mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Data (https://www.mersenneforum.org/forumdisplay.php?f=21)
-   -   (Preying for) World Record P-1! (https://www.mersenneforum.org/showthread.php?t=27406)

chalsall 2021-12-17 21:52

[QUOTE=lisanderke;595538]If someone else prefers to search and read through decades worth of posting in this particular forum, be my guest.[/QUOTE]

What you don't seem to appreciate is this is how claims of prior art works. It can be a bit onerous (which is exactly the plan).

Have you ever tried to claim a patent on an original idea? Teams of people are involved just reading tonnes of language. And even then you need to reserve a few million dollars to defend the claim.

Please forgive me for this, but I sometimes [URL="https://www.youtube.com/watch?v=3qqE_WmagjY"]try to be funny[/URL]. Or, at least, be amusing...

SethTro 2021-12-17 23:39

[QUOTE=lisanderke;595474]Be sure to keep this thread in the loop! Let us know when some feature/functionality is added/turned on so that we're able to exchange P-1 efforts specifically for this project. Thank you for the work done so far![/QUOTE]

I got thirty minutes to work on this today

* I added all the files I have (which includes backups that should be filtered)
* You can now download the files

Next up is sorting and filtering the existing files.

techn1ciaN 2021-12-18 00:50

[QUOTE=Batalov;595526]For GIMPS, P-1 is simply a crutch, a tool; a fast an relevant way to remove the chaff. For this project, this is suddenly pretending to be the proper play important on its own right.[/QUOTE]

Do you have a similar opinion about petrw1's sub-2,000 project?

Some GIMPS members are more interested in factoring than in the "proper" Mersenne prime search. These members aren't declaring factoring to be categorically more important; they just personally prefer to do this type of work. Inasmuch as work done [I]is[/I] a matter of personal preference, GIMPS isn't entitled to have anyone use their cycles in any particular way.

Obviously, the grand majority of GIMPS throughput is "relevant" TF, P-1, and primality testing. I don't think this makes it a sin for small groups to occasionally coordinate using the database for something else that they find interesting. (If nothing else, PRP-CF is an official work type and PrimeNet will happily hand out factoring AIDs for exponents far below the DC and PRP wavefronts, so it would be hard to say that GIMPS has any firm position against "useless" factoring.)

[QUOTE=chalsall;595533]I /think/ you might have missed Batalov's point.[/QUOTE]

In my opinion, Batalov could have given Mr. Viaene the basic respect of choosing to make this point directly instead of with backhanded sarcasm, which is inherently open to misinterpretation. Sarcasm almost never conveys well on an Internet forum, and some individuals (such as myself) can have significant difficulty with it even when full context is available.

Batalov 2021-12-18 02:42

As one member of the forum used to say, "You can lead a horse to water but you can't make him drink."
[SPOILER]It is of course the easiest to not even lead them to the water. What would be the use?[/SPOILER]

Far too many people believe that mersenneforum is the sole extension to mersenne.org. It isn't - solely; it grew up to be a lot of little clubs. But newcomers come and they don't want to read/search through the forum. "They have never played the violin but they don't see what is the big deal. Of course they can. And they will break a world record while doing that, too".

No, I didn't suggest to read just the forum. Read mathworld, read OEIS, read Pomerance's book, for ${Deity}'s sake, read [I]something[/I], anything at all. The danger of math is that people keep thinking that there is a king's road though it, for centuries they have. "I have thought of something that no one ever thought before." No.
"So, I have to make some errors so that I could regret them?" -- "You don't have to make errors to regret them. You have already done plenty. You simply don't know what they are." (paraphrasing "Big Kahuna")

And if you did just a little of that, then you would easily realize that people who do [I]anything [/I]below 57,000,000 aren't doing anything at all for GIMPS - altogether. They are on their own. They are searching for additional factors for the numbers that have been eliminated 20+ years ago. Some people realize that what they are trying to do is identical to helping the Cunnigham project. But not really. Most of them have no idea what I just said. That's sad.

