Re: Mersenne: Why is trial factoring of small exponents slower than large ones?
Geoffrey Reynolds wrote: For the exponents currently being assigned from primenet it takes this machine about 12 minutes to factor from 2^57 to 2^58. I thought I would try factoring some small exponents (under 1,000,000) from the nofactors.zip file. I put FactorOverride=64 into prime.ini and started mprime as usual but progress is _much_ slower, it will take about 8 hours to factor from 2^57 to 2^58. Can someone tell me why the time difference is so great? Any factor of a Mersenne number 2^p-1 (with p an odd prime) must be of the form 2kp+1, whete k is some integer and p is the exponent of 2 in the Mersenne number. If p is small, the potential factors are closer together, and thus there are more of them in a given range (such as between 2^57 and 2^58), than if p were larger. E.g., for 2^19373911-1, the values of 2kp+1 are 38747823, 77495645, 116243467, ..., spaced 38747822 apart. For 2^757-1, the values of 2kp+1 are 1515, 3029, 4543, ..., spaced 1514 apart. In any given interval (such as between 2^57 and 2^58) there are about 26,000 times as many of the 2*k*757+1 candidates as there are of the 2*k*19373911+1 candidates. So trial factoring between 2^57 and 2^58 for 2^757-1 takes about 26,000 times as long as trial factoring the same range of candidate factors for 2^19373911-1. Prime95 trial factoring code uses various sieves to eliminate some of these candidates before actually trying to divide by them, but the proportion thus eliminated is roughly the same regardless of exponent size, so the ratio of trial factoring times remains roughly equal to the inverse ratio of exponents. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Re: Mersenne Digest V1 #1038
to falsely attribute untenable opinions to that person. [snip] That's the trouble with climbing into a pulpit to preach to others, it often has a habit of swinging back around and biting you. _I_ don't feel bitten. And before you go off and waste hours trawling through all the masses of data on the project files to try and work out if I have ever poached any numbers, I'll save you the bother. Yes I have. How many? - can't remember, less than 20 or so. Will I do it ever again? - don't know. I hadn't even _thought_ of trawling through files to see whether you had ever poached any assignments, until I read that last paragraph of yours. Thank you for your revelation -- I commend your honesty. Earlier I separately posted my comments (Mersenne: 80% instead of 100%?) on the contrast between this last paragraph and your previous claim about trustworthiness of Mersenne prime discoverers. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Poaching -- Definition
Aaron wrote: For the most part though, it seemed that once upon a time there was a problem with poachers who just took small exponents from people who were actually still working on them, and simply ran them on a faster machine or something. That's the sort of thing that gives us respectable poachers a bad name. :-D ... and that's the sort of thing that happened to me. Still though, IANAL but it would seem that since these numbers are public domain, and nobody can be said to have any special claim to any of them, you probably wouldn't have much legal standing if you wanted to go after someone who poached a number from you that turned out to be a prize winning prime. Poaching in GIMPS/PrimeNet isn't a matter of legality; it's a matter of rules of the game and fair play. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Poaching -- Definition (was: Mersenne: Re: Mersenne Digest V1 #1038)
Paul Missman wrote: I know that this might be earth shattering news for you, but there is no such thing as poaching. I think that folks who've been following the poaching discussion from the beginning know that there is indeed such a thing, and what it is. But let me post a refresher for the sake of newcomers. (Sayyy ... this ought to be in the PrimeNet FAQ! I'll volunteer to write it up as such, once the discussion has proceeded to the point where it seems that we have sound definitions.) In the context of GIMPS/PrimeNet, poach is used by analogy to certain more widespread uses of the verb. From Webster's Third New International Dictionary [the square-bracketed words are expansions of the dictionary's abbreviations]: poach // [verb] ... 3a: to trespass on (a field [poach]ed too frequently by the amateur) -- often used with _on_ or _upon_ (what happens to a poet when he [poach]es upon a novelist's preserves -- Virginia Woolf) b: to take (game or fish) by illegal methods ... [verb intransitive] ... 3: to trepass for the purpose of stealing game : take game or fish illegally (had taken to [poach]ing as a means of supplying fresh meat for the table -- H.D.Quillin) 4: to play a ball in a racket game that should normally be played by one's partner In GIMPS/PrimeNet, what is being trespassed upon is an exclusive assignment by GIMPS (represented by George Woltman) or by PrimeNet to L-L test a specific Mersenne number for primality. Since GIMPS/PrimeNet is not a government entity, instead of laws we have rules, so instead of illegal we mean in violation of the established rules for assignments. Our preserves are the GIMPS/PrimeNet assignments. Our game or fish is the privilege to be the exclusive tester of the specific Mersenne number which has been assigned. That privilege includes the right to be the first person to know the result of a first-time Lucas-Lehmer primality test, and, if that result is that the number is prime, the right to be designated as the discoverer of that particular Mersenne prime. The latter right is of considerable importance and weight within the world of mathematics, and thus is deemed to have a high value, not at all trivial, in our context. For doublechecking assignments, the privilege associated with an exclusive GIMPS or PrimeNet assignment includes the right to be the first person to know the result of a doublechecking Lucas-Lehmer primality test, and, if that result is that the number is prime, the right to be designated as the discoverer of that particular Mersenne prime if it turns out that the first-time result of nonprimality was incorrect. Also, the GIMPS/PrimeNet assignment system can be considered analogous to rules for determining which partner in a racket game is to be the one to play a ball. Neither GIMPS or PrimeNet have any license to these numbers, But they have rules of the game, and poaching is a violation of those rules. The GIMPS/PrimeNEt assignment system was established for at least these reasons: (a) to make work on Mersenne numbers efficient by avoiding needless duplication of effort, and (b) to provide world-wide incentive for people to participate by establishing the privileges listed above. Poaching works against those goals. Pretending that GIMPS/Primenet poaching concerns the laws of a governmental entity, or that governmental legality is relevant to our discussion would be a straw man type of argument. nor are they the only entities testing large numbers for primality. They _are_ the established clearinghouse for testing _Mersenne_ numbers for primality. The M in GIMPS stands for Mersenne. Neither GIMPS nor PrimeNet attempts to make assignments for any other category of number. If my sister reads from her math book a method of testing large primes, knows nothing of PrimeNet or GIMPS, tests the numbers on her home computer, and finds a large prime, she is gonna publish it. ... and (provided the number were a Mersenne prime in particular) she would not be deemed to have poached if she did so in ignorance of GIMPS and Primenet. But the poachers with which this discussion is concerned are those who _do_ know about the GIMPS/Primenet assignment system but nevertheless deliberately choose to violate its assignment rules. She might choose to send any results to GIMPS, or not. Okay. She might double check it using GIMPS provided software, or not. If she did, it would weaken any claim of ignorance of the assignment system. But for sure nobody has any reason to prevent her from doing any of this. ... _if_ she genuinely was working outside GIMPS/PrimeNet and was not aware of the GIMPS/PrimeNet assignment system. There simply is no real problem here that is begging for solution. Some folks who _have_ followed this discussion and have participated in GIMPS a long time might like to pretend so. But, yes, there is a real problem here. Richard Woods
Mersenne: 80% instead of 100%? (was Re: Mersenne Digest V1 #1038)
As I finished reading Gordon Spence's latest post, I was startled by his last paragraph. Gordon Spence wrote: And before you go off and waste hours trawling through all the masses of data on the project files to try and work out if I have ever poached any numbers, I'll save you the bother. Yes I have. How many? - can't remember, less than 20 or so. Will I do it ever again? - don't know. So I checked digest #1038. There, Gordon Spence wrote, in response to an earlier question of mine: Is there any particularly _special_ relationship between being a Mersenne prime discoverer and being trusted not to poach? Unless there is some such special relationship, I imagine that thousands of non-discoverers could also be trusted not to poach. Well in actual fact, there *is* now that you come to mention it. As a Mersenne Prime discoverer I am given immediate notification of any new MP immediately it is reported, ie *before* it is verified. We are trusted to keep it quiet.because when we discovered ours we proved that we were capable of _discretion_ Back to Mr. Spence's latest post -- in item 11, just four paragraphs above his last one which I quoted above, he wrote: My point is that there are a small number of people who we know for 100% certain can be trusted to act with discretion when sensitive information is involved. Now, to me, until I got to the final paragraph, it seemed that Mr. Spence had been claiming in this discussion that Mersenne prime discoverers could be _more_ trusted not to poach than other people could. But let's look at that final paragraph again: And before you go off and waste hours trawling through all the masses of data on the project files to try and work out if I have ever poached any numbers, I'll save you the bother. Yes I have. How many? - can't remember, less than 20 or so. Will I do it ever again? - don't know. In other words, the man claiming that he could be especially trusted admits _having already poached, perhaps up to 20 or so times_. I suppose I should commend his honesty. So, maybe only 80% of Mersenne prime discoverers during the GIMPS era can be trusted not to poach, instead of 100%? Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Communication between GIMPS Forum and Mersenne mailing list
I wrote: The point I wish to make is that this sort of thing will continue to happen as long as GIMPS has significant discussions in independent media without extensive (and probably not practical) cross-communication. This is not necessarily all negative. Along with the needless duplication of effort also comes new ideas that did not rise, or might not have arisen, in the separate discussion. But IMO we still need to consider whether the separate discussions are best for GIMPS in the long run. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Poaching and related issues...
Steinar H. Gunderson wrote: Early GIMPS and the other project (Slowinski/Cray) had no common agreement or method for avoiding duplication. Umm, I've not been in this project _that_ long, but at least a year or two before Primenet got integrated into the main client. At least at that time, all communication was done by e-mailing George requests for ranges (a list of free ranges was available on the GIMPS website), George solving conflicts if two people requested the same range. (All results were also handed back to George via e-mail, of course.) Are you talking even older than that? :-) What I was referring to was not overlap of work within GIMPS, or duplication of effort by two different GIMPS participants. I was referring to there being no agreement, at that time, to avoid duplication between (a) George Woltman and (b) the team of David Slowinski and Paul Gage (who were testing Mersennes, independently of George, on Cray computers). George Woltman just _barely_ missed out on becoming a Mersenne prime discoverer back then. (But he was not a poachee bcause there existed no reservation system between Slowinski Gage, and him. It was just happenstance that Slowinski and Gage completed L-L testing M1257787 while George was in the midst of L-L testing that very same number.) Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Poaching and related issues...
