mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Aliquot Sequences (https://www.mersenneforum.org/forumdisplay.php?f=90)
-   -   Reserved for MF - Sequence 4788 (https://www.mersenneforum.org/showthread.php?t=11615)

 Batalov 2016-10-07 14:41

[QUOTE=Raman;444464]Last past time period[/QUOTE]
I hate when that happens.

 chris2be8 2016-10-09 15:59

Is it possible for the mods to prevent selected users from using coloured text?

Chris

 henryzz 2016-10-12 16:09

Is there an eta for the c184?

 ryanp 2016-10-12 16:11

[QUOTE=henryzz;444861]Is there an eta for the c184?[/QUOTE]

3 hours left on the LA, so adding in the square root phase, sometime this afternoon probably.

 Dubslow 2016-10-12 16:12

[QUOTE=henryzz;444861]Is there an eta for the c184?[/QUOTE]

I was just thinking we've been spoiled when I'm wondering "cmon what's taking so long already?!?" after merely a week or so of waiting on a f#(%$*g C184!  ryanp 2016-10-12 16:31 I may have undersieved this one a bit... been a bit busy :smile:  ryanp 2016-10-12 21:33 And, got it on the first dependency! [CODE]Wed Oct 12 14:33:15 2016 p70 factor: 5392786715992672170020355972774922016080484016352793347443385789938029 Wed Oct 12 14:33:15 2016 p114 factor: 480924752455922954122814527248090456980834275118681297460949256660360647327084111452415555580330657870922771620429[/CODE]  chalsall 2016-10-12 22:14 [QUOTE=ryanp;444883]And, got it on the first dependency!]/QUOTE] Why does this matter?  ryanp 2016-10-12 22:16 [QUOTE=chalsall;444887][QUOTE=ryanp;444883]And, got it on the first dependency!]/QUOTE] Why does this matter?[/QUOTE] I mean... because it saves me time and waiting for msieve? :smile:  Dubslow 2016-10-12 23:06 [QUOTE=chalsall;444887][QUOTE=ryanp;444883]And, got it on the first dependency![/QUOTE] Why does this matter?[/QUOTE] What do you know about the post processing phase of sieve method integer factoring? The answer to your question depends on what context you asked the question from. As far as I know, you're not particularly familiar with NFS. Is that correct?  chalsall 2016-10-12 23:18 [QUOTE=Dubslow;444893]What do you know about the post processing phase of sieve method integer factoring? The answer to your question depends on what context you asked the question from. As far as I know, you're not particularly familiar with NFS. Is that correct?[/QUOTE] I know all about that post processing phase. I paid for that post processing phase. Oh, wait, I have to pay for those females...? Never heard about them. How you dare you say I said anything about them. If you mention them I'm going to sue you. (Regaining conscience he threatens to sue you. Interesting work.)  henryzz 2016-10-12 23:19 [QUOTE=ryanp;444862]3 hours left on the LA, so adding in the square root phase, sometime this afternoon probably.[/QUOTE] A shame that it didn't lose the 3.  LaurV 2016-10-13 07:40 Lucky you Ryan! Beside of having tons of hardware at your pushbuttons, you are also a lucky son of a gun! I envy you! :smile: [QUOTE=chalsall;444887][QUOTE=ryanp;444883]And, got it on the first dependency![/QUOTE] Why does this matter?[/QUOTE] It saves a lot of time, processing each dependency takes a while, and if you are unlucky and ended up with 50 dependencies or so, which do not give you a factor and you got the factor at the 51st, then you will spend as much time in the square root phase as in the LA phase. I usually end up with ~30 dependencies for my factoring size, and get the factor after 5, or 8, or so.  pinhodecarlos 2016-10-13 08:05 [QUOTE=LaurV;444935] then you will spend as much time in the square root phase as in the LA phase.[/QUOTE] A bit exaggerated :shock:...Square root runs under one core if you don't have a script to run it in parallel while the LA uses all cores.  