EdH;624244] Instead of keeping track in the other threads, as I've been doing, should I let terminations be done in this thread now, by the "Terminator?"[/QUOTE] No, it is certainly easier for you, Edwin, to continue to operate as currently with your scripts and to put everything concerning the corresponding parities at the beginning of the thread provided for this purpose. It's not complicated for me to go see only two different places. [QUOTE=gd_barnes;624304] Please make sure you understand the context of a comment before implying that someone is not supporting the project leader.[/QUOTE] Honestly, I don't think I'm the "project leader" any more when I see all the work everyone is doing around this project. And to be frank, I think we all support each other, because we all need it, because we all work very hard often in addition to our professional and family activities. And only together can we build what we have built. Without computing enthusiasts and without their computing power, nothing can be done. Without the Karsten page, nothing can be done. Without FactroDB, nothing can be done, and so on. I don't know if we're going to find something revolutionary about aliquot suites with this project ? But we will have tried and in any case, our work will not be lost. And we will all have increased our skills in our chosen fields. |
Thank you, Jean-Luc,
I thought, perhaps, Karsten had developed a script to harvest the "update," etc., keywords from the thread. i will carry on as before. Thanks to all for all the work. |
@JL: To clear the issue with base 177792 just run the read-script in the base folder again.
To give an idea how my scripts are working to create the main page for this project including every base page and the statistics and datafiles here're some notes: 1. There're are two bases to obtain information from: a. FactorDB: For every open index of any base the last index from FactorDB will be downloaded as htm-file. This can be done for every base separately, not done all at once because of FDB restrictions. A download of a particular index is also possible but need a small change by hand. After the download is done another scripts will create the special page for that base and also the statistics for this base. This 2 scripts only runs in few seconds depending of download amount. For a new base only a new folder has to be created, copy 3 files into it and run one script from that but also in the main page to new base has to be included by hand. b. The forum-threads: only here data like who worked on any seq (terminated or cycle or only worked on). This is not done automated by scripts and has to be done by hand from JL only! As I mentioned I know that the work by YoYo is not mentioned anywhere here in the forum including terminations or merges and therefore could not updated directly. But terminations will always be found when updating a base. 2. The datafiles will be created also from those available data stored from FDB in the base folders and the scripts run over those ~9k open seqs in under a minute, create all 27 files. So those files directly depend on the accuracy and actuality of the downloaded data from FDB. Those data files I've created for everyone to get quick access of such lists without downloading and compiling their own data but give the information instantly. More special files could be included, too. That's why I suggested to create an own thread like the one for the aliseq main project: Those data are condensed in one thread without reading several posts to find the needed information. Every reservation, termination, merge as shown in the basepages has to be done by hand=JL. I hope it's now more clear how those scripts are working and that's what I meant to support Jean-Luc more than now. Keep the information small to get the needed ones for an update only. The more bases the more work has to be done although to most work is done by scripts. |
149^14 merges at index 1742 with 403620:3 but the merge info. is not shown on the page.
|
Karsten and Jean-Luc, thank you for the clarifications.
It does seem like a wholly separate thread for only work done would be helpful, like the main project has. If we did that, I would be inclined to state something like "update XX^XX, XX^XX", etc. whenever a sequence is moved only a few indexes. As it is now, I only post an update if there is a termination or if I finished an entire base to 120 digits or some other special thing like I've done a few opposite-parity sequences to cofactor C>=110 or something like that. The reason being is that if I posted everything that moved a few indexes, there would be a lot. But I would be OK doing that if we had a dedicated thread for updates. |
As for me, I am also in favor of the creation of this special thread, because I think it will make my life easier, and it will also allow me to be less forgetful.
But it should not become too painful for everyone : Nobody has to rigorously report every time a sequence has advanced a few iterations ! But if it's not too much of a hassle for you, you can report all advances, it's up to you. In any case, all the advances will be taken into account during the total update which takes place every 8 weeks or so. Note : I'm thinking of creating this thread after the next full update around the next weekend unless someone presents a major argued objection in the meantime. |
[QUOTE=garambois;624438]As for me, I am also in favor of the creation of this special thread, because I think it will make my life easier, and it will also allow me to be less forgetful.
But it should not become too painful for everyone : Nobody has to rigorously report every time a sequence has advanced a few iterations ! But if it's not too much of a hassle for you, you can report all advances, it's up to you. In any case, all the advances will be taken into account during the total update which takes place every 8 weeks or so. Note : I'm thinking of creating this thread after the next full update around the next weekend.[/QUOTE] I'm in favor of this. I'll be happy to report everything I advance a few indexes, terminations, an entire base to 120 digits, and anything that I do in a thread especially dedicated to that. It could also be used for reservations and drops of reservations. Make it like the main project status thread where the blue page picks up all of the various statuses automatically. It would probably make sense to sticky such a thread. |
[QUOTE=gd_barnes;624439]Make it like the main project status thread where the blue page picks up all of the various statuses automatically.[/QUOTE]
On that subject, would it be possible to retrofit the blue page software ([url]https://github.com/MersenneForum/MersenneForumAliquot[/url]) to support aliquot powers, importantly while having the sequence start values display nicely (in [i]exponent^power[/i] form)? That would allow us to display the last known term, drivers/guides, and any terminations much more rapidly, using a server cron job that updates a certain number of sequences per hour (I think it's configured to do 100 in the code, but that can be changed) with a priority system that prioritizes reserved and recently crunched sequences. |
[QUOTE=gd_barnes;624439]Make it like the main project status thread where the blue page picks up all of the various statuses automatically.
It would probably make sense to sticky such a thread.[/QUOTE] If you or anyone else who wants to come forward has a clear vision of how to go about starting this new thread, you can do it. I think Karsten's proposal in post #2252 is good and complete : [CODE]reserve 10^20 11^21 drop 10^21 11^22 terminate 10^22 update 12^12 new 999 [/CODE] So, we choose the same words as in the reservation thread of the main project. Note that we would have two additional words here in relation to the main project : "terminate" and "new". The word "new" would allow people to signal that they have initialized a new database. But I also think that as for the main project, people would spontaneously use variants for the words : "reserve" can be replaced by "taking" or by "take" or by "reserving". [QUOTE=Happy5214;624445]On that subject, would it be possible to retrofit the blue page software ([url]https://github.com/MersenneForum/MersenneForumAliquot[/url]) to support aliquot powers, importantly while having the sequence start values display nicely (in [i]exponent^power[/i] form)? That would allow us to display the last known term, drivers/guides, and any terminations much more rapidly, using a server cron job that updates a certain number of sequences per hour (I think it's configured to do 100 in the code, but that can be changed) with a priority system that prioritizes reserved and recently crunched sequences.[/QUOTE] I can't answer your question from post #2265 and I don't know if the people who could read the posts in this thread ? |
The new thread can be found here:
[URL="https://www.mersenneforum.org/showthread.php?t=28467"]Aliquot sequence n^i project update information[/URL] |
[QUOTE=garambois;624451]I can't answer your question from post #2265 and I don't know if the people who could read the posts in this thread ?[/QUOTE]
I asked in the blue page thread: [url]https://www.mersenneforum.org/showpost.php?p=624458&postcount=260[/url]. |
[QUOTE=EdH;624453]The new thread can be found here:
[URL="https://www.mersenneforum.org/showthread.php?t=28467"]Aliquot sequence n^i project update information[/URL][/QUOTE] OK, thank you Edwin ! This thread creation was indeed very simple. It will probably suffice without explanation, because the people who might post to it actually know the meaning of each line. When I make my next complete update in a few days, I will take into account both Threads: this one and the new one that has just been created. |
[QUOTE=Happy5214;624459]I asked in the blue page thread: [url]https://www.mersenneforum.org/showpost.php?p=624458&postcount=260[/url].[/QUOTE]
OK, thank you Happy. Let's wait and see what the answer will be. |
[QUOTE=garambois;624461]OK, thank you Edwin !
