mersenneforum.org  

Go Back   mersenneforum.org > Prime Search Projects > No Prime Left Behind

Reply
 
Thread Tools
Old 2008-11-26, 20:49   #12
Flatlander
I quite division it
 
Flatlander's Avatar
 
"Chris"
Feb 2005
England

31×67 Posts
Default

Will k=1005-2000 be for all ks or just those not reserved by others?
Flatlander is offline   Reply With Quote
Old 2008-11-27, 04:49   #13
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

235458 Posts
Default

Quote:
Originally Posted by Flatlander View Post
Will k=1005-2000 be for all ks or just those not reserved by others?

All k's. No worries. I know you're searching a few k's for k=1400-1500 for n>350K. We'll start out at n=50K sometime in Jan. It will only be several weeks to a couple of months later that we'll likely start some top-5000 ranges after we've sieved further. Like we've done with our other drives, we'll inform people ahead of time and it will be up to them to stay ahead of our drive(s), which will progress relatively slowly for ~500 k's. It should be very easy to stay ahead of our drives, even with a small amount of resources.

Alternatively, once we start top-5000 work, you could give up your reservations and search the ranges with us instead of RPS. Then you won't have to sieve your own ranges anymore!

The main benefit about searching with NPLB is that everything is sieved in one big glob at once resulting in a huge increase in sieving efficiency and very long times between when sieving is needed. It also means that only a selected few who really enjoy sieving do it. Soon we'll be searching ~350 k's for n=600K-1M; all fully sieved already and ~500 k's for n=50K-500K; the lower n-ranges which are already sieved enough with some more sieving needed and ongoing for higher n-ranges.


Gary
gd_barnes is online now   Reply With Quote
Old 2008-11-27, 05:01   #14
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

5·2,017 Posts
Default

Quote:
Originally Posted by henryzz View Post
one thing i would say dont chose k<300 as RPS will go berserk

Personally I don't care what RPS thinks. They own nothing like we own nothing. People or projects cannot claim ownership of segments of the math world. They can only coordinate efforts on them. If there are some k's < 300 that get well behind, we'll search them. But as of right now, other then k=5, most of their k's are in decent shape (mainly thanks to our 4th drive, lol). k=5 should be at n>2.5M by now but after Benson picked them off twice because they had searched it much too slowly, we won't go there.

That said, I'm all for avoiding too much controversy and I don't want to involve others in any overly controversial effort in the future. We won't be doing large #'s of k's < 300 at once anymore. Just 1 or 2 k's here and there after follow up with specific individuals as needed like I did with Beyond on k=47 and 53, which were at n<600K a month ago and are now at n=1M with 3 primes found in the range searched. That worked out well and Beyond was perfectly fine with us filling in the range since he had previously searched some ranges for them for n>1M.


Gary
gd_barnes is online now   Reply With Quote
Old 2008-11-27, 08:41   #15
henryzz
Just call me Henry
 
henryzz's Avatar
 
"David"
Sep 2007
Cambridge (GMT)

564710 Posts
Default

Quote:
Originally Posted by gd_barnes View Post
All k's. No worries. I know you're searching a few k's for k=1400-1500 for n>350K. We'll start out at n=50K sometime in Jan. It will only be several weeks to a couple of months later that we'll likely start some top-5000 ranges after we've sieved further. Like we've done with our other drives, we'll inform people ahead of time and it will be up to them to stay ahead of our drive(s), which will progress relatively slowly for ~500 k's. It should be very easy to stay ahead of our drives, even with a small amount of resources.

Alternatively, once we start top-5000 work, you could give up your reservations and search the ranges with us instead of RPS. Then you won't have to sieve your own ranges anymore!

The main benefit about searching with NPLB is that everything is sieved in one big glob at once resulting in a huge increase in sieving efficiency and very long times between when sieving is needed. It also means that only a selected few who really enjoy sieving do it. Soon we'll be searching ~350 k's for n=600K-1M; all fully sieved already and ~500 k's for n=50K-500K; the lower n-ranges which are already sieved enough with some more sieving needed and ongoing for higher n-ranges.


Gary
you mean to say that you will be wasting all my 1009 work
henryzz is offline   Reply With Quote
Old 2008-11-27, 09:24   #16
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

53148 Posts
Default

Quote:
Originally Posted by henryzz View Post
you mean to say that you will be wasting all my 1009 work
there's no waste of time!

you're searching ranges never searched before and filling a gap for k=1009.
and NPLB will doublecheck these ranges like many others before, because there're many examples when primes were missed although the ranges were said completly searched!

continue your work!
kar_bon is offline   Reply With Quote
Old 2008-11-27, 17:27   #17
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

5·2,017 Posts
Default

Quote:
Originally Posted by henryzz View Post
you mean to say that you will be wasting all my 1009 work
Quote:
Originally Posted by kar_bon View Post
there's no waste of time!

you're searching ranges never searched before and filling a gap for k=1009.
and NPLB will doublecheck these ranges like many others before, because there're many examples when primes were missed although the ranges were said completly searched!

continue your work!

Yes, please continue your work. The same thing applies to Chris (Flatlander) who is doing quite a few k's 1400-1500 for n>350K. We'll just be double-checking that work also. That's what NPLB does because as Karsten said, many primes have been missed in the past even though ranges have been reported as complete.

An excellent example was my recent prime find for k=47 at n=~540K. The k had been reported complete to n=582K but I double-checked it starting from n=210K and found the missing prime.

No prime left behind!


Gary
gd_barnes is online now   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Cannot search! retina Forum Feedback 2 2006-12-01 03:43
Prime Search on PS-3? Kosmaj Riesel Prime Search 6 2006-11-21 15:19
121*2^n-1 search justinsane 15k Search 0 2004-05-24 20:42
Welcome to the 15k*2^n-1 search! TTn 15k Search 0 2003-05-29 09:15
3*2^n-1 search PSearch 0 1970-01-01 00:00

All times are UTC. The time now is 10:30.

Tue Mar 31 10:30:19 UTC 2020 up 6 days, 8:03, 0 users, load averages: 0.97, 1.03, 1.08

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.