LaurV 2016-10-13 08:25 [QUOTE=pinhodecarlos;444937]A bit exaggerated[/QUOTE] Not at all. Last split I did for aliquot 714192 in this 2-cores laptop, LA in two cores took 15 minutes, and sqrt in single core took 30, with only 4 dependencies hammered. Well, I imagine they do not scale linearly in the 180-190 digits, of course, but thinking of 50 or a hundred dependencies... [CODE]Fri Oct 07 19:14:46 2016 commencing linear algebra Fri Oct 07 19:14:46 2016 read 421832 cycles Fri Oct 07 19:14:47 2016 cycles contain 1516465 unique relations Fri Oct 07 19:15:04 2016 read 1516465 relations Fri Oct 07 19:15:05 2016 using 20 quadratic characters above 67108352 Fri Oct 07 19:15:10 2016 building initial matrix Fri Oct 07 19:15:21 2016 memory use: 173.8 MB Fri Oct 07 19:15:26 2016 read 421832 cycles Fri Oct 07 19:15:26 2016 matrix is 421655 x 421832 (120.8 MB) with weight 40457484 (95.91/col) Fri Oct 07 19:15:26 2016 sparse part has weight 28701735 (68.04/col) Fri Oct 07 19:15:28 2016 filtering completed in 2 passes Fri Oct 07 19:15:29 2016 matrix is 420919 x 421096 (120.7 MB) with weight 40426912 (96.00/col) Fri Oct 07 19:15:29 2016 sparse part has weight 28693472 (68.14/col) Fri Oct 07 19:15:38 2016 matrix starts at (0, 0) Fri Oct 07 19:15:39 2016 matrix is 420919 x 421096 (120.7 MB) with weight 40426912 (96.00/col) Fri Oct 07 19:15:39 2016 sparse part has weight 28693472 (68.14/col) Fri Oct 07 19:15:39 2016 saving the first 48 matrix rows for later Fri Oct 07 19:15:39 2016 matrix includes 64 packed rows Fri Oct 07 19:15:39 2016 matrix is 420871 x 421096 (116.1 MB) with weight 31933087 (75.83/col) Fri Oct 07 19:15:39 2016 sparse part has weight 27904739 (66.27/col) Fri Oct 07 19:15:39 2016 using block size 32768 for processor cache size 0 kB Fri Oct 07 19:15:41 2016 commencing Lanczos iteration Fri Oct 07 19:15:41 2016 memory use: 93.8 MB Fri Oct 07 19:16:04 2016 linear algebra at 2.9%, ETA 0h12m Fri Oct 07 19:28:42 2016 lanczos halted after 6656 iterations (dim = 420871) Fri Oct 07 19:28:42 2016 [COLOR=Red]recovered 33 nontrivial dependencies[/COLOR] Fri Oct 07 19:28:43 2016 BLanczosTime: 837 Fri Oct 07 19:28:43 2016 Fri Oct 07 19:28:43 2016 commencing square root phase Fri Oct 07 19:28:43 2016 reading relations for dependency 1 Fri Oct 07 19:28:43 2016 read 210596 cycles Fri Oct 07 19:28:43 2016 cycles contain 757954 unique relations Fri Oct 07 19:28:51 2016 read 757954 relations Fri Oct 07 19:28:55 2016 multiplying 757954 relations Fri Oct 07 19:32:11 2016 multiply complete, coefficients have about 30.98 million bits Fri Oct 07 19:32:12 2016 initial square root is modulo 788896981 Fri Oct 07 19:37:07 2016 [COLOR=Red]GCD is 1, no factor found[/COLOR] Fri Oct 07 19:37:07 2016 reading relations for dependency 2 Fri Oct 07 19:37:07 2016 read 210830 cycles Fri Oct 07 19:37:07 2016 cycles contain 758980 unique relations Fri Oct 07 19:37:16 2016 read 758980 relations Fri Oct 07 19:37:20 2016 multiplying 758980 relations Fri Oct 07 19:40:35 2016 multiply complete, coefficients have about 31.02 million bits Fri Oct 07 19:40:37 2016 initial square root is modulo 811013293 Fri Oct 07 19:45:31 2016 [COLOR=Red] GCD is N, no factor found[/COLOR] Fri Oct 07 19:45:31 2016 reading relations for dependency 3 Fri Oct 07 19:45:31 2016 read 210596 cycles Fri Oct 07 19:45:32 2016 cycles contain 758060 unique relations Fri Oct 07 19:45:40 2016 read 758060 relations Fri Oct 07 19:45:44 2016 multiplying 758060 relations Fri Oct 07 19:48:59 2016 multiply complete, coefficients have about 30.98 million bits Fri Oct 07 19:49:01 2016 initial square root is modulo 789339269 Fri Oct 07 19:53:55 2016 [COLOR=Red]GCD is 1, no factor found[/COLOR] Fri Oct 07 19:53:55 2016 reading relations for dependency 4 Fri Oct 07 19:53:55 2016 read 210810 cycles Fri Oct 07 19:53:55 2016 cycles contain 758148 unique relations Fri Oct 07 19:54:04 2016 read 758148 relations Fri Oct 07 19:54:08 2016 multiplying 758148 relations Fri Oct 07 19:57:23 2016 multiply complete, coefficients have about 30.99 million bits Fri Oct 07 19:57:25 2016 initial square root is modulo 792739931 Fri Oct 07 20:02:19 2016 sqrtTime: 2016 [/CODE]  unconnected 2016-11-08 09:58 No update for a week, seems like Ryan run GNFS on c189@i10740.  