This thread creation was indeed very simple. It will probably suffice without explanation, because the people who might post to it actually know the meaning of each line. When I make my next complete update in a few days, I will take into account both Threads: this one and the new one that has just been created.[/QUOTE]I started it simple, but if desired, we can add extra info to the first post. Or, we can edit it in any way to help make things easier. |
I have completed testing of all bases > 150 all open sequences to >= 120 digits.
For this cycle, completed were bases in the 150 to 170 range. This included testing of bases 151, 157, 163, and 167. Work continues backwards by base for bases 149 to 130. There are 4 bases in this range with sequences at < 120 digits. The above would mostly complete my effort. Yoyo has completed nearly all bases < 130 to > 120 digits. Some are currently at < 120 digits but that's usually because there was a downdriver and they are still working on them. So...after I'm finished down through base 130, all that will be left are bases 77 and 88. Jean-Luc, I'm coming for your bases! (drum roll...) Don't worry, I wouldn't do anything without your permission. :-) I wasn't sure how to post this kind of status in the new thread so I posted it here. How should we state that an entire base should be updated? Would the following work for such an update?: [code] Update 151, 157, 163, 167 [/code] If we end up having something like the blue page for our sequences, some of the code would need to be a little different to account for entire base updates. |
Jean-Luc, do you know about what day you will be doing the big update of all of the bases?
|
[QUOTE=gd_barnes;624746]Jean-Luc, do you know about what day you will be doing the big update of all of the bases?[/QUOTE]
Yes : today (Sunday, February 12) ! This is a "test" update to see if everything is good now. I have a few days off and I have time ! But in about ten days, I will do a second complete update again. [QUOTE=gd_barnes;624468] I wasn't sure how to post this kind of status in the new thread so I posted it here. How should we state that an entire base should be updated? Would the following work for such an update?: [code] Update 151, 157, 163, 167 [/code] [/QUOTE] This is good for me. |
[QUOTE=EdH;624453]The new thread can be found here:
[URL="https://www.mersenneforum.org/showthread.php?t=28467"]Aliquot sequence n^i project update information[/URL][/QUOTE] Maybe we should add in the first post that there is a possibility to point out possible mergers that I would have forgotten, like Karsten does in the second post with the word "merge" ? What do you think about it ? |
[QUOTE=garambois;624759]Yes : today (Sunday, February 12) !
This is a "test" update to see if everything is good now. I have a few days off and I have time ! But in about ten days, I will do a second complete update again. This is good for me.[/QUOTE] OK, I posted the bases in the other thread that need to be updated that I'm done with to >= 120 digits. Only two bases to go! I was trying to gauge if I could complete them before your big update. Since you are doing another in 10 days, that's plenty of time for me. |
[QUOTE=garambois;624760]Maybe we should add in the first post that there is a possibility to point out possible mergers that I would have forgotten, like Karsten does in the second post with the word "merge" ?
What do you think about it ?[/QUOTE]I added a merge suggestion. I made it more brief than Karsten's, but I'm sure all such posts should help you. |
Complete update done.
See the new thread dedicated to the N^I project update. |
[QUOTE=EdH;624770]I added a merge suggestion. I made it more brief than Karsten's, but I'm sure all such posts should help you.[/QUOTE]
Thank you Edwin ! |
Page updated again.
[B]Added base : 999 (This is an experimental base that allowed to test new files for creating new bases). When you open datafiles named "easy_*.txt", you now have a sequence counter in the first line next to the date. The fourth column of the "easy_cycles*.txt" files has been modified : it now indicates the number of digits of the cycle number that ends the sequence. Other non-visible changes have been made to make my work even more convenient and to reduce the risk of errors.[/B] A very big thank you to Karsten again for all this work ! :smile: The 999 base is now available to anyone who wants to calculate their sequences. |
[QUOTE=garambois;624890]<snip>
The 999 base is now available to anyone who wants to calculate their sequences.[/QUOTE] I haven't done a new base in a while. I'll take base 999. |
Thank you very much Gary !
|
Edwin,
I do not know why, but the program "Alimerge" does not find me this merger : 149^14:i1742 merges with 403620:i3 Can you reproduce this error on your computer ? |
[QUOTE=garambois;624923]Edwin,
I do not know why, but the program "Alimerge" does not find me this merger : 149^14:i1742 merges with 403620:i3 Can you reproduce this error on your computer ?[/QUOTE]The bash script and program I use finds the merge. Are you using a version from me or the one from Allias?:[code] $ bash alimerge.sh Running base 149: 149^14:i1742 merges with 403620:i3 Done in 203 seconds.[/code] |
Allias' version (from GitHub) finds it.
[code]$ ./alimerge 149 14 14 Running base 149 from 14 through 14 . . . 149^14:i1742 merges with 403620:i3 Total running time : 31 Seconds : 859 Milliseconds (31 seconds.) Downloading file time: 31 Seconds : 842 Milliseconds Computation only time: 17 Milliseconds[/code] |
How, How, How !
Now it works. I do believe I have a problem with my computer crashing a lot lately ! Ouch, ouch, ouch ! Thank you for your answers. |
Page updated again.
[B]Reservations for Gary noted. Some errors corrected. Forgotten mergers added. Some bases updated.[/B] |
I have completed new base 999 all open sequences to >= 120 digits.
The base is released. More details in the new thread. |
I have now completed testing of all unreserved bases on the project for all open sequences to >= 120 digits.
For this cycle, completed were bases in the 130 to 150 range. This included testing of bases 131, 137, 139, and 149. Yoyo and others have already done this for sequences on bases < 130 or they are currently being worked on. Mission accomplished. :-) Jean-Luc, the only remaining sequences at < 120 digits besides a few stragglers with down-drivers that are being worked on by Yoyo are on your bases 77 and 88. Just a little nudge...:smile: |
Unfortunately, I can't work on bases 77 and 88 at the moment.
I have big problems with my personal computer (not the 64C/128T one that I have at my disposal for a few days and that is not mine). This computer is 12 years old ! In the next few days, I'm going to have to take it apart, and dust off all the components and I'm going to reinstall all the software hoping it will last me another 6-8 months. Because I'm saving the money to own my own 64C/128T in a few months. So the 77 and 88 bases are free the next week if you want to do some work on them. I should also finish all the data analysis from last summer's big harvest. It was long and complicated. I will keep you posted in a few days about this. |
[QUOTE=garambois;625164]Unfortunately, I can't work on bases 77 and 88 at the moment.
I have big problems with my personal computer (not the 64C/128T one that I have at my disposal for a few days and that is not mine). This computer is 12 years old ! In the next few days, I'm going to have to take it apart, and dust off all the components and I'm going to reinstall all the software hoping it will last me another 6-8 months. Because I'm saving the money to own my own 64C/128T in a few months. So the 77 and 88 bases are free the next week if you want to do some work on them. I should also finish all the data analysis from last summer's big harvest. It was long and complicated. I will keep you posted in a few days about this.[/QUOTE] Wow! I will be curious to hear about your quest to purchase a 64C/128T machine. I'd like to get myself one at some point so I'd like to hear about your experience with researching and purchasing one. OK, I am happy to finish off the 5 sequences each on bases 77 and 88 to >= 125 digits. That looks like how far you have been testing the bases so I will do that. I'll start on them late Saturday or sometime Sunday. Let me know if that's OK. I won't make it a formal reservation since you still have the bases reserved. |
[QUOTE=gd_barnes;625165]
I'll start on them late Saturday or sometime Sunday. Let me know if that's OK. I won't make it a formal reservation since you still have the bases reserved.[/QUOTE] Perfect for me ! |
[QUOTE=gd_barnes;625165]Wow! I will be curious to hear about your quest to purchase a 64C/128T machine. I'd like to get myself one at some point so I'd like to hear about your experience with researching and purchasing one.