Sarcasm? What sarcasm? Reality, my friend.
:answers:
[QUOTE=techn1ciaN;595557]Do you have a similar opinion about petrw1's sub-2,000 project? [/QUOTE]
Never heard of it.
Wayne probably knows what he is doing (he's been around longer than 20 random accounts combined) and I haven't heard him using too much hyperbole.

techn1ciaN 2021-12-18 04:02

[QUOTE=Batalov;595563]...you would easily realize that people who do [I]anything [/I]below 57,000,000 aren't doing anything at all for GIMPS - altogether.[/QUOTE]

I don't disagree. I just don't understand what the point of complaining about that is supposed to be.

In the first place, GIMPS is an interest project, done for fun. This isn't to say it can't have practical impacts — Prime95's immense popularity for hardware validation, for one example — but its largest / "primary" appeal is that you can set your computer doing interesting work, be part of a larger community doing the same, and hopefully learn something in the process. So, is it so bad that someone "[isn't] doing anything at all for GIMPS" when GIMPS, depending upon one's perspective, isn't doing much itself? Or to frame that more positively, levity is important; it's a fun project, so one should try to have fun with it.

In the second place, if you try to admonish someone for what they choose to run on their own computer, then they will turn off that computer and leave sooner than they will start running what you prefer instead.

[QUOTE=Batalov;595563]Never heard of it.[/QUOTE]

He is [URL="https://www.mersenneforum.org/showthread.php?t=22476"]coordinating an effort[/URL] to reach less than 20 million unfactored exponents in the mersenne.org database. He hopes to accomplish this by sub-dividing the database into many small "ranges," with an equal factoring goal for each, and then focusing on the ranges where PrimeNet's standard factoring protocols have not gotten (or probably will not get) to his goal automatically. This results in lots of work on small exponents (< 30 M, sometimes < 10 M) that were already LLed and DCed years or decades ago — in other words, something probably thoroughly useless from your point of view. The project had its first inkling in late 2017 and he projects that it might wrap up within 2022.

At the end of the day I'll simply defer to George Woltman — [URL="https://www.mersenneforum.org/showpost.php?p=577077&postcount=13"]I like factors[/URL] :smile:

Batalov 2021-12-18 06:08

First of all, you are preaching to the choir. I have seen more factors* than you will in your life. So if you count anyone who likes factors more than me, there would be very few. After all, if you cared, then you would have first found that I removed >2 million candidates from [I]this [/I]database that you keep talking about - singlehandedly - way before it became fashionable. But you can't, can you? You, too, are a writer, not a reader. Cool. Good for you. Carry on entering a community and giving lectures about how it should be.

Also, you must be hearing something that I'd never said. You are arguing with something that is inside your head. Deal with it. Where did I say who should compute something or should not compute something? I only said that those who don't think while doing it will repeat trivial mistakes that are obvious to most (or at least some) others, that's all. I warn against the "this must be easy, I just made up something and it looks true to me" attitude. This attitude also [URL="https://www.youtube.com/watch?v=rlC_z56rTik&t=153s"]leads to deep disappointments[/URL], and if you follow someone to the water you could avoid it. There is that mysterious Poisson that [I]xilman [/I]keeps bringing up (wth?), there is even some better statistics, confidence intervals, and more and more.

While in college, we had a popular (self-deprecating) anecdote.
"Scientists put a chimp in a cage, placed a tall plastic tree with a banana on top, and a stick. The chimp jumps, the chimp shakes the tree, but the banana doesn't fall. The scientists look at the chimp encouragingly and whisper to themselves, 'well, c'mon, think! think!' Then, the chimp pensively stops for a moment, then takes that stick and after a few tries, gets the banana.
Now the Scientists put a sophomore <from our college> in the same cage. The student shakes the tree vigorously for a few hours. The scientists look at him though the glass and gesture 'Think! Think!' The student yells: What is there to think about?! [B]I must shake the tree[/B]!!"

Every year we see this type of conversations, here:
[I]Person A[/I]: I see what you are trying to do and you are currently doing this wrong.
[I]Person B[/I]: how dare you tell me what to do and what not to do? I will leave this project and you will be very [B]very[/B] sorry! That will show you!

