mersenneforum.org Any way to run trial factoring on a SPECIFIC Exponent?
 Register FAQ Search Today's Posts Mark Forums Read

2021-12-25, 17:01   #12
Uncwilly
6809 > 6502

"""""""""""""""""""
Aug 2003
101×103 Posts

22×11×239 Posts

Quote:
 Originally Posted by Dr Sardonicus No one will believe you because in the first place, you gave absolutely no justification for your claim that M314159257 is prime, in the second place because you displayed your ignorance of, e.g. the fact that a PRP test can prove a number to be composite, and in the third place you have not given any reason to question the result of the PRP test on M314159257.
You left off the facts that the OP failed to investigate how to use the software. And since there was a claim of an error, the OP should have written their own code. Also, the OP suffers from the delusion that a number in one area of math will mystically be important in another.

The general public is not well fascinated with ever bigger primes. But one that is derived from pi would get more play than the norm. Also, if it was prime, that would make waves enough in the maths community. So, it would behoove the OP to test that one to their satisfaction first.

Has anyone tabulated a score for the OP yet?

2021-12-25, 17:13   #13
Uncwilly
6809 > 6502

"""""""""""""""""""
Aug 2003
101×103 Posts

244248 Posts

Quote:
 Originally Posted by IndrajithGamage So here are the said pngs. My question was/is, is that worktodo accurate. I specifically want to pause the PRP test, and do the P-1 at upper bound 50,000,000. I also want to know if the 4GB ram amount and other variables is sufficient. (if not, what are the best bounds for the Ram amount) My PC is AMD Ryzen 4500U with 6 cores at 4GHz so conservatively 15-20GHz can be given. Ram is 8GB and can allocate like 200GB from my SSD (NVMe if that's important).
Based upon https://www.mersenne.ca/prob.php?exp...&factorbits=82 you really should have 20 G of RAM to make a decent run at P-1 (with the settings suggested in the mid). Disk size is not relevant to P-1 processing (unless you wanted to do something massive and did not care about clock time.)

In order to get Prime95 to run the P-1 test first do the following:
Stop Prime95 and exit the program.
Edit the worktodo file to put the appropriate P-1 line ahead of the PRP line (adding N/A, aright after the =). Save the file.
Restart Prime95.

2021-12-25, 17:19   #14
Dr Sardonicus

Feb 2017
Nowhere

577010 Posts

Quote:
 Originally Posted by Uncwilly You left off the facts that the OP failed to investigate how to use the software. And since there was a claim of an error, the OP should have written their own code. Also, the OP suffers from the delusion that a number in one area of math will mystically be important in another.
Guilty as charged. I just hit the most basic reasons, which I thought were ample to refute the motive he falsely ascribed for nobody believing him. I didn't think it was worth the effort to go into a more detailed analysis.
Quote:
 Has anyone tabulated a score for the OP yet?
Not AFAIK. If you think it's worth doing, go for it.

2021-12-25, 17:33   #15
charybdis

Apr 2020

2·349 Posts

Quote:
 Originally Posted by Uncwilly But one that is derived from pi would get more play than the norm. Also, if it was prime, that would make waves enough in the maths community.
If this number was prime, the mathematical community would immediately understand that this was a complete coincidence and totally unrelated to the fact that the exponent looks like the base-10 expansion of pi. It would be no more interesting than meaningless coincidences such as ln(pi^4+pi^5) = 5.999999956... and the like.

 2021-12-25, 19:27 #16 kriesel     "TF79LL86GIMPS96gpu17" Mar 2017 US midwest 3×5×431 Posts GIMPS101: Never post a live AID to the forum. (png3, PRP line) Stuff like this is why I say read the reference info especially the beginner-oriented material. Last fiddled with by kriesel on 2021-12-25 at 19:29
2022-01-24, 14:55   #17
IndrajithGamage

"Indrajith Gamage"
Dec 2021
Sri Lanka

32 Posts

Quote:
 Originally Posted by kriesel GIMPS101: Never post a live AID to the forum. (png3, PRP line) Stuff like this is why I say read the reference info especially the beginner-oriented material.