[/QUOTE] There is a company in our area that builds such machines, very well ventilated. The one I'm lent has two big GPUs that consume 300W each and throw the hot air into the tower. The owner lends me 124 or 120 threads to run CADO-NFS for a few days, but he requires the GPUs to do something else in parallel. But to run the GPUs, you need some CPU. So when I run CADO-NFS, I have to run the "export OMP_DYNAMIC=true" instruction beforehand. If I don't do this, the calculation gets stuck at the linear algebra step "lingen". But according to the tests I made last summer with the same computer, it seems that the instruction "export OMP_DYNAMIC=true" slows down the calculations by a few percent, which is a pity. But this computer is a real "bomb" as we say in French : 128T and 256 GB of ram. Only weak point : the communication speed between the CPU and the ram. But this problem should be solved with the arrival of DDR5. With this future new computer, I have the ambition to relaunch the search for the "gaussian aliquot cycles" that we started with Andrew Walker, see here : [URL="https://www.mersenneforum.org/showthread.php?t=21068"]https://www.mersenneforum.org/showthread.php?t=21068[/URL] When I had started this work, I had talked to Paul Zimmermann about it and he had written a very powerful program in C language. I intend to run this program on 64 Threads for 1 or 2 years ! I will then contact Chernykh to see if my future discoveries could interest him. He seems to be interested only in Gaussian friendly pairs which remain in the set of integers and not in those which are in the set of complex numbers and which have a non zero imaginary part, [URL="https://sech.me/ap/"]see here[/URL]. But this could change, at least I hope so. Otherwise, I will be obliged to create a special page on my site which would list these discoveries, especially since Paul Zimmermann's program has discovered a gaussian cycle of length 6, [URL="https://www.mersenneforum.org/showpost.php?p=528582&postcount=28"]see here[/URL]. All this is part of a more general framework of work that I explain [URL="http://www.aliquotes.com/etendre_sigma.html"]here[/URL], where the aim is to extend the sigma function to something other than integers. I should also mention that I had also tried in some works to modify the iterative process always in order to try to generalize the problem of the aliquot sequences and to find a more general rule which would inform us about the behavior of the aliquot sequences : [URL="http://www.aliquotes.com/autres_processus_iteratifs.html"]see here[/URL]. These works were very fun and stimulating for the mind, but there was no general law of discovery ! In short, I have dozens of programs to run to continue all my work and my future 128-threaded computer will be saturated as soon as I get it ! :smile: |
Thank you, Jean-Luc, for all of the info about your process to buy a new bigger machine in the future. I look forward to hearing more about your searches with such a beast of a machine!
Meanwhile: I have completed bases 77 and 88 opposite-parities to >= 125 digits. With that, all sequences* on the project have now reached >= 120 digits. I am officially done with the effort. :smile: * There are still a few stragglers to go that are being worked on by Yoyo. |
For your information, I will do a full update again before Sunday February 26th.
Thanks Gary for the 77 and 78 bases. It looks like the dusting I did on my old machine worked. I also recompiled all the programs : everything seems to work normally. The machine does not crash anymore ! |
[B][SIZE="4"]Results of the 2022 Summer Harvest Data Analysis[/SIZE][/B]
I just finished analyzing the data from last summer's big harvest. This analysis was very long and tedious. I turned the data over and over. There is no need for me to re-explain everything here. To know the details of the analysis, you can read [URL="https://www.mersenneforum.org/showpost.php?p=612644&postcount=1845"]post #1845[/URL]. For those readers who are new to us and don't know what this "n^i project" is all about, here is a link to [URL="https://www.mersenneforum.org/showpost.php?p=586049&postcount=1328"]post #1328[/URL] which reminds us why this project was born. The analysis of the summer 2021 data had been much faster. The amount of data was much less, but more importantly, I looked for lots of new things for the 2022 data analysis compared to the previous analysis. There were many "alerts" that led me to believe that I had stumbled upon something interesting. Let me show you some examples of alerts : 1) The sequences 648^30 and 648^44 end with the prime number 53. However, since the summer of 2022, Gary has managed to finish the calculation of the sequence 648^58 which also ends with the prime number 53 ! We notice that starting from 30, we always add 14 to the exponent. The next exponent is 58 + 14 = 72. Unfortunately, 648^72 is not a terminated sequence, I remain stuck at index 3 on a C171 cofactor. Almost all such "alerts" remain unverifiable for the same reasons: the terms in the next sequence are far too large to verify. I can also go backwards and look at 648^16, because 30 - 14 = 16. And 648^16 does not end in 53. So I think we are facing a false alert. 2) Another type of example of alert with amicable numbers (1184, 1210) : 1184^3 and 1210^12 end with the prime number 73. 1184^19 and 1210^3 end with the prime number 191. No other powers of 1184 are present in the project data of all sequences that end in a prime <300. So whenever 1184 is present, its amiable number is present for a given prime. The converse is not true. So I did all sorts of tests with the other pairs of amicable numbers. Result : nothing ! And I spent weeks checking for such "alerts". We are missing data with higher exponents for each base and no one will be able to bring us that data. The 648^72 example is the "simplest" example and at the limit verifiable within 1 or 2 years, but the other examples not presented here are really out of reach. Certainly if 648^72 ended in 53, there would be something ! But I bet that won't be the case ! I therefore consider that my analyses of this type have not given anything for this year : there is nothing serious ! BUT... [SIZE="3"][B][U]A very surprising, counter-intuitive and inexplicable finding[/U][/B][/SIZE] As usual, the surprise came not from what I was