ryanp 2016-11-21 00:56 We are moving again! :smile:  Batalov 2016-11-21 01:28 Dude! Very lucky! Fifth time could be the charm, finally?  Prime95 2016-11-21 04:17 Amazing!!! :fusion::bow::bow::bow::bow::bow::bow::fusion:  Dubslow 2016-11-21 04:55 :surprised: If only my sequences caught this many down drivers...  LaurV 2016-11-21 06:39 Noooo.. That's impossible! :facepalm: Now I can not anymore wait until Ryan gives up and say "you said it pisses you off and you are going to kill it?"... :sad:  bsquared 2016-11-21 22:28 Unreal... amazing work Ryan. Hoping this one goes all the way! :popcorn:  henryzz 2016-11-21 23:43 And it starts again...  flagrantflowers 2016-12-01 01:29 So what are the possible drivers we can gain? and under what circumstances?  Dubslow 2016-12-01 01:42 [QUOTE=flagrantflowers;448112]So what are the possible drivers we can gain? and under what circumstances?[/QUOTE] The downdriver is a class 1 driver, meaning the total powers of 2 in sigma(odd part) cannot exceed 1. Given that the odd part has at least two factors, and that sigma(prime) is even and there's at least two such primes, the total powers of sigma(C) in 2*C is at least 2 (and we can go a bit further, since C is 3 mod 4, we can in fact say that there's at least three powers of 2 in the sigma of the primes in C). So no mutation. Only 2 * P (* Q^2) where P is 1 mod 4 will result in a mutation (any even powers of primes don't count, so when I say "only 2*p" I mean after ignoring perfect squares).  LaurV 2016-12-01 03:07 To be clear, we are talking about index 10796 which is 2*C185 (otherwise this talk will have no relevance after few more terms are added, nobody will know what we were talking about). Splitting C185 in 2 (or more) primes will not change the downdriver, as Dubslow said. Except of course when the split is p*q^2 with p prime and 1 (mod 4), and q a product of one or more primes, which is very unlikely at this point. We "need" the term to be a 2 multiplied by a [U]single[/U] prime (squares ignored) to have a 50% "chance" to lose the downdriver. I use quotes because we don't actually "need" that :razz:, and the "chance" in that case will be misfortune... hehe...  ATH 2016-12-01 11:07 [QUOTE=LaurV;448122]I use quotes because we don't actually "need" that :razz:, and the "chance" in that case will be misfortune... hehe...[/QUOTE] Also called the [I][B]risk[/B][/I]. It seems that the word "risk" is not used very often? At least I often see "chance" being used to describe something negative, and it looks wrong to me, and I'm thinking: "Are they really hoping for that outcome?". I guess I was brought up using "risk" so it comes natural.  ryanp 2016-12-01 18:12 [QUOTE=LaurV;448122]To be clear, we are talking about index 10796 which is 2*C185 (otherwise this talk will have no relevance after few more terms are added, nobody will know what we were talking about). Splitting C185 in 2 (or more) primes will not change the downdriver, as Dubslow said. Except of course when the split is p*q^2 with p prime and 1 (mod 4), and q a product of one or more primes, which is very unlikely at this point.[/QUOTE] This C185 has survived a lot of ECM at this point, so I think the likelihood of it having more than 2 factors is pretty low. GNFS starting now...  LaurV 2016-12-02 09:21 [offtopic] [QUOTE=ATH;448144]Also called the [I][B]risk[/B][/I]. It seems that the word "risk" is not used very often? At least I often see "chance" being used to describe something negative, and it looks wrong to me, and I'm thinking: "Are they really hoping for that outcome?". I guess I was brought up using "risk" so it comes natural.