Hi Kriesel, I'm back with a very strong rapport now aren't I?

https://www.mersenne.org/report_expo...ll=1&ecmhist=1

I see that you have completed the P1 factoring of this specific exponent to a large extent (130M in fact) and found that it returns no factors. Are you willing to give it a try with the full PRP now? (or perhaps you are still not convinced) I'm running it but seems like it's taking more than a year and a half (524 days) to complete with my PC. (still at 0.2%)

Last fiddled with by IndrajithGamage on 2022-01-24 at 15:00

 2022-01-24, 15:55 #18 Uncwilly 6809 > 6502     """"""""""""""""""" Aug 2003 101×103 Posts 1051610 Posts You obviously don't understand how P-1 is different than TF. Kriesel's factoring had a 3.88% chance of finding a factor. Using a B2 pf 130M is not the same as TF up to 130M. That took about 252 GHz-days of effort. Doing the PRP test takes about 7,010 GHz days of effort. If you want to do that, fine. And you obviously don't know that Kriesel does work on virtually all silly claims like this. He want to find factors to prove the claims false. Many other less earnest and lower claims he has done PRP on too (if no factor was found). He wants there to be no outstanding claims of Mersenne Primes that have not been proven either prime or composite. Your claim is nothing special to him. Since you are doing the PRP test, please make sure that you are using Prime95 v 30.7 (unless you have a powerful GPU that you are running the test on.) It will help to ensure that the test is error free and allow someone else to quickly verify your result.
2022-01-24, 16:03   #19
kriesel

"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

3×5×431 Posts

Quote:
 Originally Posted by IndrajithGamage I'm back with a very strong rapport now aren't I?
Really? I don't know why you would claim that.
Quote:
 I see that you have completed the P1 factoring of this specific exponent to a large extent (130M in fact) and found that it returns no factors.
About 96% of composite Mersenne numbers have no factor found by P-1 factoring to usual optimal bounds such as these. Not that they don't have factors; in P-1 factoring, no factor found means there was no sufficiently smooth factor found within the scope covered by the bounds chosen. A composite Mersenne might have 100 factors and P-1 run to appropriately economically optimal bounds find none of them.
Quote:
 Are you willing to give it a try with the full PRP now? (or perhaps you are still not convinced) I'm running it but seems like it's taking more than a year and a half (524 days) to complete with my PC. (still at 0.2%)
No. My hardware capable of tackling such a task in under a year run time is queued months deep, or deeper. It would also technically be considered poaching by many. I have other priority work waiting. So slog away on what you chose.
Quote:
 Originally Posted by Uncwilly Since you are doing the PRP test, please make sure that you are using Prime95 v 30.7 (unless you have a powerful GPU that you are running the test on.) It will help to ensure that the test is error free and allow someone else to quickly verify your result.
Since it is reporting progress, it is likely prime95. Since it has reported 0.2%, it has already passed the point where there would either be a first power 9 proof residue saved, or not. (1/29 ~ 0.195%). If the system is not yet set for adequate temporary space, that could be corrected soon and still enable proof power 8. If it was left at default settings, much lower maximum proof power would be allowed by a small power proof, and as a result much more work would be necessary eventually to certify or double check the run. About 25. GB of temporary space would be required for proof power 9 on ~385M exponent. See the CONTROLLING RESOURCE USAGE section of prime95's readme.txt and linearly extrapolate temporary space for exponent.
If it was left at default settings, it would be faster overall (test and verify) to set it properly soon, and at least stop and restart, or perhaps start over.
Quote:
 Originally Posted by Uncwilly Kriesel does work on virtually all silly claims like this. He want to find factors to prove the claims false. Many other less earnest and lower claims he has done PRP on too (if no factor was found). He wants there to be no outstanding claims of Mersenne Primes that have not been proven either prime or composite. Your claim is nothing special to him.
Uncwilly is correct that I prefer similar nonsense not go unrefuted for long, and have spent a lot of computing resources on refutation. I have sought out, tabulated, and in most cases refuted, hundreds of Predict Mxx predictions and various other dubious claims. ("Dubious" being more tactful than some other descriptors that would be apt.) There are few that have not either been proven composite by finding a factor, or conclusively been shown composite by LL primality test or PRP-C. For those that are not intractably large exponents, and have had a primality test completed but not DC or successful cert, I am working through that verification backlog this year. There are a couple that are 1G<p<2.3G that have not yet had a primality test performed. One of those is in TF now. Performing cert on either is problematic since p>1G will not be accepted by mersenne.org, and would require special handling to get it to George for cert processing. Above ~1.17G, software to perform the cert does not exist. Above ~231, software to perform prp proof generation does not yet exist. (Ernst is working on adding prp proof generation for Mlucas at last report, which will probably encompass up to ~233.) For mersenne exponents above ~233, software to perform P-1 factoring does not exist. I have promised Ernst I would help with P-1 stage 2 on F33, and committed to doing some OBD P-1 also.
By summer 2023, I may have reduced my queued work enough to run a Cert or DC of your ~999,999,8xx ppb likely composite result on your chosen exponent ~385M.