looking for but from something else entirely. Please look at the data below : [CODE]SAME PARITY OPPOSITE PARITY 0 --- 1 [2, [2, 1]] 90 [3, [2, 2], [2, 4], [2, 164], [3, 1], [3, 5], [3, 247], [6, 152], [7, 77], [10, 124], [11, 15], [12, 2], [13, 15], [14, 76], [14, 80], [15, 1], [19, 15], [20, 2], [20, 8], [21, 21], [21, 55], [22, 80], [23, 3], [23, 109], [29, 69], [30, 82], [30, 92], [31, 79], [33, 1], [35, 49], [37, 11], [38, 30], [40, 14], [43, 15], [44, 58], [45, 1], [47, 13], [52, 34], [53, 27], [54, 10], [55, 53], [58, 18], [59, 49], [60, 86], [70, 70], [74, 6], [74, 20], [74, 78], [75, 45], [77, 5], [77, 19], [79, 97], [83, 43], [84, 30], [87, 1], [87, 17], [93, 49], [95, 5], [96, 42], [99, 3], [99, 71], [104, 58], [105, 1], [107, 11], [109, 13], [131, 57], [137, 3], [139, 33], [157, 5], [167, 63], [167, 79], [197, 23], [197, 27], [220, 16], [239, 3], [251, 33], [277, 9], [293, 11], [293, 23], [338, 2], [564, 2], [648, 56], [828, 24], [966, 6], [966, 34], [968, 8], [996, 12], [1250, 10], [2310, 36], [14288, 12], [223092870, 20]] 51 [3, [2, 55], [2, 305], [2, 317], [3, 2], [5, 38], [7, 4], [11, 2], [12, 1], [26, 1], [29, 2], [30, 1], [38, 11], [41, 2], [42, 1], [45, 34], [46, 1], [50, 73], [52, 1], [53, 2], [53, 30], [54, 1], [66, 1], [71, 2], [72, 1], [78, 1], [86, 1], [89, 2], [90, 1], [98, 45], [101, 2], [102, 1], [113, 2], [173, 2], [197, 2], [211, 2], [242, 41], [257, 2], [269, 2], [288, 1], [288, 3], [306, 17], [338, 13], [396, 11], [450, 1], [450, 19], [648, 1], [722, 37], [800, 21], [882, 5], [1058, 31], [1352, 1]] 0 --- 1 [5, [5, 1]] 51 [7, [2, 10], [2, 12], [2, 278], [3, 305], [7, 1], [7, 127], [11, 5], [13, 87], [14, 130], [17, 91], [18, 52], [18, 70], [18, 134], [21, 17], [23, 11], [26, 50], [26, 80], [30, 52], [30, 70], [35, 11], [40, 2], [41, 49], [42, 2], [45, 55], [56, 38], [68, 10], [70, 48], [75, 1], [75, 63], [79, 7], [80, 28], [92, 32], [93, 29], [94, 22], [97, 5], [99, 19], [102, 2], [193, 3], [200, 24], [200, 56], [211, 65], [271, 3], [277, 25], [293, 55], [385, 57], [450, 12], [648, 40], [968, 60], [8128, 38], [15015, 35], [131071, 25]] 78 [7, [2, 3], [2, 141], [2, 387], [2, 421], [3, 6], [3, 8], [3, 118], [3, 198], [7, 2], [7, 8], [10, 1], [12, 21], [13, 2], [14, 1], [14, 19], [14, 21], [15, 10], [17, 24], [18, 13], [19, 2], [20, 1], [21, 8], [22, 1], [22, 19], [26, 3], [26, 19], [28, 9], [28, 47], [34, 1], [34, 7], [34, 9], [37, 2], [38, 1], [38, 13], [43, 20], [45, 4], [46, 33], [47, 8], [47, 42], [48, 61], [50, 17], [53, 8], [58, 5], [58, 27], [61, 2], [62, 1], [66, 5], [66, 31], [72, 25], [74, 11], [76, 39], [77, 2], [77, 16], [78, 23], [84, 31], [96, 3], [96, 7], [105, 2], [151, 2], [163, 4], [220, 15], [223, 4], [227, 10], [229, 4], [242, 39], [271, 34], [338, 39], [450, 17], [496, 31], [564, 21], [696, 7], [722, 51], [800, 51], [888, 5], [14264, 7], [15472, 3], [19116, 15], [131071, 8]] 66 [11, [2, 60], [2, 316], [2, 480], [3, 15], [3, 189], [3, 303], [5, 15], [7, 143], [10, 20], [11, 1], [11, 137], [13, 31], [14, 14], [14, 28], [18, 76], [21, 1], [21, 47], [24, 2], [26, 36], [35, 83], [47, 47], [51, 1], [53, 15], [53, 67], [58, 2], [58, 24], [59, 81], [63, 85], [65, 33], [67, 3], [70, 2], [71, 61], [72, 4], [72, 58], [73, 3], [76, 28], [77, 15], [79, 3], [82, 2], [82, 54], [88, 54], [91, 1], [92, 30], [95, 63], [98, 2], [99, 13], [109, 15], [109, 75], [139, 5], [162, 70], [193, 57], [210, 58], [271, 23], [283, 27], [288, 4], [564, 30], [578, 50], [648, 42], [800, 54], [882, 50], [888, 26], [968, 32], [14288, 2], [14288, 4], [31704, 16], [223092870, 4]] 12 [11, [2, 499], [17, 2], [18, 1], [18, 55], [37, 50], [65, 2], [72, 63], [80, 3], [242, 25], [284, 19], [648, 41], [14288, 9]] 47 [13, [2, 358], [3, 3], [3, 31], [3, 67], [5, 9], [11, 3], [11, 27], [13, 1], [20, 72], [23, 49], [23, 77], [24, 70], [35, 1], [38, 86], [39, 91], [43, 3], [43, 17], [50, 62], [56, 8], [62, 2], [69, 1], [70, 34], [73, 11], [75, 13], [79, 23], [93, 1], [94, 12], [96, 6], [103, 5], [104, 24], [104, 64], [131, 9], [167, 17], [173, 69], [179, 25], [181, 23], [197, 3], [229, 21], [263, 7], [263, 15], [392, 48], [392, 54], [770, 28], [8191, 17], [15015, 9], [15015, 33], [6469693230, 2]] 5 [13, [50, 99], [271, 4], [648, 7], [968, 1], [1058, 7]] 27 [17, [3, 63], [6, 2], [7, 55], [13, 21], [15, 79], [17, 1], [19, 97], [24, 4], [38, 42], [39, 1], [45, 5], [46, 10], [51, 31], [55, 1], [58, 70], [60, 28], [78, 52], [84, 78], [85, 63], [92, 4], [96, 34], [99, 23], [119, 79], [131, 15], [157, 65], [271, 21], [396, 6]] 4 [17, [23, 2], [24, 1], [162, 39], [1152, 31]] 59 [19, [2, 76], [2, 190], [3, 275], [3, 327], [5, 233], [10, 2], [10, 12], [10, 44], [12, 4], [12, 150], [13, 3], [13, 141], [15, 21], [19, 1], [22, 32], [34, 96], [40, 6], [41, 7], [42, 10], [42, 22], [45, 15], [55, 29], [59, 85], [60, 30], [65, 1], [70, 28], [71, 21], [72, 74], [76, 66], [77, 1], [79, 13], [88, 62], [89, 9], [89, 19], [90, 2], [94, 70], [97, 29], [98, 6], [102, 54], [109, 27], [210, 12], [233, 41], [241, 51], [263, 21], [269, 11], [276, 10], [277, 11], [288, 16], [293, 17], [392, 6], [392, 26], [392, 42], [496, 30], [722, 30], [770, 14], [1058, 44], [8191, 15], [19116, 24], [131071, 21]] 7 [19, [2, 39], [2, 219], [2, 505], [41, 68], [193, 2], [338, 59], [1152, 27]] 44 [23, [3, 181], [7, 3], [7, 9], [11, 127], [17, 79], [18, 6], [18, 64], [21, 61], [23, 1], [40, 36], [44, 68], [47, 3], [54, 80], [55, 15], [57, 1], [57, 17], [65, 3], [69, 9], [71, 65], [74, 64], [77, 23], [77, 37], [79, 49], [85, 1], [86, 54], [88, 26], [93, 3], [99, 1], [101, 65], [105, 47], [127, 3], [137, 43], [173, 35], [200, 18], [242, 28], [263, 13], [293, 7], [496, 22], [564, 50], [800, 2], [8128, 8], [14536, 32], [19116, 8], [510510, 8]] 3 [23, [3, 12], [18, 17], [578, 1]] 21 [29, [5, 41], [14, 92], [18, 50], [18, 116], [22, 8], [26, 68], [29, 1], [46, 36], [47, 39], [48, 4], [61, 5], [63, 67], [83, 11], [88, 2], [88, 78], [91, 13], [97, 3], [229, 51], [283, 5], [306, 56], [82589933, 3]] 4 [29, [72, 43], [162, 11], [338, 57], [450, 21]] 39 [31, [2, 146], [3, 169], [5, 3], [5, 161], [11, 93], [11, 107], [12, 14], [19, 35], [19, 37], [31, 1], [31, 11], [40, 30], [44, 30], [56, 52], [61, 55], [70, 54], [70, 58], [80, 30], [87, 13], [91, 15], [92, 2], [92, 14], [95, 27], [97, 49], [103, 9], [113, 3], [119, 53], [179, 21], [210, 14], [220, 4], [227, 19], [239, 33], [269, 37], [281, 61], [882, 6], [1152, 4], [1152, 28], [1210, 10], [8191, 