## Other

drummers-lowrise

Message boards : General discussion : Assumptions and predictions for the new subrooject ... Wieferich !

 Post to thread Subscribe SortOldest firstNewest firstHighest rated posts first
Author Message
Grzegorz Roman Granowski

Joined: 12 Mar 10
Posts: 83
ID: 56804
Credit: 15,034,803,024
RAC: 33,096,340

Message 145863 - Posted: 30 Nov 2020 | 11:22:03 UTC

Greetings and warm welcome ..,. !!!

are there predictions of how many tasks ... are needed to make a find ... ???

I started counting yesterday ... and ... I had hope ... to make some finds ... but after ... 3000 wu's ... there is no find ...

my last fing was made years ago, so ... it will be nice to get Wiferich !!!

regards, Grzegorz Roman Granowski ...
____________

Yves Gallot
Volunteer developer
Project scientist

Joined: 19 Aug 12
Posts: 669
ID: 164101
Credit: 305,042,960
RAC: 0

Message 145865 - Posted: 30 Nov 2020 | 12:40:46 UTC - in response to Message 145863.

For each prime p, the probability is 1/p.
If p is the nth prime, we have p ~ n log n then the expected number of solutions for n1 log n1 <= p <= n2 log n2 is Integrate[n1, n2] dx / (x log x) = log log n2 - log log n1.

The range is p1 = 14·1016 => n1 ~ 3.546·1015; p2 = 262 ~ 4.6·1018 => n2 ~ 1.073·1017.

Expected number of solutions for the project = 3.669 - 3.578 = 0.091.

Grzegorz Roman Granowski

Joined: 12 Mar 10
Posts: 83
ID: 56804
Credit: 15,034,803,024
RAC: 33,096,340

Message 145868 - Posted: 30 Nov 2020 | 12:57:24 UTC - in response to Message 145865.

Dear Mr. Project scientist ... thank You for the answer ...

regards, Grzegorz Roman Granowski ...
____________

JeppeSN

Joined: 5 Apr 14
Posts: 1502
ID: 306875
Credit: 33,917,665
RAC: 17,786

Message 145881 - Posted: 30 Nov 2020 | 15:06:37 UTC

Also, for each time there is one Wieferich (A=0), there is going to be 2000 near-Wieferichs (A=-1000,-999,-998,...,-1 and +1,+2,+3,...,+1000). So multiplying Yves's number by 2000 I get about 180 near-Wieferich.

And also 180 near-Wall-Sun-Suns.

However, at the moment we are double checking work that was done years ago on PRPNet. The near-finds we meet that have already been discovered back then, are not shown. But at some moment we will make a near-find that was missed by the original search (malfunctioning computers were not identified on PRPNet).

/JeppeSN

Grzegorz Roman Granowski

Joined: 12 Mar 10
Posts: 83
ID: 56804
Credit: 15,034,803,024
RAC: 33,096,340

Message 145975 - Posted: 1 Dec 2020 | 18:08:22 UTC - in response to Message 145881.

However, at the moment we are double checking work that was done years ago on PRPNet. The near-finds we meet that have already been discovered back then, are not shown. But at some moment we will make a near-find that was missed by the original search (malfunctioning computers were not identified on PRPNet).

/JeppeSN

greetings ...

how long will it take ... those doublecheck works made by prpnet ????

regards, Grzegorz Roman Granowski ...

____________

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13625
ID: 53948
Credit: 274,799,588
RAC: 248,110

Message 145977 - Posted: 1 Dec 2020 | 18:39:28 UTC - in response to Message 145975.

However, at the moment we are double checking work that was done years ago on PRPNet. The near-finds we meet that have already been discovered back then, are not shown. But at some moment we will make a near-find that was missed by the original search (malfunctioning computers were not identified on PRPNet).

/JeppeSN

greetings ...

how long will it take ... those doublecheck works made by prpnet ????

regards, Grzegorz Roman Granowski ...

