Riesel base 3 - mini-drive I

[michaf]

Did you make a typo with that first one? It isn't in my sieve-file, nor in my list of deleted k's

[Flatlander]

Originally Posted by michaf

Did you make a typo with that first one? It isn't in my sieve-file, nor in my list of deleted k's

They were all cut and pasted from pfgw-prime.log after being found prp with LLR.exe, so I don't think so.

From lresults.txt

22148702*3^32952-1 is a probable prime. Time: 10.148 sec.
Please credit George Woltman's PRP for this result!

[mdettweiler]

Originally Posted by Flatlander

Is phrot quicker then? (Windows 32bit)
If so, where can I get it, how do I run it?

Yes, Phrot is generally quicker for non-base-2 PRP tests. Carlos's instructiosn pretty much cover all the bases; the only difference to keep in mind between tnerual's instructions that Carlos linked to (written for version 0.50), and the new version (0.51, which is what Geoff's binaries later in the thread are), is that 0.51 adds a .ini file that keeps track of the number currently being tested, so that if you have to stop in the middle of a file and resume testing, it will automatically pick up where it left off without you having to mess with the input file. (Note that Phrot still doesn't keep savefiles, thus you lose any progress on the current number; however, for such tiny tests as these, it's only a matter of seconds that would be potentially lost.)

[Flatlander]

6539872*3^34772-1 is confirmed prime.

Phrot is about 28% quicker than LLR here. (Core 2 quad, Win32)
I'll use it when I reserve my next ranges. I assume I'll need to use the -o option to produce a results.out file; or are you not collecting them?


btw: Which time zone is this:

All k's with primes have been removed from all files as of 18 Oct 08, 01:10 AM.

[michaf]

Originally Posted by Flatlander

They were all cut and pasted from pfgw-prime.log after being found prp with LLR.exe, so I don't think so.

From lresults.txt

Ah... It was in post 19, of henry.

I didn't include those yet when I replied :)

Sorry for the inconvenience, problem solved

[mdettweiler]

Originally Posted by Flatlander

6539872*3^34772-1 is confirmed prime.

Phrot is about 28% quicker than LLR here. (Core 2 quad, Win32)
I'll use it when I reserve my next ranges. I assume I'll need to use the -o option to produce a results.out file; or are you not collecting them?


btw: Which time zone is this:

Yes, we're collecting results files so you'll need to use the -o option. (The results.out file is comparable to the lresults.txt file that LLR would produce.)

As for the time zone: whoops, forgot to include that there. (I just copied and pasted it off of Gary's last post reporting that he had removed the primes from the posted files.) My time zone is currently EDT (GMT-4), so that's what the times I see on the forum (i.e. what I copied and pasted from) would be. I'll go and add that information to the first post so as to avoid any further confusion.

[gd_barnes]

Originally Posted by Flatlander

22148702*3^32952-1
8915336*3^33067-1
24980686*3^33130-1

are confirmed primes.


Thanks, I'll check it out.

Chris and Max,

Whenever more than one prime is found for a k-value, we only list the lowest prime found. Otherwise it's too easy to miscount the # of k's remaining.

Because 22148702*3^27950-1 was already found prime by Henry, I'll remove the prime for n=32952.

Max,

Please watch for this. You've incorrectly reduced the # of k's remaining twice for the same k-value in the 1st post of this thread. I will correct it. This is very important because it is how I make sure that the # of k's remaining is correct whenever I remove k's with primes. Srfile reports the number of sequences remaining when it writes out the file and that has to match the # of k's remaining in the 1st post. I'm constantly double-checking myself in that reguard.


Gary

[Flatlander]

93999322*3^33244-1
43666918*3^33296-1
54852866*3^33318-1
42277192*3^33524-1
94922096*3^33734-1

67520326*3^34843-1
83781844*3^34927-1
87236956*3^34955-1
58220776*3^34977-1
42353782*3^35211-1
confirmed primes.

All checked against found ks in post #1 and appear to be new, but at my age ...

[KEP]

@Flatlander:

Wow I feel like I should soon give you a fine or 2 You're truly amazing at finding primes. Now 56 primes with n>25000 has been found. The last ones is not yet posted on the Riesel base 3 attack website. Also I should add, that site may still be active in the future, but expect no more than one update per week. Also I'm propably not going to take this conjecture further than k<=500M, so if anyone is interested in overtaking the conjecture and the construction of the proofs for k>2M, please let me know. Sorry for tausing a bad news among all the good and great prime news. But I'm starting to gain interest in another project that I began of my own, which has no relation to any scientific projects nor relation to any conjectures or other mathematical claims So if you're the one interested in overtaking this conjecture, please PM me or e-mail me, doesn't matter what you do, both options is very welcome

Happy crunching, and happy prime hunting. Looking forward to see if you find more than or less than the 200 primes Gary expects for this range. Also if I should add 1 last thing, the timeframe of 4-8 weeks for k>250M to k<=500M seems to hold pretty well, in fact the range might end before 4 weeks, but no promises on this

Regards

KEP

[henryzz]

76605434 28065
56468192 28306
88742494 28361
66302042 28394
67994904 30400
37875256 30426
52641746 30493
10772464 30700
40623532 30728
are all prime
i am beating u flatlander
edit just realized ur primes hadnt been added so u r winning

[mdettweiler]

Originally Posted by gd_barnes

Chris and Max,

Whenever more than one prime is found for a k-value, we only list the lowest prime found. Otherwise it's too easy to miscount the # of k's remaining.

Because 22148702*3^27950-1 was already found prime by Henry, I'll remove the prime for n=32952.

Max,

Please watch for this. You've incorrectly reduced the # of k's remaining twice for the same k-value in the 1st post of this thread. I will correct it. This is very important because it is how I make sure that the # of k's remaining is correct whenever I remove k's with primes. Srfile reports the number of sequences remaining when it writes out the file and that has to match the # of k's remaining in the 1st post. I'm constantly double-checking myself in that reguard.


Gary

Oh, whoops. Sorry--I forgot completely about the whole issue of additional primes for k's that have already been eliminated! Thanks for reminding me on this--I'll be sure to watch for those now.

Max

P.S.: I checked both Flatlander's and henryzz's loads of primes prior to entering them. Flatlander's was, as he said, all new, and henryzz's only had one prime for a k that had already been eliminated (76605434*3^28065-1), which I left out of the status post.