7]] 17 [31, [2, 5], [2, 101], [6, 17], [11, 38], [31, 2], [38, 19], [54, 67], [58, 1], [67, 2], [68, 1], [74, 21], [98, 3], [162, 27], [163, 8], [231, 2], [882, 1], [12496, 3]] 30 [37, [2, 68], [7, 79], [10, 74], [12, 98], [14, 2], [14, 138], [22, 74], [24, 6], [37, 1], [42, 16], [45, 11], [53, 87], [55, 43], [60, 94], [63, 3], [78, 4], [84, 32], [85, 81], [94, 16], [95, 37], [96, 48], [99, 5], [104, 4], [131, 45], [149, 43], [241, 47], [283, 33], [552, 18], [14316, 2], [15015, 27]] 43 [37, [2, 125], [2, 243], [3, 90], [5, 4], [7, 6], [10, 3], [10, 5], [11, 40], [14, 35], [22, 7], [22, 21], [22, 23], [23, 10], [26, 31], [28, 3], [28, 11], [29, 50], [35, 2], [44, 3], [44, 19], [47, 16], [51, 2], [67, 74], [71, 4], [74, 5], [75, 8], [77, 34], [79, 6], [83, 2], [83, 10], [84, 1], [88, 3], [90, 9], [96, 1], [97, 4], [131, 6], [139, 2], [220, 27], [239, 18], [241, 28], [251, 26], [277, 10], [293, 10]] 52 [41, [2, 6], [2, 8], [2, 112], [7, 11], [11, 57], [12, 32], [12, 56], [17, 65], [24, 18], [37, 73], [38, 54], [40, 20], [40, 52], [41, 1], [43, 25], [43, 81], [50, 16], [55, 21], [59, 39], [61, 9], [62, 40], [63, 1], [66, 78], [70, 38], [71, 5], [71, 7], [73, 27], [78, 64], [79, 17], [84, 2], [88, 22], [94, 46], [96, 50], [97, 7], [98, 10], [103, 21], [107, 3], [107, 61], [119, 29], [167, 13], [173, 9], [181, 5], [210, 4], [283, 17], [385, 11], [966, 2], [968, 2], [1210, 24], [1352, 4], [2310, 8], [9699690, 20], [223092870, 2]] 90 [41, [2, 23], [2, 47], [2, 117], [2, 281], [2, 373], [2, 405], [2, 411], [6, 5], [6, 13], [11, 14], [12, 119], [15, 2], [15, 6], [18, 39], [20, 3], [22, 67], [33, 2], [35, 8], [35, 34], [37, 20], [40, 11], [40, 21], [42, 5], [44, 49], [45, 8], [45, 22], [45, 36], [47, 2], [48, 1], [48, 3], [51, 18], [52, 25], [56, 1], [57, 2], [62, 3], [62, 13], [65, 4], [65, 26], [66, 25], [69, 2], [70, 13], [70, 23], [75, 2], [76, 1], [78, 15], [79, 2], [79, 4], [80, 1], [82, 11], [85, 8], [88, 1], [92, 1], [94, 11], [97, 48], [98, 41], [103, 2], [104, 1], [105, 10], [105, 30], [109, 2], [113, 16], [127, 26], [127, 44], [127, 52], [163, 2], [167, 20], [193, 32], [197, 10], [197, 20], [220, 17], [223, 2], [227, 2], [229, 2], [231, 4], [257, 28], [263, 8], [277, 14], [284, 11], [288, 57], [293, 4], [385, 42], [439, 2], [578, 3], [720, 3], [770, 13], [966, 31], [14288, 3], [14536, 5], [31704, 9], [8589869056, 3]] 97 [43, [2, 62], [2, 210], [2, 510], [6, 26], [6, 32], [7, 51], [7, 189], [10, 42], [11, 13], [13, 95], [17, 101], [17, 129], [18, 34], [19, 55], [20, 70], [22, 2], [22, 104], [24, 10], [24, 36], [26, 6], [26, 38], [26, 78], [26, 88], [28, 18], [29, 3], [29, 59], [33, 3], [33, 23], [33, 45], [34, 28], [34, 44], [37, 97], [38, 38], [43, 1], [43, 55], [47, 17], [51, 5], [51, 13], [60, 6], [62, 8], [62, 50], [66, 56], [67, 63], [68, 14], [68, 18], [69, 3], [69, 43], [70, 8], [70, 22], [75, 11], [75, 73], [79, 31], [80, 36], [80, 68], [82, 72], [83, 3], [88, 76], [93, 27], [95, 21], [102, 22], [102, 26], [104, 26], [104, 68], [104, 70], [105, 3], [107, 5], [109, 11], [127, 11], [137, 71], [157, 25], [167, 39], [179, 3], [179, 5], [191, 5], [191, 31], [199, 29], [227, 57], [231, 1], [241, 55], [257, 3], [263, 43], [269, 43], [276, 4], [277, 45], [439, 3], [496, 18], [496, 36], [552, 28], [660, 30], [828, 14], [1152, 12], [1155, 1], [1210, 28], [2310, 16], [19116, 4], [2147483647, 5], [200560490130, 10]] 121 [43, [2, 9], [2, 271], [3, 4], [3, 22], [3, 80], [3, 86], [6, 77], [7, 16], [7, 28], [11, 28], [11, 56], [13, 4], [13, 8], [13, 40], [13, 98], [14, 5], [14, 23], [14, 25], [15, 14], [17, 12], [17, 16], [19, 4], [20, 21], [21, 2], [21, 6], [21, 98], [23, 4], [24, 19], [24, 31], [28, 13], [31, 14], [31, 26], [34, 3], [35, 4], [39, 6], [40, 1], [40, 35], [42, 13], [43, 2], [44, 1], [45, 2], [46, 3], [46, 5], [48, 9], [50, 1], [51, 50], [52, 9], [53, 4], [54, 49], [57, 26], [58, 71], [59, 2], [59, 6], [59, 18], [60, 1], [62, 9], [63, 6], [65, 12], [65, 32], [66, 21], [67, 10], [70, 1], [70, 27], [71, 26], [73, 2], [73, 14], [74, 1], [75, 6], [75, 46], [78, 17], [79, 30], [79, 56], [80, 5], [82, 1], [86, 11], [87, 6], [87, 28], [90, 35], [91, 2], [91, 44], [92, 13], [92, 19], [93, 24], [94, 1], [96, 5], [97, 40], [98, 5], [98, 29], [99, 8], [104, 3], [105, 6], [107, 2], [119, 38], [131, 2], [139, 4], [157, 2], [181, 2], [191, 2], [200, 7], [223, 12], [227, 6], [227, 12], [227, 14], [231, 6], [233, 10], [271, 2], [271, 20], [277, 2], [306, 11], [385, 8], [396, 5], [648, 45], [722, 15], [770, 1], [968, 29], [1152, 9], [1152, 29], [1155, 2], [1250, 47], [15015, 12], [6469693230, 1]] 14 [47, [3, 9], [28, 34], [30, 10], [40, 22], [47, 1], [65, 47], [77, 13], [78, 80], [82, 26], [93, 17], [120, 44], [578, 22], [696, 2], [968, 40]] 3 [47, [2, 453], [98, 75], [200, 37]] 19 [53, [2, 20], [2, 78], [2, 214], [2, 450], [23, 101], [48, 50], [48, 70], [53, 1], [72, 18], [75, 31], [94, 2], [163, 61], [181, 53], [241, 3], [306, 8], [306, 50], [648, 30], [648, 44], [888, 14]] 1 [53, [2, 347]] 9 [59, [6, 128], [19, 25], [59, 1], [78, 44], [105, 19], [127, 25], [151, 27], [162, 62], [269, 47]] 108 [59, [3, 94], [7, 14], [7, 24], [7, 64], [10, 7], [10, 23], [11, 8], [11, 58], [11, 80], [13, 6], [13, 38], [14, 9], [14, 49], [15, 36], [17, 64], [19, 6], [20, 59], [20, 71], [24, 9], [24, 13], [26, 59], [26, 71], [28, 7], [28, 23], [29, 4], [29, 36], [31, 8], [31, 10], [33, 4], [33, 6], [35, 6], [38, 77], [39, 2], [39, 12], [42, 3], [44, 5], [45, 6], [45, 28], [45, 30], [51, 32], [51, 78], [52, 3], [52, 11], [55, 4], [56, 61], [57, 4], [58, 17], [60, 3], [60, 17], [61, 12], [65, 50], [66, 17], [69, 4], [69, 10], [69, 24], [70, 19], [70, 39], [71, 28], [73, 4], [73, 10], [74, 33], [75, 18], [75, 38], [77, 28], [84, 5], [86, 37], [86, 49], [87, 4], [87, 8], [89, 16], [90, 11], [95, 2], [97, 46], [99, 2], [103, 14], [103, 38], [104, 5], [119, 2], [131, 22], [137, 2], [149, 2], [151, 6], [167, 2], [173, 6], [179, 4], [179, 16], [191, 20], [193, 4], [193, 16], [197, 4], [199, 2], [199, 4], [200, 1], [211, 4], [220, 55], [233, 2], [251, 2], [263, 2], [271, 6], [281, 12], [385, 10], [439, 4], [564, 5], [720, 11], [14288, 7], [15472, 5], [131071, 4], [7420738134810, 1]] 15 [61, [30, 90], [39, 3], [43, 23], [45, 7], [52, 6], [52, 36], [61, 1], [67, 61], [75, 43], [76, 68], [104, 54], [137, 21], [284, 22], [496, 50], [1058, 12]] 1 [61, [60, 9]] 8 [67, [14, 6], [19, 53], [45, 47], [57, 13], [67, 1], [99, 