mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Conjectures 'R Us (https://www.mersenneforum.org/forumdisplay.php?f=81)
-   -   Sierp base 3 - mini-drive Ib (https://www.mersenneforum.org/showthread.php?t=10955)

 MooooMoo 2008-12-20 21:08

54320036*3^64148+1 is prime. It was found as a PRP by LLR, and it was confirmed as prime by Proth.

Here's the output:

[CODE]
54320036*3^64148+1 is a probable prime. Time: 67.384 sec.
...
54320036*3^64148 + 1 may be prime. (a = 2)
54320036*3^64148 + 1 is prime! (a = 5) [30615 digits]
[/CODE]

 mdettweiler 2008-12-21 06:20

[quote=MooooMoo;154311]54320036*3^64148+1 is prime. It was found as a PRP by LLR, and it was confirmed as prime by Proth.

Here's the output:

[code]
54320036*3^64148+1 is a probable prime. Time: 67.384 sec.
...
54320036*3^64148 + 1 may be prime. (a = 2)
54320036*3^64148 + 1 is prime! (a = 5) [30615 digits]
[/code][/quote]
Um...just to clarify, are you reserving 64K-65K or any other ranges on this drive? Or was this found independently?

 MyDogBuster 2008-12-21 16:30

50K - 60K complete 22 proven primes

77976142*3^50169+1 is prime! (62.8740s+0.0014s)
69026702*3^50280+1 is prime! (135.2472s+0.0014s)
91379272*3^50340+1 is prime! (97.0802s+0.0014s)
92793016*3^50687+1 is prime! (55.3105s+0.0014s)
55075498*3^51708+1 is prime! (56.4614s+0.0014s)
70009502*3^51837+1 is prime! (54.1643s+0.0015s)
71192354*3^52042+1 is prime! (96.1381s+0.0014s)
52912976*3^52588+1 is prime! (58.2575s+0.0015s)
89106848*3^53018+1 is prime! (54.5909s+0.0015s)
77554786*3^53248+1 is prime! (55.0181s+0.0015s)
86000056*3^53788+1 is prime! (55.5784s+0.0015s)
66048746*3^54336+1 is prime! (173.5886s+0.0015s)
53109118*3^54460+1 is prime! (65.8118s+0.0016s)
64133198*3^54595+1 is prime! (66.5338s+0.0015s)
64341406*3^56947+1 is prime! (184.2590s+0.0017s)
81956716*3^56961+1 is prime! (70.2281s+0.0017s)
96746836*3^57284+1 is prime! (298.3549s+0.0017s)
74481242*3^58110+1 is prime! (71.3721s+0.0017s)
52898312*3^58616+1 is prime! (72.4967s+0.0018s)
97132676*3^58743+1 is prime! (71.9681s+0.0017s)
90784756*3^59179+1 is prime! (77.6378s+0.0017s)
71744534*3^59498+1 is prime! (74.0493s+0.0018s)

Results emailed

I will not be taking anymore files for this drive until we figure out who is doing what.

 MooooMoo 2008-12-21 21:27

[QUOTE=mdettweiler;154364]Um...just to clarify, are you reserving 64K-65K or any other ranges on this drive? Or was this found independently?[/QUOTE]
I found this prime during a fixed n-search a few months ago, and I didn't know that this was one of the remaining k's left in the base 3 conjecture until I checked this thread yesterday.

I don't know whether or not there are any other primes in the 64K-65K range.

 mdettweiler 2008-12-21 22:48

[quote=MooooMoo;154459]I found this prime during a fixed n-search a few months ago, and I didn't know that this was one of the remaining k's left in the base 3 conjecture until I checked this thread yesterday.

I don't know whether or not there are any other primes in the 64K-65K range.[/quote]
Okay, then, in that case I'll mark it down as found by you but with a special side note indicating that this was found independently of the 64K-65K range in this drive.

Gary, should we mark this down as found in the 64K-65K range on the reservations table when we get to that point?

 MyDogBuster 2008-12-26 05:32

Taking 60K - 65K

 gd_barnes 2008-12-27 08:00

[quote=mdettweiler;154466]Okay, then, in that case I'll mark it down as found by you but with a special side note indicating that this was found independently of the 64K-65K range in this drive.

Gary, should we mark this down as found in the 64K-65K range on the reservations table when we get to that point?[/quote]

Mike,

That was a most astute observation from an old effort! Thanks for sharing it with us.

Do you have any others that we have remaining?