[/QUOTE] We didn't know we can use "risk" in this context. We learn something new everyday. For us, risk is the money we lose when we trade Forex... :smile: For us, "chance" is a percent or a number between 0 and 1. We were never thinking to "risk" as a percent or under-unit number (like in probabilities). We see "risk factor" as possibility (unquantifiable) not as a probability (numbers). Even now after your comment, we are tempted to say "the chance of risk". OTOH, we guessed somehow that a word for "negative chance" or "chance of jinxing it" must exist in English, only we didn't know the word. Thanks for the info... [/offtopic]  Dubslow 2016-12-02 11:07 Consider "the risk of failure" versus "the chance of success". Risk can be used in much the same way, i.e. as a probability between 0 and 1. "Chance of risk" would be seen as redundant -- not a phrase a native speaker would use (at least not in my experience; languages, especially English, have a certain way of contradicting any firm statement one might make about them).  ryanp 2016-12-05 03:17 [CODE]Mon Dec 5 03:10:41 2016 p84 factor: 185596989654008443816154517215981298046822949687813530269384349965625218099819665917 Mon Dec 5 03:10:41 2016 p102 factor: 144103322597284487319028362827830047182446040144943320906819406487398249398651395159960993902986495231[/CODE]  flagrantflowers 2016-12-09 16:00 Below 180 digits.  Batalov 2016-12-14 03:45 2 Attachment(s) Oh noes!  axn 2016-12-14 04:26 There's still hope ...  Stargate38 2016-12-14 23:10 Given the fact that the sequence has 2^4, but doesn't have a 3 or 31, hopefully the sequence will keep decreasing or flatten out until it catches the downdriver again. If it terminates, maybe RyanP can help us out with another sequence, like 660. However, that sequence is already reserved by Paul Zimmerman, and I don't think he'll be unreserving it anytime soon. The smallest unreserved sequence (as of this post) is 305460, so maybe that could be Ryan's next sequence.  science_man_88 2016-12-14 23:14 [QUOTE=Stargate38;449179]Given the fact that the sequence has 2^4, but doesn't have a 3 or 31, hopefully the sequence will keep decreasing or flatten out until it catches the downdriver again. If it terminates, maybe RyanP can help us out with another sequence, like 660. However, that sequence is already reserved by Paul Zimmerman, and I don't think he'll be unreserving it anytime soon. The smallest unreserved sequence (as of this post) is 305460, so maybe that could be Ryan's next sequence.[/QUOTE] it caught a 5 at one point and climbed I think that's what axn's post was about.  ryanp 2016-12-15 00:03 Dodged a bullet, perhaps? We had a 2^4 · 7 briefly but it went away.  Batalov 2016-12-15 02:26 2 Attachment(s) Oh yez!  LaurV 2016-12-15 05:11 [QUOTE=science_man_88;449180]it caught a 5 at one point and climbed I think that's what axn's post was about.[/QUOTE] [QUOTE=ryanp;449183]Dodged a bullet, perhaps? We had a 2^4 · 7 briefly but it went away.[/QUOTE] 5 and 7 can not stick with 2^4. Only 31 is dangerous here (it was, actually, now we are back to DD).  Dubslow 2016-12-15 21:42 :w00t:  Batalov 2016-12-17 19:35 1 Attachment(s) [COLOR=LemonChiffon].[/COLOR]  mshelikoff 2016-12-17 21:02 2 Attachment(s) [ATTACH]15343[/ATTACH][ATTACH]15345[/ATTACH]  Prime95 2016-12-17 21:14 Below 160 digits. Go, ryan, go.  Stargate38 2016-12-18 01:16 Over 11000 terms! Current index value is a palindrome (11011). Hopefully this downdriver will stick. This could beat the record that 933436 is holding.  ryanp 2016-12-18 07:09 [QUOTE=Prime95;449387]Below 160 digits. Go, ryan, go.[/QUOTE] Almost there! :smile:  rajula 2016-12-18 14:02 And now below 140. Let's see what happens this time..  science_man_88 2016-12-18 17:17 now below 130 digits  axn 2016-12-18 17:40 [QUOTE=Stargate38;449395]Hopefully this downdriver will stick. [/QUOTE] Most probably not. We'll probably see another 20000 lines!  unconnected 2016-12-18 22:20 Now below 120 digits! :cool:  Stargate38 2016-12-18 22:21 I meant until it was below 30 digits. What's the chance that it'll hold it that long? Seems pretty high, given that the sequence has dropped this far. EDIT: Oops! I was hoping it would hold it longer.  richs 2016-12-18 22:26 Just lost the downdriver at i11181. :bangheadonwall:  Stargate38 2016-12-18 22:34 It may regain it again, like it did at i10935. There's no driver at the moment, only a mutating guide. At the current rate, we're getting a new term every 1-2 minutes.  