Nathan Russell wrote: Okay, to start with, GIMPS lost the very first prime we ever found to a member of another project who beat George to finding the exponent by a matter of hours. This is simply the way math and other fields of research work. Darwin's theory of evolution was very nearly duplicated by another researcher working independently. So were some of Edison's improvements on the telegraph. But with regard to Primenet poaching, there is a crucial difference from each of the three cases you cite. Within Primenet, just as within many other distributed computational projects, there is a reserving/assigning system for the purpose of minimizing unnecessary duplication or overlap of work. The cases of conflict you cite had no such method for avoiding duplication/overlap. Early GIMPS and the other project (Slowinski/Cray) had no common agreement or method for avoiding duplication. Darwin and the other guy (Wallace) had made no arrangement to work on separate theories. Edison was in frank competition with other inventors; he wouldn't have even tried to cooperate for nonduplication, I think. However, Primenet cannot _enforce_ its nonduplication/nonoverlap policy. Primenet can control exclusiveness of its assignments (who shall perform this work unit), but the exclusiveness of its reservations (only one participant shall perform this work unit while assigned) depends on voluntary cooperation by participants. Poaching, in the Primenet context, is noncooperation with the voluntary, cooperative reservation rules. If a 'poacher' beat me to a prime I'd be very upset. That's one reason why some of us want to prevent poaching if possible. It's not likely to happen to me, because I run an Athlon XP 2000+ Yes, slow systems are more likely to get poached. So users with slow systems may be more motivated to try to prevent poaching than users with fast systems. That doesn't mean poaching is right, it does mean you're making yourself something of a tempting target. Increasing the difficulty for a poacher to _find_ a tempting target would mean other participants could be less concerned about making themselves into such a target, and just concentrate on doing the work they considered most suitable within the rules. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Re: Mersenne Digest V1 #1036
Gordon Spence wrote: Of course, as this is a *public* volunteer project, there are a lot of us, who have been in the project for a long-time (6+ years) ... which got me wondering when I started, which was: Sun, 08 Dec 1996 (at least that's when I requested my first range) 6 years, 48 days as of today. who regularly look through these for no other reason than we *want* to. Okay, I *want* to, too. But suppose there's a correlation between ability to browse _other_ people's assignment status info (you can always see your own complete assignment info), and ability to select poaching targets on the basis of other people's assignment status info? It seems worth discussing possible ideas for reducing the second even if it means reducing the first, not necessarily by the same proportion. That would deny target-selecting information to would-be poachers, right? No. If I was setting out to poach numbers - which in itself is a moot point. You don't *own* an exponent, they are after all simply numbers. I think there's a fairly well-established consensus that in the context of GIMPS/Primenet, to poach means to run a (L-L, usually) test while it is assigned by Primenet to a different GIMPS participant or something similar. There's another consensus that Primenet assignments mean something like a reservation as is used in other cooperative computational projects to avoid duplicated effort. If I was setting out to poach numbers, then I would simply setup a few 3.06 Ghz P4's and just start at the bottom of the list (smallest exponents) and let rip. So, unlike many other poachers who've declared themselves and their motives on this list or in the GIMPS Forum, you wouldn't care whether any of those exponents were, say, only 2 days from completion by the Primenet assignee? Is that correct? You wouldn't take the trouble to distinguish between an assignment that has an estimated 2 days to completion and one that had 200 days to completion? Complete an exponent every day or so. So some of them might be completed before me, so what, we then have a triple check. If someone wants to do it, you won't stop them. My proposal was not aimed at stopping that sort of blind poaching. (And I disclaimed that it would stop ALL of any type of poaching.) You are missing the point about it being useful to have triple checks. No, I'm not. I readily agree that triplechecks have some value. Perhaps you and I differ as to how that value ranks relative to values of some other things, like the value of a poach-free Primenet assignment? Make the current assignments report password-protected, then substitute a new public assignments report that omits the above four items. Do the system administrators currently need the eyes and attention of others to detect stragglers _about whom action needs to be taken_? If not, then why provide this information to poachers? If so, just give some other trusted individuals the password for the full assignments report. Are you putting yourself forward as one of the trusted individuals? Trusted not to poach (as is meant in this context) -- yes, just as would thousands of other GIMPSers be, I imagine. But I don't have time to devote to the sort of report monitoring or usage that some others do, so I wouldn't request report access or volunteer to monitor in the first place. Or how about myself, as one of the *very* exclusive club of people who have actually discovered a Mersenne prime? As long as you could be trusted by system administrators not to poach, sure. Is there any particularly _special_ relationship between being a Mersenne prime discoverer and being trusted not to poach? Unless there is some such special relationship, I imagine that thousands of non-discoverers could also be trusted not to poach. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Communication between GIMPS Forum and Mersenne mailing list (was: Poaching -- Discouragement thereof)
Mikus Grinbergs wrote: several good ideas mikus (not looking at the GIMPS forums) Mikus is just one of the contributors to the Mersenne list discussion who has not only independently reached some of the same conclusions posted slightly earlier on the GIMPS Forum, but also come up with new wrinkles that I don't recall having seen there. This is not unexpected; it makes sense that there would be both parallels and differences between two separate discussions that had been seeded with the same proposal. But consider the potential loss of ideas, as well as needless duplication of effort, because some readers of each medium will not see the discussion in the other medium. The point I wish to make is that this sort of thing will continue to happen as long as GIMPS has significant discussions in independent media without extensive (and probably not practical) cross-communication. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Re: Mersenne Digest V1 #1037
