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)

techn1ciaN 2021-12-17 15:04

[QUOTE=lisanderke;595466]Question: why does my composite factor show up in the results history of this exponent? Shouldn't Primenet already have been aware of this composite factor?[/QUOTE]

You're experiencing this problem: [URL]https://www.mersenneforum.org/showthread.php?t=27346[/URL]

[QUOTE=Batalov;595456]...reading this thread is full of surprises, like "oh, he also doesn't know that", "oh, and that too", but surely the attitude is "what is that qualifying for regional/divisional college competition? ...No, I will go straight for the world record"...[/QUOTE]

I'll say that I admire Mr. Viaene's tenacity in being willing to jump right in, get something rolling even if it's ambitious, and learn as he goes. Given the extent of advice already offered, I don't think much would have been gained from him cautiously poking the topic for months and making absolutely sure he had a completely firm handle on it before starting a thread. At the end of the day GIMPS searches for prime numbers for fun; as Mr. Woltman points out, the absolute worst "failure" case is finding a few interesting factors and "wasting" (open to interpretation) a few cycles.

VBCurtis 2021-12-17 16:40

[QUOTE=bur;595460]I'm not sure I got that right, but is the plan to run consecutive P-1 with increasing bounds on the same exponent? Isn't that a waste of the previous runs? Or does round apply to ranges of larger exponents?[/QUOTE]

A Stage-1 save file can be resumed from that B1 to a higher B1 without losing any work. So, maybe?

Also, one could run stage 1 on small-memory machines to rather big bounds, and post the work to a repository where someone with big-memory could run stage 2 to quite high bounds.

lisanderke 2021-12-17 19:49

[QUOTE=James Heinrich;595491]Why wouldn't it show up in the exponent history? Maybe you didn't find any new factors, but we still need to record the space that you searched and what you found in it.[/QUOTE]
:picard: Should've realized, that makes sense! :grin:

[QUOTE=techn1ciaN;595495]
I'll say that I admire Mr. Viaene's tenacity in being willing to jump right in, get something rolling even if it's ambitious, and learn as he goes. Given the extent of advice already offered, I don't think much would have been gained from him cautiously poking the topic for months and making absolutely sure he had a completely firm handle on it before starting a thread. At the end of the day GIMPS searches for prime numbers for fun; as Mr. Woltman points out, the absolute worst "failure" case is finding a few interesting factors and "wasting" (open to interpretation) a few cycles.[/QUOTE]
Thanks for the kind words! Another reason for jumping in before doing lots more research by myself was that I knew there had been many before me trying (and perhaps succeeding in doing) the same thing! I now know it was the right decision to dive straight in, instead of trying to figure out a lot of things on my own. I have previous posts in this thread to prove that lots of people are keen to help others understand these things better. :smile: (I did count on others helping me from the start, as I pointed out in my original post:)
[QUOTE]With all of that out of the way, I need help devising up proper bounds for all of these ranges, exponents... As I've said before, the math escapes me, but hopefully I can learn from this project![/QUOTE]

Batalov 2021-12-17 20:03

[QUOTE=lisanderke;595475]I'm afraid I must not have clarified enough that this project would limit itself to Mersenne numbers. With that in mind,
...The 'bit-size' to beat is 173,22. [/QUOTE]
Nope, higher.
58 digits prime factor is a factor of a [URL="https://mathworld.wolfram.com/MersenneNumber.html"]Mersenne number[/URL] and was found with P-1 (in 2005, too). That's [URL="http://factordb.com/index.php?id=1100000000012855105"]189.80 bits[/URL].

Also, keep in mind that composite factors were and should remain ineligible. One can trivially find insanely large composite factors with P-1. One [I]"excellent" [/I]way to find them is indeed not include already known factors while queueing a "World Record" job.
___

This little project is somewhat reminiscent of "[URL="https://en.wikipedia.org/wiki/Rosencrantz_and_Guildenstern"]Rosencrantz and Guildenstern[/URL] are dead". With good execution, it might be excellent. With a sophomoric implementation, it might be mediocre.
How is it reminiscent? Simple. We take a popular play and pick totally decorative characters and make them the center characters for "your" play. 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. "Let's create a World Record for a general method, but only when it is applied to a tiny fraction of its possible uses." = "Let's set up and start fighting for a World Record in long jumps with both hands tied behind the jumpers' backs." Cute, but only as a joke. My 2 cents.

I understand George's interest, sure. It is generally called "riding out a hype", that is -- it doesn't matter why hype evolved. As long as it can help the main project, that's great. That is a valid reason, I have nothing against that. And also, a great reason to revisit and debug a rarely visited branch of the code. :rolleyes:

lisanderke 2021-12-17 20:22

[QUOTE=Batalov;595526]Nope, higher.
58 digits prime factor is a factor of a [URL="https://mathworld.wolfram.com/MersenneNumber.html"]Mersenne number[/URL] and was found with P-1 (in 2005, too). That's 189.80 bits.

Also, keep in mind that composite factors were and should remain ineligible. One can trivially find insanely large composite factors with P-1. One [I]"excellent" [/I]way to find them is indeed not include already known factors while queueing a "World Record" job.[/QUOTE]


Ah, Mersenne Numbers with prime exponents then, those that are listed one mersenne.ca and mersenne.org. That is what I meant. Apologies for assuming this was implied.