55], [968, 4], [968, 58]] 1 [67, [2, 147]] 25 [71, [6, 14], [12, 58], [13, 53], [13, 113], [21, 11], [22, 36], [26, 66], [50, 2], [61, 11], [63, 33], [66, 70], [71, 1], [92, 60], [95, 25], [137, 33], [179, 23], [239, 45], [271, 7], [271, 53], [439, 31], [496, 8], [722, 12], [780, 30], [1210, 26], [510510, 4]] 5 [71, [2, 43], [2, 177], [18, 77], [162, 1], [722, 27]] 17 [73, [7, 133], [20, 4], [22, 94], [31, 3], [34, 2], [57, 47], [57, 81], [60, 62], [65, 57], [73, 1], [74, 74], [127, 19], [200, 2], [229, 3], [1210, 12], [131071, 3], [33550336, 18]] 30 [73, [2, 67], [3, 98], [5, 10], [6, 3], [10, 11], [10, 13], [12, 3], [17, 48], [23, 6], [28, 5], [34, 5], [43, 14], [46, 7], [63, 14], [73, 12], [76, 9], [78, 3], [86, 7], [97, 2], [98, 1], [105, 12], [137, 4], [173, 14], [263, 10], [496, 43], [968, 11], [1184, 3], [2310, 25], [19116, 5], [30030, 11]] 11 [79, [3, 113], [6, 4], [17, 39], [34, 56], [50, 6], [79, 1], [96, 26], [151, 67], [181, 45], [211, 53], [1250, 2]] 1 [79, [82, 5]] 10 [83, [2, 376], [5, 5], [5, 57], [56, 44], [60, 88], [78, 12], [83, 1], [84, 46], [95, 57], [120, 28]] 1 [83, [5, 36]] 13 [89, [2, 32], [2, 82], [3, 111], [7, 101], [37, 23], [47, 31], [57, 5], [68, 12], [89, 1], [113, 73], [450, 28], [660, 44], [510510, 2]] --- 11 [97, [7, 43], [38, 34], [43, 57], [68, 2], [97, 1], [98, 80], [103, 43], [107, 67], [131, 17], [293, 57], [392, 18]] 4 [97, [2, 51], [2, 73], [19, 38], [1250, 9]] 9 [101, [31, 29], [45, 93], [55, 45], [73, 43], [73, 79], [99, 69], [101, 1], [1210, 2], [7420738134810, 8]] 4 [101, [2, 301], [2, 425], [29, 12], [181, 16]] 8 [103, [2, 50], [6, 16], [24, 90], [44, 74], [50, 12], [72, 62], [103, 1], [396, 10]] 1 [103, [242, 47]] 20 [107, [5, 223], [7, 165], [7, 171], [14, 94], [41, 29], [76, 78], [85, 31], [101, 31], [105, 61], [107, 1], [119, 3], [220, 2], [227, 41], [229, 17], [229, 39], [231, 57], [251, 3], [293, 9], [496, 60], [82589933, 13]] 1 [107, [2, 255]] 14 [109, [3, 249], [10, 80], [18, 10], [28, 98], [30, 46], [55, 49], [79, 33], [83, 7], [109, 1], [131, 21], [276, 6], [338, 4], [1352, 16], [14288, 6]] 3 [109, [85, 4], [578, 11], [660, 5]] 7 [113, [5, 31], [55, 3], [58, 50], [102, 12], [113, 1], [239, 31], [578, 28]] 9 [113, [2, 11], [3, 38], [19, 50], [20, 5], [21, 70], [34, 15], [38, 3], [72, 21], [73, 22]] 8 [127, [5, 53], [53, 39], [97, 73], [101, 75], [127, 1], [139, 29], [211, 9], [1250, 6]] 2 [127, [2, 7], [127, 2]] 8 [131, [3, 141], [19, 3], [30, 2], [33, 9], [66, 2], [66, 88], [119, 49], [131, 1]] 1 [131, [162, 37]] 8 [137, [2, 396], [42, 42], [54, 28], [82, 16], [97, 57], [137, 1], [191, 47], [968, 54]] 0 --- 4 [139, [61, 27], [77, 49], [139, 1], [284, 4]] 0 --- 3 [149, [3, 55], [59, 45], [149, 1]] 0 --- 6 [151, [96, 70], [103, 63], [151, 1], [162, 16], [210, 40], [15472, 38]] 1 [151, [18, 79]] 5 [157, [20, 18], [20, 34], [60, 12], [65, 83], [157, 1]] 2 [157, [241, 2], [242, 1]] 8 [163, [3, 47], [35, 9], [163, 1], [163, 51], [220, 26], [284, 32], [780, 4], [800, 8]] 3 [163, [281, 2], [293, 2]] 3 [167, [45, 35], [109, 3], [167, 1]] 0 --- 4 [173, [7, 31], [57, 59], [173, 1], [277, 3]] 0 --- 11 [179, [14, 40], [40, 72], [60, 36], [61, 31], [87, 25], [109, 29], [179, 1], [200, 52], [227, 13], [241, 5], [14316, 8]] 0 --- 3 [181, [12, 6], [149, 63], [181, 1]] 0 --- 12 [191, [7, 63], [18, 112], [40, 84], [54, 22], [56, 2], [78, 14], [89, 15], [191, 1], [306, 34], [385, 1], [439, 13], [1152, 2]] 10 [191, [12, 5], [47, 48], [68, 25], [94, 3], [220, 3], [284, 5], [450, 3], [578, 5], [1184, 19], [1210, 3]] 9 [193, [23, 39], [26, 2], [51, 67], [63, 5], [98, 4], [193, 1], [242, 10], [1155, 19], [15472, 40]] 3 [193, [37, 10], [62, 55], [882, 31]] 2 [197, [15, 85], [197, 1]] 0 --- 4 [199, [2, 110], [23, 71], [199, 1], [450, 36]] 0 --- 8 [211, [2, 46], [3, 211], [33, 21], [43, 21], [48, 58], [65, 63], [76, 14], [211, 1]] 1 [211, [338, 1]] 4 [223, [69, 19], [105, 23], [162, 68], [223, 1]] 0 --- 2 [227, [58, 76], [227, 1]] 0 --- 3 [229, [23, 33], [229, 1], [241, 49]] 0 --- 4 [233, [14, 4], [120, 72], [229, 63], [233, 1]] 0 --- 6 [239, [5, 93], [6, 204], [19, 7], [20, 14], [181, 3], [239, 1]] 1 [239, [1152, 1]] 7 [241, [2, 346], [19, 11], [33, 89], [57, 7], [59, 25], [241, 1], [269, 35]] 1 [241, [2, 279]] 3 [251, [191, 3], [251, 1], [263, 3]] 0 --- 13 [257, [20, 112], [33, 39], [46, 2], [46, 56], [47, 19], [54, 4], [101, 15], [104, 8], [104, 40], [193, 13], [257, 1], [288, 44], [2147483647, 9]] 4 [257, [13, 10], [18, 3], [93, 2], [96, 9]] 3 [263, [29, 7], [263, 1], [648, 8]] 0 --- 5 [269, [53, 71], [101, 7], [127, 7], [269, 1], [1155, 41]] 0 --- 1 [271, [271, 1]] 1 [271, [18, 7]] 3 [277, [13, 83], [79, 61], [277, 1]] 4 [277, [2, 103], [71, 6], [720, 1], [510510, 5]] 12 [281, [6, 40], [14, 46], [19, 27], [34, 4], [42, 48], [50, 38], [93, 21], [96, 28], [181, 7], [197, 5], [281, 1], [828, 4]] 1 [281, [37, 74]] 3 [283, [2, 86], [48, 10], [283, 1]] 1 [283, [392, 55]] 3 [293, [15, 47], [281, 67], [293, 1]] 1 [293, [2, 53]][/CODE] How to read this data ? For example, if we take the two lines below : [CODE]5 [157, [20, 18], [20, 34], [60, 12], [65, 83], [157, 1]] 2 [157, [241, 2], [242, 1]][/CODE] This means that there are 5 sequences of the same parity that end in the prime 157 (which are 20^18, 20^34, 60^12, 65^83 and 157^1) and 2 sequences of opposite parity that end in the prime 157 (which are 241^2 and 242^1). [U]Now, given the rarity of sequences of opposite parity which end, there should be for each prime number p many more sequences of the same parity which end with this prime number than of sequences of opposite parity. This is not the case ![/U] You can see that the prime numbers that are exceptions are [B]2, 5, 7, 37, 41, 43, 59, 73, 113, and 277.[/B] I remind you that the work only analyzed prime numbers <300. There may be other primes >300 for which this will also be the case. I am checking with Karsten to see if he can modify his scripts to make such phenomena directly apparent. The case of the prime number 59 is particularly spectacular, because 108 sequences of opposite parity end in 59 while there are only 9 sequences of the same parity which end in 59 ! I see there a real paradox and I cannot explain it, because I noticed that the bases which are the doubles of squares do not explain this paradox ! [B]If someone finds an explanation, I would be very interested ![/B] [SIZE="3"][B][U]What will we do now ?