That depends on participation.
____________
My lucky number is 75898524288+1

Grzegorz Roman Granowski

Joined: 12 Mar 10
Posts: 83
ID: 56804
Credit: 15,034,803,024
RAC: 33,096,340

Message 145978 - Posted: 1 Dec 2020 | 18:43:01 UTC - in response to Message 145977.

all right !!! Thank You ...

and, Dear Mr. Goetz , if You please, what was the max value ... that was checked by prpnet???

regards, Grzegorz Roman Granowski ...
____________

JeppeSN

Joined: 5 Apr 14
Posts: 1502
ID: 306875
Credit: 33,917,665
RAC: 17,786

Message 145984 - Posted: 1 Dec 2020 | 21:45:34 UTC - in response to Message 145978.

Did you see the thread https://www.primegrid.com/forum_thread.php?id=9446? 5.97077e17 for Wieferich (2.63691e17 for Wall-Sun-Sun). /JeppeSN

Grzegorz Roman Granowski

Joined: 12 Mar 10
Posts: 83
ID: 56804
Credit: 15,034,803,024
RAC: 33,096,340

Message 145985 - Posted: 1 Dec 2020 | 22:10:39 UTC - in response to Message 145984.

Did you see the thread https://www.primegrid.com/forum_thread.php?id=9446? 5.97077e17 for Wieferich (2.63691e17 for Wall-Sun-Sun). /JeppeSN

Thank You,

Greatly appreciated ...

regards, Grzegorz Roman Granowski ...
____________

mikey

Joined: 17 Mar 09
Posts: 1268
ID: 37043
Credit: 525,568,701
RAC: 10,704

Message 146038 - Posted: 3 Dec 2020 | 12:22:24 UTC - in response to Message 145985.

Did you see the thread https://www.primegrid.com/forum_thread.php?id=9446? 5.97077e17 for Wieferich (2.63691e17 for Wall-Sun-Sun). /JeppeSN

Thank You,

Greatly appreciated ...

regards, Grzegorz Roman Granowski ...

Are you looking for ANY primes or mostly big ones that get reported into the top 5000 list? Because in the last few days I've found several primes in Genefer-15 using my gpu's but the project does NOT find primes that are big enough to be reported in the top 5000 database. Now OBVIOUSLY past finding of primes does NOT equal future finding of primes and to be honest this little stretch here is unusual even for me, I usually find maybe one a month running Genefer-15 tasks. I My gpu usually runs a bit over 700 tasks per day, the tasks are short, and around 40% of those are 'firsts'. Right now I do have several pc's using their gpu's running Genefer-15 so that could be why there are more recently.

Here's the PM I get when I find one:
"Dear Primefinder,

Congratulations! Our records indicate that a computer registered by you has found a unique prime number. This computer is assigned to the Generalized Fermat Prime Search n=15 (GFN-15). Since candidates on this subproject are not large enough to report to the Top 5000 Primes List, your prime is visible immediately.

Workunit 687479372 : 206325772^32768+1 (272452 digits)

If you have any questions or concerns, please contact us and we will surely resolve any problems.

Once again, congratulations on your find! Thank you for participating in PrimeGrid.

Sincerely,
The PrimeGrid Staff"

Bur
Volunteer tester

Joined: 25 Feb 20
Posts: 400
ID: 1241833
Credit: 161,795,275
RAC: 1,012,044

Message 146063 - Posted: 3 Dec 2020 | 17:46:26 UTC - in response to Message 146038.

As there was some discussion how to find the current upper bound. Instead of deducing from the WU name, you can simply check the stderr output for the task. It shows the commandline, which contains the range that was tested.

Current tasks show we're at around 90P or 0.9E17, i.e. about 20% of old range already done after 4 days.

A question I was pondering: will the results we get be deemed valid for use in peer-reviewed publications? Since we'll likely not find a prime, the main result will be the upper limit we reached. So it'd be nice if it has some kind of impact.
____________
Primes: 1281979 & 12+8+1979 & 1+2+8+1+9+7+9 & 1^2+2^2+8^2+1^2+9^2+7^2+9^2 & 12*8+19*79 & 12^8-1979 & 1281979 + 4 (cousin prime)

Message boards : General discussion : Assumptions and predictions for the new subrooject ... Wieferich !