Gordon Spence wrote: [snip] I think there's a fairly well-established consensus that in the context of GIMPS/Primenet, to poach means to run a (L-L, usually) test while it is assigned by Primenet to a different GIMPS participant or something similar. There's another consensus that Primenet assignments mean something like a reservation as is used in other cooperative computational projects to avoid duplicated effort. I think we all agree on how it's supposed to work So you agree that there should be no poaching of Primenet assignments -- right? Or by it, were you not including Primenet? [snip] they are after all *just numbers*. Nobody owns them and So by they and them you _do_ mean just the numbers, without any consideration of GIMPS or Primenet - correct? anyone in the world can work on whatever they want Well, sure. But that's irrelevant to my proposal. My proposal concerned the GIMPS/Primenet system, not the whole world. without anyone's permission. Isn't there something in the current GIMPS/Primenet software along the lines of if you use our software, you agree to abide by our rules? It's actually more complicated, and it wasn't there in its present form when you used GIMPS software to discover that 2^2976221 - 1 is prime, but isn't that the gist of the current provision? [snip] If I was setting out to poach numbers, then I would simply setup a few 3.06 Ghz P4's and just start at the bottom of the list (smallest exponents) and let rip. So, unlike many other poachers who've declared themselves and their motives on this list or in the GIMPS Forum, you wouldn't care whether any of those exponents were, say, only 2 days from completion by the Primenet assignee? Is that correct? You wouldn't take the trouble to distinguish between an assignment that has an estimated 2 days to completion and one that had 200 days to completion? 1) Do you care to give us a direct answer to any of the questions I posed in the above paragraph, so that we have a clearer idea of just what you were referring to when you used that in your next sentence? 2) When you wrote bottom of the list, were you referring to a list derived from a Primenet-generated report? If anyone wanted to systematically poach, then that is a very simple approach. By that, do you mean an approach that excludes checking whether any of the Primenet assignments were very close to completion? [snip] Or how about myself, as one of the *very* exclusive club of people who have actually discovered a Mersenne prime? [[snip]] Is there any particularly _special_ relationship between being a Mersenne prime discoverer and being trusted not to poach? Unless there is some such special relationship, I imagine that thousands of non-discoverers could also be trusted not to poach. Well in actual fact, there *is* now that you come to mention it. As a Mersenne Prime discoverer I am given immediate notification of any new MP immediately it is reported, ie *before* it is verified. We are trusted to keep it quiet.because when we discovered ours we proved that we were capable of _discretion_ Only the very, very few people who had the luck to choose, or to be assigned, to L-L test a Mersenne number that happened to be prime, along with a small number of others who were directly involved in the verification process, have had the very exclusive chance to demonstrate their discretion during the post-discovery verification phase. None of the other thousands of GIMPS participants have been given even a _chance_ to demonstrate that particular, very exclusive type of discretion. Can none of the latter category be trusted not to poach? Is there any special reason _why_ discretion during the Mersenne prime verification process should have a stronger correlation with nonpoaching trustworthiness than any other demonstration of discretion has? Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Communication between GIMPS Forum and Mersenne mailing list (was: Poaching -- Discouragement thereof)
Recently I wrote: I asked this in the GIMPS forum, but haven't seen any answer there yet. So will you please point out what I overlooked? ... to which Brian Beesley replied: Sorry, I don't read the forum. It's inconvenient expensive for those of us that have pay-as-you-go dialup access; whilst I'm at work I simply don't have the time to mess with such things. There is no cable service within 40 miles of me, the nearest ADSL-enabled exchange is about the same distance away, satellite broadband is ferociously expensive still depends on a dial-up line for outbound data; the density of population round here is such that there's almost no chance of a wireless mesh working, either. I hadn't meant to suggest that Mr. Beesley should actually go look at the GIMPS Forum discussion. I meant only that I'd failed to get an answer when I posed my question elsewhere. BUT ... put that together with: Earlier in Mr. Beesley's response, he suggested something (time limits on right to an exponent assignment) that happened to have already been significantly discussed in the GIMPS Forum. In fact, if I correctly interpret the Forum discussion of that idea, it seems almost certain that that function will be added in the next Primenet server software upgrade. So Mr. Beesley and anyone else who hadn't read the Forum, whatever their reasons, _didn't yet know that a major Primenet server software change had already been discussed in the recent past and had practically been decided-upon_, because none of that Forum discussion had been reported to the Mersenne mailing list. * It looks to me like we have the potential for a major information schism between the Forum-haves and Forum-have-nots, if there continue to be major Forum discussions that go unreported to the Mersenne mailing list. * What should/can be done to ensure that those who can't/don't read the Forum are alerted to, and can contribute to, important topics discussd on the GIMPS Forum and are informed of important announcements/decisions posted on the Forum? Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
RE: Mersenne: Poaching -- Discouragement thereof