Remarkably constant.
__________
*and primes? Primes, too :rolleyes:

techn1ciaN 2021-12-18 14:41

[QUOTE=Batalov;595570]I have seen more factors* than you will in your life ... I removed >2 million candidates from [I]this [/I]database that you keep talking about - singlehandedly - way before it became fashionable.[/QUOTE]

Congratulations on having joined the project when lots of easy breadth-first TF was still available — not all of us can be so lucky. There's no need to diminish those who are doing the best they can with the factoring assignments that are currently available.

[QUOTE=Batalov;595570]You, too, are a writer, not a reader ... Carry on entering a community and giving lectures about how it should be.[/QUOTE]

I would never make any categorical statement. I am purely voicing my opinion and anyone is free to ignore it, just as they are free to ignore Mr. Viaene. If he truly is blundering headfirst into a losing endeavor (and he does not learn enough to rectify this), then he will either fail to recruit any participants or his participants will fail to see any results, and his project will fizzle out within a few months organically.

FWIW, I've been a GIMPS member for five years (although I'm sure that still means nothing to you), and I spent about a year lurking on the forum without an account before opening one. How recently someone joined here doesn't necessarily mean anything.

LaurV 2021-12-18 15:44

I can pee higher than all three of you!

Dr Sardonicus 2021-12-18 15:49

[QUOTE=techn1ciaN;595583]<snip>
I would never make any categorical statement.
<snip>[/QUOTE]
[quote][b]ALL:[/b] What, never?

[b]CAPTAIN:[/b] No, never!

[b]ALL:[/b] What, never?

[b]CAPTAIN:[/b] Well, hardly ever![/quote] -- [i]HMS Pinafore[/i] by Gilbert and Sullivan

Uncwilly 2021-12-18 19:22

I like factors. I have sent many actual CPU wall clock years only doing factors (out in the 332M range). I have also done mainly DC's for a while now. And I am doing PRP-CF. Etc. I think each of us can do work in multiple parts of the project, what ever excites us at the time. As is there is the primary goal (find the next MP, and maybe all of them). The secondary goal (confirm that we did not miss an MP). Then there are the tertiary goals (find factors for Mersenne numbers.) And then there are the non-Mersenne related projects hosted on the MF. There is no need for any one to :poop: on any of the other goals, projects, nor individuals involved. And unless your name is Lucas, Robinson, Slowinski, or Woltman, there is always someone else that has bigger bragging rights.

Many newbies in most forums go through a similar arc of enthusiasm/knowledge/hubris/etc. I think that several of the new folks in the last year have been fine additions to the community. The older members might get jaded in their views and need to take a breath and count to 127 before posting a tirade.

Batalov 2021-12-18 21:25

[QUOTE=techn1ciaN;595583]If he truly is blundering headfirst into a losing endeavor (and he does not learn enough to rectify this), then he will either fail to recruit any participants or his participants will fail to see any results, and his project will fizzle out within a few months organically.[/QUOTE]
If?

[QUOTE=petrw1;464177]Thinking out loud about getting under 20M unfactored exponents

Breaking it down I'm thinking ... <good!>

So I did some Excel ciphering looking at: <very good!>
- how many more factors are required in each range
- how many exponents need to be TF'd at the current bit level to get there (could require several bit levels to complete)
- how many GhzDays each assignment would take.[/QUOTE]
Good project, good goals. Steady and useful work. No talk of "world records" - but likes and makes others like factors. This is where world records really will come - as a byproduct, with a lot of diligent work.

[QUOTE=lisanderke;595271]World Record P-1!

I'd like to organize a thread to find World Record Factors with P-1 on small exponents.
I'll now refer to this project as WR P-1 (at the risk of it sounding too pretentious, perhaps...)
Similar to Petrw1's <2k (sub twok, <twok...)[/quote]
Pretentious? You said it, man.

Like I said, he is simply trying to reinvent Cunnigham's project (and only 2- portion of it and only with P-1), and had not put any thought into it. No resource planning. No proof of concept. No knowledge of what was done before. "Zimmermann? Who the hell is he?"
Worse of all - this is how one can successfully [I]slow down[/I] Wayne's project and the whole project that he says his is a sub-sub-project of.

"I want to be a ... lion tamer!"

P.S. The in-joke. Can you guess who was they last person who changed the thread's title (which is our old little game)? You'd be surprised and it actually makes so much sense once you find out :missingteeth:


All times are UTC. The time now is 17:53.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.