Max,

This is no big deal. Yes, show it as found by Mike as you have done. For the n=60K-65K range, I will just show it as (plus 1 other) like I've done with the Riesel and Sierp base 16 drives where primes are found by other efforts such as when a prime was found for Riesel base-2 odd-n that converted to a Riesel base 16 prime.

Ian,

Please go ahead and include the k in your searching of the n=60K-65K range to see if we find a smaller prime for it. I like to find the smallest prime for each k. Otherwise, we could just search for huge primes n>200K and list those to make us look good. That would be "cheating". lol :smile:

This is bound to happen from time to time. I found a Riesel base 256 prime for k=1478 here that NPLB confirmed for k=739 base 2 two weeks later.

For the very high Riesel base 16 prime for k=13854 shown in the 1st post of that drive, I was doing a search for the Riesel base 2 odd-n conjecture for k=6927. It just so happens that the effort overlapped with bases 4 and 16.

The nature of this effort will overlap a few different other efforts at times, although we attempt to minimize it as much as possible. In other words, if I know someone else is searching something that will encompass all of an effort that we have here, I will show it as reserved by them and not hand it out in the drives. An example is k=443 that Curtis at RPS is working on for base 2 but that overlaps our Riesel base 16. I show it as reserved by him on base 16 so that we don't include it in our drives. Since it's so low weight, he is well past where our drives are at so it's not an issue.

In the case of a fixed-n search like this, there's no way that it can encompass an entire k on any base that we have so there's no way to show it reserved by anyone.

Gary

 mdettweiler 2008-12-27 08:12

[quote=gd_barnes;155267]Max,

This is no big deal. Yes, show it as found by Mike as you have done. For the n=60K-65K range, I will just show it as (plus 1 other) like I've done with the Riesel and Sierp base 16 drives where primes are found by other efforts such as when a prime was found for Riesel base-2 odd-n that converted to a Riesel base 16 prime.[/quote]
Oh, duh--I had forgotten at the time about how we have already been using the "plus x other" notation for primes found in an individual-k reservation on team-drive bases. Thanks for reminding me. :smile:

One other question, though: When I marked it down I counted it in the "this drive started with x k's, has found x k's with primes, and has x k's remaining" section. Is this correct? Or do we not count it there for non-drive k's?

 MooooMoo 2008-12-27 08:48

[QUOTE=gd_barnes;155267]Mike,

That was a most astute observation from an old effort! Thanks for sharing it with us.

Do you have any others that we have remaining?
[/QUOTE]

I didn't even start from k=1 or anywhere near k=1 for this fixed-n search, so there weren't many base 3 primes found. I was quite surprised that one of them turned out to be a k that CRUS was working on.

 gd_barnes 2008-12-27 08:52

More files have been posted with k=54320036 removed...a total of 108 k's remaining.

Ian, if you find it useful, I reposted your reserved file below the reservations with all of the k's removed that you found primes for n=50K-60K. I didn't remove k=54320036 so there are a total of 109 k's remaining in the file.

Gary

 gd_barnes 2008-12-27 09:02

[quote=mdettweiler;155268]Oh, duh--I had forgotten at the time about how we have already been using the "plus x other" notation for primes found in an individual-k reservation on team-drive bases. Thanks for reminding me. :smile:

One other question, though: When I marked it down I counted it in the "this drive started with x k's, has found x k's with primes, and has x k's remaining" section. Is this correct? Or do we not count it there for non-drive k's?[/quote]

Yes, we definitely remove the k from that count even though we haven't technically searched up to that range yet. I checked that and you had it correct. The reasoning:

1. Balancing: The # of k's remaining shown in the 1st post must equal the # of k's we started with minus the # of primes found as shown in that post. I use that constantly as a double-check against k's remaining on my web pages that include far more ranges than just this drive.

2. Several other prior efforts: There were a # of top-5000 and other previous base 3 primes that we used to eliminate k's before searching for n>25K. The montra was: Search everything to n=25K and see what remains to attempt to find the lowest primes. For the k's remaining at that point, then use higher primes found from other efforts to eliminate them. In this case, Mike's effort is one of those "other" efforts.

I'm not going to be overly anal about finding the lowest prime on all k's; just most of them. lol Had Mike's prime been n>~100K, I would have had us remove it from the rest of our drive like we, in effect did, for other efforts before the drive started. But since we're close, we may as well see if it is the lowest. The thinking being that we don't want to spend too much extra CPU time to find the lowest prime but a little bit is OK.

Gary

All times are UTC. The time now is 14:10.