I sometimes make forecasts about number of exponents in some ranges to be checked in in particular periods. Just for fun. So the system administrators could either (a) provide you the password for the full version of the assignments report, or (b) produce a report with aggregate data that is adequate for your and similar purposes, without details for any particular assignment. I believe the number of poached exponents is very low. One's perception of low might depend on whether one were a poachee or not. I've been poached eight or nine times, and I'm tired of it. When a milistone is within reach a few last ones are sometimes poached, but it is not a structural problem imho. Recently we just had 26 or 28 (I think) such assignments poached. Is 26 or 28 a few, if you're one of those who had been steadliy working on an assignment for a year and was almost finished? Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Poaching -- Discouragement thereof
Brian J. Beesley wrote: Sure. So would eliminating the report altogether. 1) Exaggeration is unnecessary. My proposal was not that extreme. :) 2) After much discussion on the GIMPS forum, I've withdrawn my current proposal (which had been extended from the initial one I sent to this mailing list) because I have a better idea that needs mulling-over before publication. I detect righteous indignation here. So? Poaching is an official GIMPS No-No, in case you've forgotten. Might I respectfully point out that, if you stick to LL test assignments, it makes practically no difference whether you get poached, since there will eventually have to be a double-check anyway. Even though I've explained at least one other significance difference, some folks just don ... Let's put it this way: Maybe you don't give a fig for fame, but some of the rest of us do. A chance at real, honest-to-gosh mathematical fame has a value not measurable in CPU years, but poaching steals that. I don't have the hardware to climb the Top Producers list, and I don't have the financial resources to acquire much, but GIMPS was intended to give even a humble system the right to reserve Mnumbers to L-L test. Poachers steal that. I've loved mathematics ever since I learned what numbers were. I was locally considered something of a math whiz kid. My first college major was mathematics. If fate hadn't intervened, I could have been a math professor now. I'll trade you 50,000 CPU-years and the $100,000 prize in exchange for being the honest-to-gosh, no-cheating-involved discoverer of a new Mersenne prime. Have you no soul? I might also point out that there is sufficient information in the other reports (particularly the hrf3 lucas_v database files) to enable the poachers to be identified, Better read my proposal again. Its intent was to _PREVENT_ (not perfectly, but to some extent) poaching, not identify it after-the-fact. BTW, exactly which data fields in either the HRF3.TXT or LUCAS_V.TXT file provide information about currently-assigned, in-progress, incomplete assignments (which are the poachable ones)? I asked this in the GIMPS forum, but haven't seen any answer there yet. So will you please point out what I overlooked? Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Poaching -- Discouragement thereof
Here's what I've just posted in the GIMPS Forum. - - - _IF_ PrimeNet has automatic time limits on assignments, ordinarily requiring no manual intervention to expire assignments or re-assign them, then why would any GIMPS participant, other than a system administrator or a would-be poacher, need to know someone else's: (a) current iteration, (b) days-to-go, (c) days-to-expire, or (d) last date-updated? If there's no non-poaching non-administrating user's need-to-know for those items, then just stop including them in public reports. Include them only in administrative reports and private password-requiring individual reports. That would deny target-selecting information to would-be poachers, right? I'm not claiming this would stop poaching, just that it denies poachers the information they currently use (I presume) to select straggling assignments to poach. A poacher could still target the assignments of userids that had historicallly been slow or simply the ones he/she'd poached before, but then runs a greater risk of poaching a LL that's actually on-schedule, perhaps even one that's running on that user's freshly-upgraded faster system the poacher doesn't know about. Then the would-be poacher might need to consider that his/her efforts would (a) fail to help achieve milestones, and (b) more obviously interfere with GIMPS's orderly progress. (We could help the denser poachers figure this out by publishing this reasoning. :) ) - - - Indeed, couldn't my suggestion be done NOW? Make the current assignments report password-protected, then substitute a new public assignments report that omits the above four items. Do the system administrators currently need the eyes and attention of others to detect stragglers _about whom action needs to be taken_? If not, then why provide this information to poachers? If so, just give some other trusted individuals the password for the full assignments report. Does anyone see any problem with this scheme? (... other than that some individuals' curiosities would go unsatisfied?) Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Index of Mersenne Digest number vs. date - part 2b of 4
285 = 1998 January 1,2,3,4 286 =4,5 287 =6,7 288 =9 289 = 12 290 = 13,14,15 291 = 15,16 292 = 18,19 293 = 20,21 294 = 22,23,24,25 295 = 25 296 = 1980 January 1, 1998 January 25,26,27 297 = 1998 January 27,28 298 = 28,29 299 = 29,30,31 300 = 1998 January 31, February 1 301 = February 1,2,3 302 = 3,4 303 = 4,5 304 = 1996 January 10, 1998 February 5,6 305 = 1998 February 6,7,8 306 = 7 307 = 8,9 308 = 9 309 =10,11 310 =11 311 =12,13,14 312 =15 313 =15,16,17 314 =17,18,19 315 =19,20,21 316 =21 317 =22,23,24 318 =24,25 319 =25 320 =26,27,28 321 = 1998 March 1,2,3 322 = 3,4,5 323 = 5,6,7 324 = 7,8 325 = 8,9 326 = 9 327 = 10,11 328 = 12,13,14,15 329 = 15,16 330 = 1997 December 17, 1998 March 17 331 = 1998 March 18,19 332 = 19 333 = 1997 March 22, 1998 March 19,22,23,24 334 = 1998 March 24,25 335 = 26,27,28 336 = 28,29 337 = 29,30,31 338 = 30,31 339 = 1998 March 31, April 1 340 = April 2,3 341 = 3,4 342 = 5,6 343 = 7,8 344 = 9,10,11 345 = 12,13,14 346 = 14,15 347 = 15,16,17,18 348 = 19,20 349 = 21,22,23 350 = 23,24 351 = 24,25,26 352 = 26,27 353 = 1998 April 28,29,30, June 29 354 = April 30, May 1 355 = May 1,2,3 356 =5 357 =7,8,9,10 358 = 10,11,12 359 = 12 360 = 12 361 = 14,15,16 362 = 15,16 363 = 17,18 364 = 18,19 365 = 19,20,21 366 = 21,22,23 367 = 24,25 368 = 25,26,27 369 = 28 370 = 28,29 371 = 29,30,31 372 = 1998 May 31, June 1 373 = June 2,3 374 = 3 375 = 4,5 376 = 7,8,9 377 = 8 378 = 9,10 379 =11,12 380 =14,15 381 =16,17 382 =17,18,19,20 383 =19,20 384 =20,21,22 385 =21,22,23,24 386 =24,25,27 387 =28,29,30 388 = 1998 June 28,30, July 1 389 = July 1 390 = 2,3,4 391 = 4 392 = 5,6,7 393 = 7,8 394 = 8,9 395 = 9,10,11 396 =12,13 397 =14,15 398 =16,17,18 399 =19,20,21 400 =21,22 401 =23 402 =26,27 403 =28,29 404 = 1998 July 31, August 1 405 = August 2,3 406 = 4 407 = 6,7,8 408 = 9,10 409 = 11,12 410 = 13,14 411 = 16 412 = 18,19,20 413 = 20,21,22 414 = 23,24,25 415 = 25,26 416 = 26,27,29 417 = 30,31 418 = 31 419 = 1998 September 1,2 420 = 3,4 421 = 5 422 = 7,8 423 = 4,8,9 424 = 10,11,12,13 425 = 14,15 426 = 15 427 = 15,16 428 = 16,17 429 = 16,17 430 = 17 431 = 13,17,18 432 = 18,19 433 = 19,20 434 = 20,21,22 435 = 22,23 436 = 24,25,26,27 437 = 27,28 438 = 29,30 439 = 1998 September 30, October 1 440 = October 1,2,3 441 =4,5 442 =6,7,8 443 =8,9,10 444 = 11,12,13 445 = 13,14 446 = 14,15 447 = 15,16 448 = 17 449 = 18,19 450 = 20,21,22 451 = 22 452 = 22,23,24 453 = 25,26,27 454 = 26,27 455 = 27,28 456 = 28 457 = 29,30 458 = 30,31 459 = 1998 November 1,2,3 460 = 3,4 461 = 5,6,7 462 = 9.10