Also, you may be confusing the main topic of this thread with another conversation that was started between George Woltman and I about some weird behavior of Prime95 I had been having with v30.8b5. I am aware that composite factors found with P-1 are not prime factors, and, consequently, not eligible for taking the '[B]World Record Title for P-1 Prime Factor Found on Mersenne Number With Prime Exponent.[/B]" Sigh. See below for the reason I posted a composite factor result:

[QUOTE=Prime95;595403]-snip- Here's some good news. You can check if your save files are good. Create a worktodo.txt entry with a slightly larger B1 and [B]no known factors[/B]. Let 30.7b9 run that and see if it finds some or all of the known factors. I think you'll find you're in good shape.

Example: Pminus1=N/A,1,2,5003,-1,200000999,200000999[/QUOTE]

[QUOTE=lisanderke;595406]-snip- I used the following worktodo entry as suggested: Pminus1=N/A,1,2,5903,-1,200000999,200000999
Copied save file for 5903 from 30.8b5 to 30.7b9. And Prime95 turned it into a 'bad' save file.[/QUOTE]
[QUOTE=Prime95;595416]My bad. 30.7 cannot read a 30.8 save file. Try your save file and "Pminus1=N/A,1,2,5903,-1,200000999,200000999" with 30.8.[/QUOTE]

[QUOTE=lisanderke;595429]Oh, that did finish the P-1, but it didn't report a factor. Could it be that the save file already knows the 8 factors?[/QUOTE]

[QUOTE=Prime95;595441]That's not good. The save file knows nothing about the known factors.

Do the save files prior to the 100% reporting find known factors when the B1 bound is increased by 999?[/QUOTE]
The rest is history :)


EDIT: In retrospect, I could have prompted @Batalov to simply re-read parts of this thread to make them better understand why I posted a composite factor result in this particular thread. Though this probably makes for a more pleasant reading experience.

Batalov 2021-12-17 20:29

[QUOTE=lisanderke;595528]EDIT: In retrospect, I could have prompted @Batalov to simply re-read parts of this thread to make them better understand why I posted a composite factor result in this particular thread.This way it's probably easier to follow than a re-reading a conversation with lots of other posts going on in between![/QUOTE]
Retrospects are great! :thumbs-up:
Because in retrospect you are simply retelling what was already rotating on this forum 10 years ago, 5 years ago. Every few years. But did we recommend you to simply scan the forum and re-read the old threads? No, that would be cruel. Why on earth would you want to do that. :rolleyes: you are a writer, not a reader.

lisanderke 2021-12-17 20:41

[QUOTE=Batalov;595530]-snip- you are a writer, not a reader.[/QUOTE]
On the flip side, writing gives me something to do while I take range 40k to B1=10^9. Speaking of, I'll have this completed by the end of the weekend (approx.) and I might make some more attempts at stage 2 for these exponents with 30.8b5 (since 5k was giving me such a hard time!)

chalsall 2021-12-17 20:52

[QUOTE=lisanderke;595532]On the flip side, writing gives me something to do...[/QUOTE]

I /think/ you might have missed Batalov's point.

Put another way, [URL="https://liberalarts.vt.edu/magazine/2017/history-repeating.html"]History Repeating[/URL], [URL="https://en.wikipedia.org/wiki/Prior_art"]Prior art[/URL], etc, etc, etc.

You are relatively new here. And have bitten off quite a bit.

We tend to be rather patient around these here parts for new ideas and new people.

But our patience is somewhat limited when the seriousness is not understood and appreciated by those who choose to play in this space.

FWIW...

lisanderke 2021-12-17 20:56

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

Put another way, [URL="https://liberalarts.vt.edu/magazine/2017/history-repeating.html"]History Repeating[/URL], [URL="https://en.wikipedia.org/wiki/Prior_art"]Prior art[/URL], etc, etc, etc.

You are relatively new here. And have bitten off quite a bit.

We tend to be rather patient around these here parts for new ideas and new people.

But our patience is somewhat limited when the seriousness is not understood and appreciated by those who choose to play in this space.

FWIW...[/QUOTE]Oh, I quite got the point. I just saw no reason to reply to it specifically and I'd rather keep this thread on topic. With all due respect, I value everyone's patience but I'm not going to re-iterate the many thanks I've given to those that deserve it, for their patience, and redirect my thanks to Batalov for pointing out the obvious instead. (Amongst the obvious things being: Yes, it's been done before. Yes, I don't know more than most people posting here. Yes, I'm eager to do things I don't quite understand.)

chalsall 2021-12-17 21:10

[QUOTE=lisanderke;595535]Yes, I'm eager to do things I don't quite understand.[/QUOTE]

Excellent. That is how one learns.

However...

Please understand that others might understand things better than you do. The counsel given was sound. And you seemed to ignore it.

lisanderke 2021-12-17 21:31

[QUOTE=chalsall;595536]Excellent. That is how one learns.

However...

Please understand that others might understand things better than you do. The counsel given was sound. And you seemed to ignore it.[/QUOTE]


See, the way you put it was much easier to understand and much nicer to read. In reference to the counsel given, I'd rather not re-read years worth of posting, many of which diverges from the topic at hand quite quickly (these posts being one of those examples!) making it much harder to find relevant information. Especially if it's a simple and short question. I'd rather 'write' first (when comprehensive tools such as google, the search function on this forum itself, which does not even seem to take "P-1" as a valid search term, software documentation and the sort don't provide me with an answer) and 'read' the reply from someone that already possesses the information I need afterwards. If someone else prefers to search and read through decades worth of posting in this particular forum, be my guest.


All times are UTC. The time now is 06:11.

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