My vote is for letting this thread run a while. It occurs to me as still mildly amusing.

Last fiddled with by kriesel on 2022-01-24 at 17:00

 2022-01-24, 16:21 #20 mathwiz   Mar 2019 32·29 Posts Why are we still entertaining this troll / charlatan? Maybe a mod should lock the thread.
2022-01-24, 16:45   #21
Uncwilly
6809 > 6502

"""""""""""""""""""
Aug 2003
101×103 Posts

22×11×239 Posts

Quote:
 Originally Posted by mathwiz Why are we still entertaining this troll / charlatan?
Because the best way to deal with people that think that they know the next prime is to show them that their "special" number isn't special. That may take a few minutes, days, or never. Those that never accept it eventually leave the forum, or do something that gets them banned. Those that eventually understand that their wild guess is as good as picking lottery numbers either leave, wander off, or become more educated about things. If we can convert them, to educated, so much the better. You are free to ignore them (there is a tool for that) or this thread.

BTW, carping about each troll (repeatedly) is not an endearing trait.

Last fiddled with by Uncwilly on 2022-01-24 at 16:46

2022-01-24, 17:29   #22
Dr Sardonicus

Feb 2017
Nowhere

2·5·577 Posts

Quote:
 Originally Posted by IndrajithGamage Hi Kriesel, 384989999 Are you willing to give it a try with the full PRP now? I'm running it but seems like it's taking more than a year and a half (524 days) to complete with my PC.
I've already seen this movie. You have forfeited the right to ask him (or anyone else) to do a PRP test on any exponent you claim yields a Mersenne prime.

In this post, you claimed that M314159257 was prime. In this post he pointed out that he had already run a PRP test that proved the number composite six months previous:
Code:
[Fri Jun 11 12:40:01 2021]
{"status":"C", "exponent":314159257, "worktype":"PRP-3", "res64":"3A7F291B4DA3A___", "residue-type":1, "res2048":"8D5A...", "fft-length":17694720, "shift-count":232515581, "error-code":"00000000", "security-code":"________", "program":{"name":"Prime95", "version":"30.6", "build":4, "port":4}, "timestamp":"2021-06-11 17:40:01", "errors":{"gerbicz":0}, "proof":{"version":2, "power":9, "hashsize":64, "md5":"2675..."}, "user":"Kriesel", "computer":"moa"}

Quote:
 Originally Posted by IndrajithGamage M314159257 is prime, and if any methods currently used are saying otherwise, I can only think that some sort of error is causing that.
So we already know the script: If he does a PRP test on your current pet exponent, and it proves the number composite, you'll just say "It must be a mistake."

I find it interesting that you're apparently not willing just to let your own test run to completion.

 Similar Threads Thread Thread Starter Forum Replies Last Post king Information & Answers 5 2018-02-21 18:15 Karl M Johnson Software 12 2015-10-12 15:56 Qubit Software 4 2014-01-15 05:04 137ben Software 9 2012-05-26 19:16 Unregistered Information & Answers 19 2008-05-30 08:34

All times are UTC. The time now is 12:58.

Thu May 19 12:58:05 UTC 2022 up 35 days, 10:59, 0 users, load averages: 1.58, 1.68, 1.69