Mersenne: Index of Mersenne Digest number vs. date - part 2a of 4
96 = 1997 January 2,3,4 97 =6,7 98 =7,8 99 =9,10,11,12 100 = 12,13,14,15 101 = 15 102 = 16,17,18 103 = 18,19 104 = 19,20,21 105 = 21,22,23 106 = 23,24 107 = 25,26 108 = 26,27 109 = 27,28 110 = 28,29 111 = 29,30 112 = 1997 January 30,31, February 1,2 113 = February 2,3,4 114 = 5,6 115 = 6,7,8,9 116 = 9,10 117 =10 118 =11,12 119 =12,13 120 =13,14,15 121 =15,16 122 =13,16,17,18 123 =18 124 =18,19,20 125 =20,21,22,23 126 =23,24 127 =24,25 128 =25,26,27 129 = 1997 February 27,28, March 1 130 = February 18,20, March 2,3,4 131 = March 4 132 = 1996 December 4, 1997 March 4,5,6 133 = 1997 March 6,7,8,9 134 = 9,10,11 135 = 10,11,12 136 = 13,14,15,16 137 = 14,17,18 138 = 18,19,20 139 = 20,21,22 140 = 20,22,23 141 = 23,24,25 142 = 25,26 143 = 27,29 144 = 30,31 145 = 1997 April 1,2,3 146 = 3,4,5,6 147 = 6,7 148 = 8,9 149 = 10,11,12 150 = 14,15,16 151 = 16 152 = 17,18,19 153 = 20,21 154 = 22,23 155 = 24,25 156 = 25,26 157 = 27,28 158 = 28 159 = 1997 April 28,29,30, May 1 160 = May 1,3,4,5,6 161 =5,6,7 162 =8,9,10 163 = 11,12 164 = 13,14 165 = 15,16,17 166 = 19 167 = 20,21 168 = February 22, May 22,23,24 169 = May 25,26 170 = 27 171 = March 1, May 29,30 172 = 1997 March 4, June 1,2 173 = March 7, June 3,4 174 = June 4,5,6 175 = 8,9 176 =10,11,12 177 =12,13,14 178 =16,17 179 =17,18 180 =19,20 181 =22,23,24 182 =24,25 183 =26,27,28,29 184 =28 185 = 1997 June 29,30, July 1 186 = July 1,2 187 = 3,4,5 188 = 7,8 189 = 8,9 190 = 1,9,10 191 =10,11,12,13 192 =13,14 193 =15,16 194 =16,17 195 =17,18,19 196 =20,21 197 =22,23 198 =23 199 = 1980 January 10, 1997 July 24,25,26,27 200 = 1997 July 28,29 201 =29,30 202 = 1997 July 31, August 2 203 = August 4,5 204 = 5,6 205 = 7,8,10 206 = 11 207 = 12,13 208 = 15,16 209 = 18 210 = 19,20 211 = 13,20,21,22 212 = 24,25,26 213 = 26,27,28 214 = 28,29 215 = 1997 August 31, September 1,2 216 = September 2,3,4 217 = August 22, September 4,5 218 = September 5,6,7 219 = 6 220 = 7.8 221 = 8.9 222 = 9,10 223 = 10,11 224 = 11,12,13 225 = 13,14 226 = 13,14 227 = 14,15 228 = 15 229 = 16,17 230 = 17,18 231 = 18,19,20 232 = 22,23 233 = 23,24 234 = 25,26,27 235 = 26,27,28 236 = 28,29 237 = 29 238 = 1997 September 30, October 1 239 = October 1,2 240 =2,3,4 241 =5,6,7 242 =7,8 243 =9,10,11 244 = 12,13 245 = 14,15 246 = 16,17,18 247 = 18,19,20,21 248 = 21,22 249 = 22 250 = 23,24 251 = 24,25,26 252 = 26 253 = 28,29,30 254 = 1997 October 29,30,31, November 1 255 = November 2,3,4 256 = 3 257 = 4,5,6 258 = 6,7 259 = 7,8 260 = 9,10,11,12,13,14 261 =14,15,16,17,18 262 =21,22 263 =24 264 =25,26 265 =26,27 266 =26,27,28 267 =
Re: Mersenne: trial factoring
Norbert.Pfannerer wrote: Why is the range 219. not included for TF? Before Christmas, for several weeks it looked like someone was systematically working through 2191 (and maybe earlier -- I didn't check) to 2194, judging by weekly changes to factors.zip and nofactor.zip. My guess is that someone has reserved the 219x range by e-mail with George W. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Index of Mersenne Digest number vs. date - part 2 of 2
Did anyone on the Mersenne mailing list receive either of the two postings I sent to the list yesterday with the subjects Mersenne: Index of Mersenne Digest number vs. date and Mersenne: Index of Mersenne Digest number vs. date - part 2 of 2? If not, then apparently I've discovered yet another of my ISP's e-mail flaws. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Index of Mersenne Digest number vs. date - part 0 of 4
Thanks to those who responded to my inquiry about my 2-part posting. Since no one seems to have received either of those parts, I'm going to try again with a 4-part posting. If that doesn't work, ... hmmm ... I'll try reformatting the index text. So stay tuned. My next posting to [EMAIL PROTECTED] will have the index's introductory heading and number vs. date for the first 95 Digests (through the end of 1996). Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Index of Mersenne Digest number vs. date - part 3 of 4
490 = 1999 January 3,4 491 =4,5,6 492 =6,7 493 =7,8 494 =9 495 = 10,11,12 496 = 12,13 497 = 12,14,15,16 498 = 16,17,18 499 = 19,20 500 = 24 501 = 26,27 502 = 28,29 503 = 29,30 504 = 1999 January 31, February 1,2 505 = February 2,3 506 = 4,5,6 507 = 7,8,9 508 = 9,10 509 =11,12,13,14 510 =15,16,17 511 =16,17,18 512 =19,20,21 513 =22,23,24,25 514 =25,26 515 =26,27,28 516 =28 517 = 1999 February 28, March 1,2 518 = March 2 519 = 2,3 520 = 2,3 521 = 4,5,6 522 = 7 523 = 2,7 524 = February 8, March 7,8 525 = March 8,9 526 = 3,8,9,10 527 = 10 528 = 10,11 529 = February 12, March 8,11,12,13 530 = March 13 531 = 14,15 532 = 15,16 533 = 16,17 534 = 17 535 = 18,19 536 = 19,20,22 537 = 22,23,25 538 = 25,26,30 539 = 1999 March 30,31, April 1 540 = April 1,2,3 541 = 3,4,5,6 542 = 6,7,8 543 = 8,9,10,11 544 = 11,12 545 = 12,13,14,15 546 = 15,16,17 547 = 17,18,19 548 = 20,21,22 549 = 23,24,27 550 = 27,29 551 = 1999 April 30, May 3 552 = May 3,4,5 553 =4,5,6,7,8 554 =8,9,10 555 = 10,11,12 556 = 11,13,14 557 = 14,15 558 = 14,15,16 559 = 16 560 = 16,17 561 = 18 562 = 18,19,20,21 563 = 21,23,25 564 = 25,27 565 = 28,29,30,31 566 = 1999 May 31, June 1,2 567 = June 2,3,4 568 = 4,5,6 569 = 5,6,7 570 = 7,8 571 = 8,9 572 = 9,10,11 573 =10,11 574 =11,12 575 =12,13,14 576 =13,14 577 =14 578 =14 579 =14,15 580 =15,16 581 =15,16 582 =16,17 583 =17,18 584 =18,19 585 =19,20,21 586 =18,20,21,22 587 =22,23,24 588 =24,25,26 589 =26,27 590 =27,28,29 591 = 1999 June 28,29,30, July 1 592 = July 1,2,3 593 = 3,4,5,6 594 = 5,6,7,8 595 = 8,9 596 = 9,10,11 597 =12,13,14 598 =14,15,16 599 =16,17,18 600 =18,19 601 =20,21,22 602 =22,23 603 =23,24,25 604 =25,26 605 =26,27,28,29 606 =28,29 607 =29,30,31 608 = 1999 July 31, August 1,2 609 = August 3,4 610 = 4,5 611 = 5,6,7 612 = 7,8,9,10 613 = 10,11,12 614 = 12,13,14,15 615 = 15,16,17 616 = 17,18,19,20 617 = 19,20,21 618 = 21,22,23,24 619 = 24,25,26,28 620 = 29,31 621 = 1999 August 31, September 1,2,3 622 = September 4,5,7 623 = 7,8,9 624 = 9,10,11,12 625 = 10,11,12,13,14 626 = 14,15 627 = 15,16,17 628 = 17,18,19 629 = 19,20,21 630 = 20,21,22,23 631 = 22,23,24 632 = 24,25 633 = 19,25,26 634 = 26,27,28 635 = 1999 September 28,29,30, October 1 636 = September 30, October 1,2,3 637 = September 22, October 3,4,5 638 = October 5,6,7 639 =7,8,9,10 640 = 10,11 641 = 11,12 642 = 12,13 643 = 13,14,15 644 = 14,15 645 = 15,16 646 = 16,17 647 = 17,18,19 648 = 19,20 649 = 20,21 650 = 21,22 651 = 21,22,23,24 652 = 24,25,26 653 = 26,27,28 654 = 1999 October 28,29, November 1 655 = November 1,2,3 656 = 3,4,5,6 657 = 6,7,8 658 = 8,9,10,11 659 =12 660 =13,16 661 =16,19 662 =