Batalov 2016-12-18 22:49 To make it a little easier for the database, let's [URL="http://factordb.com/sequences.php?se=1&aq=91901038409919778877643157776565267901484711624094452804992470776726931834980411313523560256435290609998758178617234&action=last&fr=0&to=100"]watch progress[/URL] from element #11180 ?  Stargate38 2016-12-18 22:55 That's smart. Using a later element from the sequence prevents the DB from getting flooded with 4788 requests. Uh-oh! It has 2^5*7. Thank goodness there's no 3. EDIT: Just lost the 2^5*7.  Prime95 2016-12-18 23:37 [QUOTE=richs;449466]Just lost the downdriver at i11181. :bangheadonwall:[/QUOTE] This sequence must hold the record for downdrivers lost (and the record for reacquired).  henryzz 2016-12-18 23:56 [QUOTE=Prime95;449472]This sequence must hold the record for downdrivers lost (and the record for reacquired).[/QUOTE] That may be the case although it probably needs checking. My other thought is whether it is possible to construct a sequence with more.  science_man_88 2016-12-19 00:36 2^4*5  henryzz 2016-12-19 00:54 Downdriver again. We very narrowly missed 2*3.  Batalov 2016-12-19 00:55 Yo! = [URL="http://factordb.com/index.php?id=2"][COLOR=#000000]2[/COLOR][/URL] · [URL="http://factordb.com/index.php?id=541"][COLOR=#000000]541[/COLOR][/URL] · [URL="http://factordb.com/index.php?id=2161"][COLOR=#000000]2161[/COLOR][/URL] · [URL="http://factordb.com/index.php?id=1100000000888668536"][COLOR=#000000]5680960716188988151[SUB]<19>[/SUB][/COLOR][/URL] · ...  ryanp 2016-12-19 01:02 Onward and (hopefully) downward...  science_man_88 2016-12-19 02:47 now below 110 digits sorry spamming the thread.  Happy5214 2016-12-19 03:08 At 100 digits right now.  Batalov 2016-12-19 03:10 ...and under 100...  firejuggler 2016-12-19 03:43 Shoot, 2^2*3  Batalov 2016-12-19 03:50 [QUOTE=firejuggler;449487]Shoot, 2^2*3[/QUOTE] [YOUTUBE]HoSOuYNNXjU[/YOUTUBE]  LaurV 2016-12-19 03:57 2^2*3 is good as long as no 7... 3s come and go... It starts looking like my forex charts... Elliott waves... trendlines... grrr... hahaha... On another front, we just added ~1800 terms to 991080 overnight, which also lost the DD and now is raising... :sad:  Batalov 2016-12-19 04:10 That Ryan chap, he is luckier than a dog with ... ehrmm...  firejuggler 2016-12-19 04:25 90 digit, it is a matter of time before we despair or rejoice.  mshelikoff 2016-12-19 04:29 Is there an archive of the name changes to this thread and when they took place?  Batalov 2016-12-19 04:34 Yes, there is.  mshelikoff 2016-12-19 04:45 Oooo...how is it visible? Can mere mortals see it? Who changes the name? Am I asking too many questions? There should be a huge graph of the sequence with a label every time there was a change in the name of the thread and the new name each time. It would be a little like Minard's map of Napoleon's 1812 Russian campaign.  Batalov 2016-12-19 05:00 I see some 2^2 * 7 ahead. :bangheadonwall:  Batalov 2016-12-19 05:12 [QUOTE=mshelikoff;449499]Oooo...how is it visible? Can mere mortals see it? Who changes the name? Am I asking too many questions? There should be a huge graph of the sequence with a label every time there was a change in the name of the thread and the new name each time. It would be a little like Minard's map of Napoleon's 1812 Russian campaign.[/QUOTE] [CODE] 04:24, 20th Jul 2009 10metreh Thread title (original 'Sequence 4788') changed 11:12, 20th Jul 2009 10metreh Thread title (original 'Frdhrapr 4788') changed 00:24, 21st Feb 2010 10metreh Thread title (original 'Sequence 4788') changed 18:56, 21st Feb 2010 Mini-Geek Thread title (original 'ecneuqeS 8874') changed 00:58, 1st Apr 2010 10metreh Thread title (original 'Sequence 4788') changed 11:32, 1st Apr 2010 10metreh Thread title (original ' (upside-down letters)) changed 02:11, 15th Jul 2013 Batalov Thread title (original 'Sequence 4788') changed 02:11, 15th Jul 2013 Batalov Thread title (original 'Sequence 4788 and/or 3048') changed 08:28, 4th Aug 2013 henryzz Thread title (original 'Sequence 4788 and/or 3408') changed 12:59, 