[/U][/B][/SIZE] Again, I want to thank all the people who have worked so hard on this project so far. I would like to remind you that we are not at all sure that we will reach the goal we have set for ourselves. The laws linking the base, the exponent and the prime number or the cycle on which the sequences end may not exist. So to all, I say : continue this project only if it pleases you, because I can't guarantee any result. It's no problem for me if someone wants to withdraw. But I hope, of course, that you will all continue the work for this project ! If some of you want to continue to do calculations for this project, here are some possible directions. 1) It seems difficult to add exponents to the bases we have started so far. The calculations would become too long and so this option is to be discarded for the moment. We have done it for base 2 and it is very difficult : I see that even with a 128 threads machine, it is very long. On the other hand, it is very important that we continue as far as possible to complete as many high exponent sequences of the same parity as possible to add high "y" coordinate points on the curves I am looking for. 2) The other option is to add bases to add points with high "x" coordinates on the possible curves I am looking for. We now have all the bases for prime numbers up to 500 and that should be enough. If some of you want to add bases, I recommend adding bases that will follow the bases 106, (107), 108, (109), 110, 111, 112 and (113) that I am currently working on, in order to have all the bases exhaustively beyond 113. It would also be interesting to add 2 or 3 couples of amicable numbers and to continue to initialize bases that are part of the C28 cycle that we started a long time ago, and also some bases which would be factorial numbers. [B]And also, if anyone has their own ideas of things they would like to look at, they can suggest the bases that interest them.[/B] On my side, I will continue to look for new ideas to look at the data in other different ways. [B]Do not hesitate if you have any comments, or even criticisms to make following this post. [/B] :smile: |
[QUOTE=garambois;625591]
We notice that starting from 30, we always add 14 to the exponent. The next exponent is 58 + 14 = 72. Unfortunately, 648^72 is not a terminated sequence, I remain stuck at index 3 on a C171 cofactor. Almost all such "alerts" remain unverifiable for the same reasons: the terms in the next sequence are far too large to verify. :smile:[/QUOTE] I can crack the C171 for you, if you like. While I doubt we'll see the pattern continue, satisfying such curiosity is as good a reason as any to run a factoring job! |
This is great if you can afford to crack it.
But at the next indexes, you may come across a C200, and there, it will be another level of difficulty ! But I'm excited if you go for it... For the ECM methods, I went up to t40 and it still runs on my PC with one thread. Let me know if you take over ! :smile: |
I ran up through t50 on the c171 today.
I "might" attempt t55 tomorrow, but I'm not sure my cluster will actually do t55. I know it didn't like t60 when I tried it on something a while ago. |
Thank you both very much !
Me, I'm working on 2^577 and 2^583, as I'll still be able to run some CADO decompositions on the 128-threaded computer : my loan has been extended a bit. |
Thanks, Ed! I'll decide this weekend whether to run it locally or feed it to NFS@home e-small queue. It'll be done quicker on e-small.
I'll run a little ECM on it, too, maybe 0.25 * t55 |
[QUOTE=VBCurtis;625645]Thanks, Ed! I'll decide this weekend whether to run it locally or feed it to NFS@home e-small queue. It'll be done quicker on e-small.
I'll run a little ECM on it, too, maybe 0.25 * t55[/QUOTE]If you choose not to run it locally, would you want me to use adjusted params from the [URL="https://www.mersenneforum.org/showpost.php?p=622406&postcount=84"]harvest thread[/URL] and run it? If I interpret the post correctly, the only changes from the c170 would be:[code] tasks.lpb0 = 31 tasks.lpb1 = 31 tasks.sieve.mfb1 = 89[/code] |
Sure, be my guest! You have more hardware than I, by about a factor of three. I'm presently entertained with this "get 13^i to 160 digits" mini-project, and C171 would be nearly 2 weeks for me.
I have a second params adjustment to suggest: increase nq from 15625 to 78125, and decrease admax by a factor of 5. I'm finding better lognorm (stage 1) and poly scores at small ad-ranges, so searching more deeply down low seems profitable. I'm still "in testing" with this adjustment, but for C100-115 it seems to produce polys 5-7% higher scoring, with a smaller advantage around C140 (only tested on 2 jobs so far at C14x). |
My cluster completed t55 without success. I will be interested in what size does show up. I'm not 100% confident in my cluster as this is the first successful completion at t55.
Additional info on how I got to t55:[code] In addition to t20, t25, t30, t35, t40 runs, 2750 @ 11e6 4500 @ 43e6 4320 @ 26e7[/code]The last three calculated to 55.057 by wraithx.net. |
92^80 ends in a cycle
|
WOW !
And which cycle ? The one of length 28 ! This had already happened only for 3^286, outside the bases themselves belonging to the C28 cycle ! This will be the only sequence ending with a cycle for base 92 as it was for base 3. But for base 3, we were dealing with a sequence of opposite parity. Did you Sergiosi terminate this sequence ? |
Yes, I did the calculations for AS 92^80, starting at i4. The c144 there was the largest factorization needed for completing the sequence.
|
I have completed testing of new bases 106 & 108 all open sequences to >= 120 digits.
|
Great job, Sergiosi and Gary.
Thank you very much to you. |
48^5 merge info. has a typo. It shows as merged with sequence 15302802. It should be sequence 1530280.
Specifically: 48^5:i15 merges with 1530280:i2 A point of interest: 15302802 does merge with the main project: 15302802:i189 merges with 27860:i19. I also found it strange that 31^36 shows as merged with sequence 3762570. I verified that the merge info. is correct but I didn't know you were showing merges with sequences outside of the main project. I confirmed that neither 31^36 nor 3762570 merge with any sequences < 3M. I only caught these due to the new modifications to the page to include new data files by Karsten. |
OK for the typo, I will correct it for the next update.