Mersenne: Index of Mersenne Digest number vs. date - part 1 of 4
Mersenne Mailing List -- Digest Number vs. Date Index The date range listed for a digest issue is that of the messages posted within that issue, and does not necessarily include the publication date in the issue's header. There are anomalous dates in some message headers. Example: In digest Volume 1, Number 171, published on 1 June 1997, one of the posted messages is dated Sat, 1 Mar 1997. Was that message actually posted three months after it was sent, or was it mis-dated by the sender's software for one reason or another? Rather than try to determine what happened, I simply index all message header dates as they appeared in the digests. On each line of this index, the message dates within each digest are listed in chronological order, but the messages were not necessarily posted in that order in the digest. Also, more than one message with a given date may have appeared in a digest issue, but this index lists each unique date only once for each digest issue. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Before first digest issue = 1996 April 21 through 1996 July 1 - - - Volume 1, Number 1 = 1996 July 7,8,9 2 = 9 3 = 9 4 = 9,10,11 5 =11 6 =11,12 7 =12,13 8 =14,15 9 =16 10 =18,20 11 =22 12 =25,26,27 13 = 1996 July 30,31, August 1 14 = August 2 15 = 4,5 16 (Number 16 contained only spam and responses to it.) 17 = 8,9,10 18 = 14,15 19 = 15 20 = 15 21 = 15,16,17 22 = 17,18 23 = 18,19 24 = 20,21 25 = 22,23,24 26 = 26 27 = 1996 September 1 28 = 3,4 29 = 3,4,5 30 = 5,6,7,8 31 = 9,10 32 = 10,11 33 = 12,13 34 = 16 35 = 18 36 = 19,20,21 37 = 22,23,24 38 = 24,25 39 = 27,28 40 = 29,30 41 = 1996 October 1,2 42 =3,4,5 43 =6 44 =8,9 45 = 10 46 = 14,15 47 = 15 48 = 17 49 = 24,26 50 = 28,29 51 = 29,30 52 = 1996 October 31, November 1,2 53 = November 3 54 = 5 55 = 7,8,9 56 =10,11,12 57 =12,13 58 =14 59 =17 60 =19,20 61 =21 62 (Number 62 contained only repetitions of earlier messages.) 63 =21 64 =21,23 65 =24,25,26 66 =26,27 67 =28,29 68 =29,30 69 =28,29,30 70 = 1996 November 30, December 1 71 = December 1,2 72 = 2,3 73 = November 29, December 1,3 74 = November 29, December 3,4 75 = December 3,4 76 = December 5 77 = 5 78 = 5,6 79 = 6,7 80 = 7,8 81 = 8,9 82 = 9,10 83 = 9,10,11,12 84 =11 85 =12,13 86 =13,14 87 =15,16 88 =17,18,19 89 =18,19 90 =19,20 91 =20,21 92 =22,23 93 =24,25 94 =26,28 95 = 1996 December 29,30 _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Index of Mersenne Digest number vs. date - part 4 of 4
807 = 2001 January 3,5,7 808 =7,8,9,14 809 = 15,16,17 810 = 19,20,22 811 = 22,23,27 812 = 2001 February 2,3,4 813 = 3,4,5,6 814 = 6,7,8 815 = 8,9,10,11 816 =11,12,13 817 =13,15 818 =15,17 819 =17,18,19 820 =20,21,22,23 821 =23,24,25,27 822 = 2001 February 27,28, March 1,3 823 = March 3,4,5,6 824 = 6,7,8,9 825 = 9,10 826 = 2001 March 10,11, 2002 March 11 827 = 2001 March 11,12 828 = 12,13 829 = 13,14,15 830 = 16,18 831 = 18,20 832 = 20,21,22 833 = 25,26,27 834 = 27,28,29 835 = 2001 March 30,31, April 1,2 836 = April 2,3,4 837 = 5,6,7 838 = 7 839 = 10,11,12 840 = 15,16,17 841 = 17,18,19 842 = 20,22 843 = 22,23,24,25 844 = 25,26,28 845 = 2001 April 28,29, May 1 846 = March 31, May 2,6 847 = May 6,7,8,9 848 =9,12 849 = 12,13 850 = 12,13,14 851 = 14 852 = 13,14,15,16 853 = 15,16 854 = 1998 January 29, 2001 May 16,17,18,19 855 = 2001 May 19,20,21 856 = 21,22,25 857 = 25 858 = 2001 May 31, June 1,2 859 = June 3,4,5,8 860 = 8,9,10,11 861 =12,13,17 862 = 2001 June 17,18,19, 2002 June 19 863 = 2001 June 20,21 864 =22,23,25 865 =26,27,30 866 = 2001 July 1,3 867 = 5,10 868 =17 869 =18,20 870 =21,22,23 871 =23,24,25,26 872 =26,27,28 873 =28,29,30 874 = 2001 July 30,31, August 1,7 875 = August 8,9,12 876 = 14,16 877 = 17,18,22 878 = 22,29 879 = 2001 August 30,31, September 3 880 = September 3,4,5,6 881 = 7,8,9 882 = 9,10 883 =10,11,12,13 884 =17 885 =21,22,23 886 =22,23,25,26,27 887 =27,28,30 888 = 2001 September 30, October 1,2 889 = October 3,4,5 890 =6,7,9 891 =9,10,12 892 = 12,14 893 = 14,15,16,17 894 = 19,20,21 895 = 21,22,23,25 896 = 25,26,27 897 = 27,28,29 898 = 2001 October 29,30,31, November 1 899 = October 31, November 1,2 900 = November 2,3,4 901 = 4,5 902 = 5,6,7 903 = 8,9,10 904 =10,11,12 905 =12,13,14 906 =14,15,17 907 =17,18,19 908 =19,20,21 909 =22,23,24 910 =24,25,26,27,28 911 =27,28,29 912 = 2001 November 30, December 1,2 913 = December 2,3,4 914 = 4,5 915 = 5,6 916 = 6,7 917 = 7,9,10 918 =10,11,12 919 =12,13,14 920 =14,15,16 921 =17,20 922 =20,21,22 923 =22,23,24 924 =25,26,27 925 =28,30 926 = 2001 December 31, 2002 January 1,5 927 = 2002 January 6,8,9 928 = 10,11,12,13 929 = 13,14 930 = 14,15,16 931 = 16,17,19 932 = 23,26 933 = 26,27,28,30 934 = 2002 January 31, February 1,2 935 = February 2,3,4 936 = 4,5,7 937 = 7,8,9,11 938 =11,12,13 939 =13,14,15 940 =15,16,17 941 =17,18,26 942 =26,27,28 943 = 2002 February 28, March 1,2,3 944 = March 2,4,5 945 = 1,7,8,12 946 = 13,15 947 = 16,17,18 948 = 18,19,20 949 = 20,21,22 950 = 22,23,24 951 = 24,25,26,27 952 = 27,28,29,30 953 = 2002 March 30, April 1 954 = April 2,8 955 = 8,13 956 = 13,14,15,16 957 = 16,21 958 = 22,23,24 959 = 25,26,27 960 = 2002 April 27,28, May 1 961 = May 4,8 962 =9,10 963 = 21,22,24 964 = 24,25,26 965 = 26,27,28,30 966 = 2002 May 31, June 3 967 = June 3,5,6 968 = 6,7,8,9 969 = 9,10,11 970 =12,13,14,15 971 =
Re: Mersenne: Dissed again
Let's not be _too_ eager to emulate SETIhome's popularity and user-friendliness -- http://www.zdnet.com.au/newstech/security/story/0,224985,20269509,00 .htm Cheats wreak havoc on SETIhome: participants SETIhome administrators are allegedly ignoring claims that the project is being sabotaged by miscreants who are threatening to derail its reputation and that of many valuable Internet-based distributed computing projects. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Hyper-threading