6th Jul 2016 Batalov Thread title (original 'Sequence 4788') changed 18:31, 18th Sep 2016 Prime95 Thread title (original 'Sequence 4788 - the new miracle :-)') changed 14:08, 20th Sep 2016 Prime95 Thread title (original 'Sequence 4788 - high hopes dashed :(') changed 15:14, 24th Sep 2016 Batalov Thread title (original 'Sequence 4788 - hope renewed :)') changed 17:32, 20th Nov 2016 Batalov Thread title (original 'Sequence 4788 - hope reneged :]') changed 13:46, 15th Dec 2016 Batalov Thread title (original 'Sequence 4788 - hope renewed again!') changed 15:30, 18th Dec 2016 Batalov Thread title (original 'Sequence 4788 - hope reviewed again!') changed [/CODE]  axn 2016-12-19 05:28 [QUOTE=Batalov;449500]I see some 2^2 * 7 ahead. :bangheadonwall:[/QUOTE] Well, I guess that's it for another 2000 iterations.  ryanp 2016-12-19 18:15 Sigh. What do we need to get rid of the 2^2*7?  science_man_88 2016-12-19 18:47 [QUOTE=ryanp;449540]Sigh. What do we need to get rid of the 2^2*7?[/QUOTE] [QUOTE="http://www.rechenkraft.net/aliquot/analysis.html"]For example, when the 2^2 · 7 driver takes the form 2^2 · 7^2, its class of -1 temporarily increases by 3 (the 2s count of 7) so a mutation will occur when the 2s count of t is 2 or 1.[/QUOTE] a change in certain values to allow for a class change so a mutation can occur and get out of it potentially.  Batalov 2016-12-19 20:53 You do need a 2^2 * 7^2 (or 2^2 * 7^4 or other even degree) and one prime* ...or two primes of 4k+1 type. The excitement (and the shortcoming!) of aliquot sequence riding is sort of like surfing: you need a good wave and then you have even less control than in surfing; you just ride the wave. It can be even more adequately compared to the joy of playing pinball. [YOUTUBE]4AKbUm8GrbM[/YOUTUBE] _____________ * and that prime needs to be bla, bla, bla, you know: Many additional conditions.  Drdmitry 2016-12-23 16:19 The driver has been lost [URL="http://factordb.com/sequences.php?se=1&aq=4788&action=last20"]again[/URL]. This song will play forever. :)  science_man_88 2016-12-23 16:33 [QUOTE=Drdmitry;449805]The driver has been lost [URL="http://factordb.com/sequences.php?se=1&aq=4788&action=last20"]again[/URL]. This song will play forever. :)[/QUOTE] too bad all it will take for it to be a driver again ( okay 2^3 is a guide technically) is gaining a 5 or a 3. [URL="http://factordb.com/sequences.php?se=1&aq=255145911385930884813469706779422046222428384495923114580386266092670514219708202002510068356025542943673015531053573837009278762195360327910824476973993091288765112152&action=last&fr=0&to=100"]this[/URL] is a better link potentially if you don't want to look up the whole sequence and just see the last line.  Drdmitry 2016-12-23 16:42 [QUOTE=science_man_88;449806]too bad all it will take for it to be a driver again ( okay 2^3 is a guide technically) is gaining a 5 or a 3. [URL="http://factordb.com/sequences.php?se=1&aq=255145911385930884813469706779422046222428384495923114580386266092670514219708202002510068356025542943673015531053573837009278762195360327910824476973993091288765112152&action=last&fr=0&to=100"]this[/URL] is a better link potentially if you don't want ot look up the whole sequence and just see the last line.[/QUOTE] 2[SUP]3[/SUP] without a factor of 5 or 3 is not a guide. It will be stable until the power of two changes.  science_man_88 2016-12-23 16:44 [QUOTE=Drdmitry;449808]2[SUP]3[/SUP] without a factor of 5 or 3 is not a guide. It will be stable until the power of two changes.[/QUOTE] [url]http://www.rechenkraft.net/aliquot/analysis.html[/url] list it as a class 3 guide it may not class as a driver though.  Dubslow 2016-12-23 17:48 [QUOTE=Drdmitry;449808]2[SUP]3[/SUP] without a factor of 5 or 3 is not a guide. It will be stable until the power of two changes.[/QUOTE] In SM_88's link, I followed [URL="https://www.rechenkraft.net/aliquot/analysis.html"]Clifford Stern's definition[/URL]: [quote]A guide consists of$2^a$($a \gt 0$) along with a subset of the prime factors of$\sigma(2^a)\$.[/quote]