Regarding merging with a number >3000000 outside the main project, you are right. Normally such merges are not noted. I don't understand how this merge can appear in the project, I don't understand its origin ? But should I delete this information which is correct ? |
From post# 956:
[QUOTE=RichD;558264]I think 31^36 may have merged.[/QUOTE]I traced the merge to 3762570. |
It happened a long time ago.
I didn't remember it at all ! Let's put this merger on the page, since we had already discussed it in the past. |
[QUOTE=garambois;626205]OK for the typo, I will correct it for the next update.
Regarding merging with a number >3000000 outside the main project, you are right. Normally such merges are not noted. I don't understand how this merge can appear in the project, I don't understand its origin ? But should I delete this information which is correct ?[/QUOTE] For consistency, I would delete it. But maybe store it somewhere in case the main project ever gets extended to 5M or 10M. The reason being is that there is no formal work being done on 3762570 and removing the merge info. allows our project to work on it if the interest is there. That's just my two cents. [QUOTE=EdH;626206]From post# 956: I traced the merge to 3762570.[/QUOTE] I did not see anything in post 956 about it. I do see it in post 835 here: [url]https://mersenneforum.org/showpost.php?p=572146&postcount=835[/url] That was in response to posts 830 and 833. It appears that Jean-Luc thought you had an error in your script causing it to not find a merge. In fact, there was no error if your script was only looking for merges with sequences <= 3000000. |
I found a typo for the 239^6 merge:
The page states: 239^6:i1681 merges with 2667000:i1 It should be: 239^6:i681 merges with 2667000:i1 1681 vs. 681 This one stuck out at me because it is the only merge on the n^i project in Karsten's new list that is at current size < 136 digits. It's only at 120 digits. |
[QUOTE=gd_barnes;626213]. . .
I did not see anything in post 956 about it. I do see it in post 835 here: [url]https://mersenneforum.org/showpost.php?p=572146&postcount=835[/url] That was in response to posts 830 and 833. . .[/QUOTE]The number should have been [URL="https://www.mersenneforum.org/showthread.php?p=558264#post558264"]596[/URL] instead of 956. That's where RichD first brought it up. |
[QUOTE=gd_barnes;626220]I found a typo for the 239^6 merge:
The page states: 239^6:i1681 merges with 2667000:i1 It should be: 239^6:i681 merges with 2667000:i1 [/QUOTE] Congratulations for noticing this new error ! I hope there are no more like it ! Of course, I will correct it in the next update. |
[QUOTE=gd_barnes;626213]For consistency, I would delete it. But maybe store it somewhere in case the main project ever gets extended to 5M or 10M. The reason being is that there is no formal work being done on 3762570 and removing the merge info. allows our project to work on it if the interest is there. That's just my two cents.
[/QUOTE] Thank you Gary for your opinion, that's how we move forward. I think the main project should be expanded to at least 4 or 5 million in not too long. Indeed, it has been a very long time since it has been expanded. So this sequence will inevitably move forward in a while. And since this information about its fusion is quite valid, I think I'll leave it for now. If this sequence should be of particular interest, we will soon notice again that it is not included in the main project if the main project should not be extended and we could then add the terms that we need. |
[QUOTE=garambois;626304]Thank you Gary for your opinion, that's how we move forward.
I think the main project should be expanded to at least 4 or 5 million in not too long. Indeed, it has been a very long time since it has been expanded. So this sequence will inevitably move forward in a while. And since this information about its fusion is quite valid, I think I'll leave it for now. If this sequence should be of particular interest, we will soon notice again that it is not included in the main project if the main project should not be extended and we could then add the terms that we need.[/QUOTE] OK. Sounds like a good explanation to me. One of the sequences that I tested to 120 digits in the last few months dropped to between 3000000 and 4000000 3 or 4 different times but it somehow did not merge with the main project. I will see if I can find it. Maybe you can note it as merged also. Maybe someone can do a merge analysis looking for merges with sequences < 5000000. |
I found the sequence from the previous post that I was talking about. It is 401^8. See here:
[url]https://mersenneforum.org/showpost.php?p=618494&postcount=2082[/url] It dropped as low as 3086468 at index 1061 and had 5 indexes in that area that were < 4000000. Since it never merged with any sequences < 3000000, 3086468 is likely the merge sequence. If so: 401^8:i1061 merges with 3086468:i0 This was very unusual! Is this the type of sequence you want to show a merge for? It's even more obvious than 31^36. |
I know, I'm kind of spamming the threads...
I have completed testing of new base 110 all open sequences to >= 120 digits. |
[QUOTE=gd_barnes;626306]
One of the sequences that I tested to 120 digits in the last few months dropped to between 3000000 and 4000000 3 or 4 different times but it somehow did not merge with the main project. I will see if I can find it. Maybe you can note it as merged also. [/QUOTE] Thanks a lot, I will add this merge in the next update. [QUOTE=gd_barnes;626306] Maybe someone can do a merge analysis looking for merges with sequences < 5000000.[/QUOTE] This is not without risk, because it is possible that sequences between 3,000,000 and 5,000,000 merge together ! To avoid such problems, when I run my program that allows to get the list of the last C80 of each sequence of the main project ([URL="https://www.mersenneforum.org/showthread.php?t=24423"]see here[/URL]), I upload the sequences on [URL="https://www.rechenkraft.net/aliquot/AllSeq.txt"]this page[/URL] to avoid merging. Ewin's "Alimerge" program can then work. |
Page updated
Many thanks to all for your work ! [URL="https://www.mersenneforum.org/showpost.php?p=626628&postcount=25"]See here.[/URL] |
I have completed testing of new bases 111 & 112 all open sequences to >= 120 digits.
|
Excellent Gary !
|
Page updated
Many thanks to all for your work ! [URL="https://www.mersenneforum.org/showpost.php?p=627002&postcount=31"]See here[/URL]. |
Page updated
Many thanks to all for your work ! [URL="https://www.mersenneforum.org/showpost.php?p=627703&postcount=35"]See here[/URL]. I noticed something quite unusual for the new base 116 that has just been added. Two sequences from this same base merge : 116^19:i849 = 116^25:i950 = 6552:i4 I must also point out that my computer is having very serious problems. It is becoming difficult for me to do any work. I have to restart the sequences every day. Dilemma : wait a few more months to be able to afford the computer of my dreams (64C/128T) or change now and get a smaller machine ? That's the question... |
Thanks Jean-Luc!
The merge is interesting to me in particular, because it shows the merging of the two base 116 sequences only a couple terms prior to the merge with 6552. This makes me wonder if there may be several sequences within the project which have merged with each other, but gone unnoticed because they don't merge with the main project. If 6552 wasn't involved, would the two merging have been caught? I may have to set up a program to see if I can find any such merges within the whole set. Bummer on the computer issues cropping up again. I can only suggest checking the things you've already done in the past: heat issues and power supply failure. If it is the power supply and you keep restarting, it may damage the motherboard. As to when to upgrade, that will be quite a decision to make - sorry I can't be of more help. More directly to the project, a lot of the matched parity sequences have current terms above 150 digits, which take me a while to bring down to 135* for that sub-project. Your addition of base 116 has helped that thread and the others you currently have pending will help, too. In light of the above, I'm wondering if I should initialize some new bases rather than bring existing sequences that have a current term >150 down to 135. Do you think it would be better to add more bases or extend existing sequences in regards to your research efforts? A side effect, is the addition of index 1 sequences. Many of those will be taken care of through the matched parity work, but others will have to be addressed alternately. A last note: As you already know, I will from time to time, at least, want to move my machines to other areas, which may include team projects that take me away for extended periods. I should be able to still maintain the threads, but I might not be able to work on sequences. * You might want to change the 145 to 135 on the main project page where you mention the matched parity thread. |
| All times are UTC. The time now is 07:08. |
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2023, Jelsoft Enterprises Ltd.