Daran wrote: Could this feature of forthcoming Intel processors be used to do trial factorisation without adversely impacting upon a simultaneous LL? This was discussed in March, starting in Digest Number 945. Short answer: no. Medium answer: If an application were not already optimized for maximum pipelining, hyperthreading would allow another application to use unused pipeline capacity. But Prime95 is already so optimized. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Re: http://mersenne.org/primenet/ - the world test
Try accessing the Primenet status summary report directly at http://mersenne.org/primenet/summary.txt Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Primenet status summary report Exponent Range
Speaking of the Primenet status summary report at http://mersenne.org/primenet/summary.txt , someone needs to allocate another couple of columns to Exponent Range if it's going to be reporting on exponents over . Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Order of TF and P-1
Daran wrote: When the TF limits were originally decided, it was assumed that a sucessful TF would save 1.03 or 2.03 LLs. I can't remember whether George has ever said whether they have been lowered to take the P-1 step into account. to which Steve Harris replied: I don't think the TF limits were ever lowered; it seems they may have been raised, as I have gotten several 8.3M DC exponents which first had to be factored from 63 to 64 Whether it's raising or lowering depends on which way we're looking at the TF limits: (a) the upper limit on power-of-two to which Prime95 TFs a given Mnumber, or (b) the lower or upper limits on Mnumber exponent for which Prime95 TFs to a given power-of-two. Recent Prime95 changes raised (a) for some Mnumber exponents, and thus lowered (b) for some powers of two. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Re: Any Comments On The Significance For GIMPS Of The New Kanpur Primality Algorithm?
Referring to the work, recently in the news, of Mssrs. Manindra Agrawal, Neeraj Kayal and Nitin Saxena. This is the same work to which Jens Pfeiffer referred in a Wed, 7 Aug 2002 17:37:27 +0200 posting with subject Mersenne: find primes in polynomial time? Parallelling Paul Leyland's response then: The Lucas-Lehmer primality testing algorithm in GIMPS software is restricted to only a very special form of prime, 2^p - 1. In contrast, the new AKS algorithm applies to _all_ primes. On primes of the form 2^p -1, the Lucas-Lehmer test in GIMPS software is *MUCH, MUCH* faster than the new AKS algorithm. However, the new AKS algorithm will work on other forms of primes to which the Lucas-Lehmer test is not applicable. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Re: P-1 limits
by P-1 factoring are more complicated than the simple ( number of squarings = exponent - 2 ) one used for L-L test cost because P-1 factoring operation is inherently more complex than L-L testing. Furthermore, Prime95's P-1 operation differs according to how much memory is available -- it will perform a tradeoff of more memory use for faster operation when possible, and this tradeoff is taken into account by the limit-choosing algorithm. Outputs from the algorithm: optimal B1 and B2, total number of squarings for a P-1 run using that combination, and probability of finding a factor using that combination. Though both P-1 and L-L costs are dominated by the number of FFT squarings required, the nonlinear dependence of P-1 squaring cost on exponent and bounds means that there can be a discontinuity in chosen bounds when the exponent passes from the range for one FFT size to that for another FFT size -- the balance point shifts differently for P-1 and L-L. Richard Woods P.S. When replying to this message, check that your reply's To: field contains [EMAIL PROTECTED]. My ISP's mailer has been mangling the Reply-To: field of messages I send. _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Re: P-1 limits
Recently I wrote: Well, if a Mnumber has been trial-factored to 2^63 (with no success, or else we wouldn't be considering P-1 now), then P-1 will not find any factors of 63 bits or less. Tom Cage has reminded me that he and others have found factors with P-1 that are smaller than the posted bit limit in the current nofactor.cmp at http://www.mersenne.org/gimps. AFAIK, the number of such findings is too small to make any significant difference in the probabilities calculated by the P-1 limit-choosing algorithm, but it would be well to remember that it can still happen. Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Re: Mersenne: Re: P-1 limits
Brian J. Beesley wrote: Shouldn't make much difference Of course. By Daran's example (quoted here in reverse order), 7714127B1=4, B2=65 7786567B1=4, B2=64 one can easily see that the effect of differing FFT sizes (384K for exponent 7714127, 448K for 7786567) on choice of P-1 limits was small in this case. But Daran's question was about why the the _smaller_ exponent should get the _higher_ B2 limit at all. My point was that that was a result of the FFT difference on the algorithm's output. P-1 computation rate is directly linked to LL testing computation rate, since the same crossover points are used and the main cost of both is FFT multiplication. True, but the question here was about the choice of B1 and B2 limits. I think the most likely cause is that the P-1 limits depend on the relative speed of P-1/LL computation speed and trial factoring speed. No. Neither trial factoring speed nor its ratio to P-1/LL computation speed enters into the P-1 limit-choosing algorithm. If RollingAverage was slightly different, the relative computation rate might be guessed to be different enough for the P-1 limits to vary slightly. No. The P-1 limit-choosing algorithm does not consider RollingAverage either. Richard Woods P.S. When replying to this message, check that your reply's To: field contains [EMAIL PROTECTED]. My ISP's mail handler has been mangling the Reply-To: field of messages I send. _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Mersenne-digest numbering
Ahem ... Will the Mersenne-digest following Volume 01 : Number 999 be Volume 02 : Number 001? Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
Mersenne: Mersenne-digest numbering
How will the Mersenne digest following Volume 01 : Number 999 Richard Woods _ Unsubscribe list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