I interpreted it in the following way, which I believe to be at least perfectly reasonable if really just a question of meaningless semantics.

[quote]This motivates the following definition: a number of the form 2^b \cdot v, where v divides \sigma(2^b), is called a guide. (Note that when v=1, the guide is just the power of 2.)[/quote]

Actually, technically by your statement the down"driver" wouldn't be considered a guide. I think that is certainly enough to justify the empty powers of two as guides also...

 ryanp 2016-12-24 00:42

Maybe it's time to change my username to [URL="https://en.wikipedia.org/wiki/Sisyphus"]Sisyphus[/URL]...

 LaurV 2016-12-24 02:58

[QUOTE=ryanp;449837]Maybe it's time to change my username to [URL="https://en.wikipedia.org/wiki/Sisyphus"]Sisyphus[/URL]...[/QUOTE]
Told you so... Trust the [URL="https://en.wikipedia.org/wiki/Old_salt"]Old Salt [/URL]next time... :razz:
Now, don't be sad, from how the graph looks like, we may find the first cycle in the history of man, with a period 7000 or so... :w00t:

 VBCurtis 2017-01-03 23:20

Ugh. The 7.

 LaurV 2017-01-06 12:23

good, now it is me again who owns the largest downdriver! :razz:
(which is 613068, at 153 digits, not yet updated into Dubslow's table)

 Batalov 2017-01-06 15:26

Well, don't get your hopes too low! :rolleyes:

 ryanp 2017-01-07 01:37

[QUOTE=Batalov;450579]Well, don't get your hopes too low! :rolleyes:[/QUOTE]

Except that we're rapidly approaching 200 digits... :bangheadonwall:

Need that downdriver ASAP.

 unconnected 2017-02-22 09:43

No progress from Jan 18, did Ryan still working on it?

 ryanp 2017-02-23 00:09

[QUOTE=unconnected;453449]No progress from Jan 18, did Ryan still working on it?[/QUOTE]

I'm taking a break for now and handing this one back to the forum for the time being. :sad:

 swellman 2017-03-24 20:12

Do my eyes deceive me or is this train rolling again?

 unconnected 2017-03-24 21:07

A little bit. BTW, C142 is ready for GNFS.

 Stargate38 2017-03-27 16:35

Only 68 terms needed to become the new record sequence (in terms of length)! :smile:

 LaurV 2017-03-28 09:50

With 68 terms more, with the current driver and without losing the 3, it will grow about 19 digits in size too, therefore it will not only be the length record, but also (at 196+19=215 digits) one "absolute" record.

 ryanp 2017-03-28 22:55

I'll take this a bit further to see if we can avoid that fate. :smile:

GNFS for the c184 blocker is underway.

 LaurV 2017-03-29 01:39