TPS Discovery Rally: January 14-16

Oddball · 2011-01-09, 08:20

On January 15 2007, TPS discovered its first record-breaking twin: 2003663613*2^195000+/1. Four years later, we're holding a weekend rally to commemorate that event. It starts on 0:00 CST January 14 (6:00 GMT), and it ends exactly three days later.

The main theme for this rally will be PRPnet vs. LLRnet. Pick a server you are more comfortable with, and try to stick with that server for the duration of the rally. There are also several side competitions: first person to discover a prime, person with the most candidates tested during the rally, and person with the most primes discovered during the rally. Of course, the rally also features a contest to be the first person to discover a 145,000+ digit twin, and the prize for winning that is a spot at the top of this list: http://primes.utm.edu/top20/page.php?id=1

Nobody has to be part of a team to participate in the rally, but if you want to form one, post here and say what team you're on. The individual stats for both servers are updated every 15 minutes here: http://noprimeleftbehind.net/tps/ , and they (along with the team scores) will be posted on this thread at the end of each day. One k/n pair takes a single core roughly 5 minutes to test.

Here is a guide to PRPnet and LLRnet for newcomers: http://www.mersenneforum.org/showthread.php?t=13805

PRPnet config info.:
server = "www.noprimeleftbehind.net"
port = 12000

LLRnet config info.:
server = "www.noprimeleftbehind.net"
port = 12050

Total rally stats:

Code:

User                k/n pairs tested      Primes found
sm5ymt                21529                   3
mdettweiler            7884                   1
Flatlander             7128                   3
Oddball                5317                   0
Ralf_Recker            4010                   3
Mini-Geek              3236                   1
cipher                 2962                   0
MiHost                 1046                   0
Robert47                 66                   0
acer                      1                   0
 
Team                           k/n pairs tested    Primes found
Everyone else                   34931                  6
Raiders of the Lost Primes      18248                  5
 
Server           k/n pairs tested       Primes found
PRPnet             39688                     8
LLRnet             13491                     3
 
Total: 53179 pairs tested, 11 primes found, 0 twins found
Rate: 1 k/n pair every 4.87 seconds
Density: 1 prime every 4834 candidates

January 14 stats:

Code:

User                k/n pairs tested
Flatlander             2621
mdettweiler            2492
Oddball                1780
cipher                 1228
Mini-Geek              1192
Ralf_Recker             967
 
Team                           k/n pairs tested
Raiders of the Lost Primes       6305
Everyone else                    3975
 
Server           k/n pairs tested
PRPnet              5879
LLRnet              4401
 
Total: 10280 pairs tested
Rate: 1 candidate every 8.40 seconds

January 15 stats:

Code:

User                k/n pairs tested
sm5ymt                 2684
mdettweiler            2680
Flatlander             2638
Oddball                1766
Ralf_Recker            1537
Mini-Geek              1188
cipher                  444
Robert47                  1
 
Team                           k/n pairs tested
Raiders of the Lost Primes       6506
Everyone else                    6432
 
Server           k/n pairs tested
PRPnet              8534
LLRnet              4404
 
Total: 12938 pairs tested
Rate: 1 candidate every 6.68 seconds

January 16 stats:

Code:

User                k/n pairs tested
sm5ymt                18845
mdettweiler            2712
Flatlander             1869
Oddball                1771
Ralf_Recker            1506
cipher                 1290
MiHost                 1046
Mini-Geek               856
Robert47                 65
acer                      1
 
Team                           k/n pairs tested
Everyone else                   24524
Raiders of the Lost Primes       5437
 
Server           k/n pairs tested
PRPnet             25275
LLRnet              4686
 
Total: 29961 pairs tested
Rate: 1 candidate every 2.88 seconds

Primes found during the rally:

Code:

31239*2^484153-1 (Flatlander)
7077*2^482292-1 (Ralf_Recker)
10641*2^482399-1 (mdettweiler)
30705*2^482399-1 (Mini-Geek)
82719*2^482469-1 (Ralf_Recker)
52125*2^482492-1 (sm5ymt)
8847*2^484394-1 (Flatlander)
78081*2^482529-1 (sm5ymt)
7533*2^484406-1 (Flatlander)
67917*2^482936-1 (Ralf_Recker)
10071*2^482995-1 (sm5ymt)

Twins found during the rally:

Code:

None

Happy prime hunting!

cipher · 2011-01-09, 16:40

Awesome Idea!!

mdettweiler · 2011-01-09, 18:45

I'm in.

I should be able to participate with at least one quad (hopefully two) plus I'll see how much of my dualcore I can free up in time for the rally. I'll be on PRPnet, FYI.

Edit: ooh, this reminds me, I need to get some more work loaded into the servers (PRPnet in particular) before the rally. I just spotted gribozavr's latest small k file (January 7) in the sieving thread...I'll download that and get the servers loaded up shortly.

Flatlander · 2011-01-09, 22:30

Sounds fun.

I'll check my diary ... yep, empty for 2011. (I exaggerate, I have to worm the dog in March.)
I'm having issues with my quad but will try to be ready for the 15th.

edit:14th.

Oddball · 2011-01-12, 04:14

Two days to go until the rally starts!

Max, could you replace the n=484105-484899 candidates in the LLRnet server with the n=484105-484899 candidates from the new sieve file? The new file has a much higher sieve depth.

mdettweiler · 2011-01-12, 06:15

Quote:

Originally Posted by Oddball

Two days to go until the rally starts!

Max, could you replace the n=484105-484899 candidates in the LLRnet server with the n=484105-484899 candidates from the new sieve file? The new file has a much higher sieve depth.

Good idea--done.

mdettweiler · 2011-01-13, 21:14

One quad has been added to PRPnet for the rally. The other quad and dualcore will follow sometime later today.

Flatlander · 2011-01-13, 21:29

Quote:

Originally Posted by mdettweiler

One quad has been added to PRPnet for the rally. The other quad and dualcore will follow sometime later today.

Great minds...
I added a quad to LLRNet a short while ago. I'll get my laptop on later. Maybe also the kids' PC.

Oddball · 2011-01-14, 02:11

Quote:

Originally Posted by Flatlander

Great minds...
I added a quad to LLRNet a short while ago. I'll get my laptop on later. Maybe also the kids' PC.

The rally hasn't even started yet, and you already found a prime. Congrats!

Doublecheck:
20559*2^484107-1 is prime! Time : 269.115 sec.
20559*2^484107+1 is not prime. Proth RES64: E0BE74E18434F813 Time : 341.045 sec.

I'll add a few cores to LLRnet an hour or two from now.

Ralf Recker · 2011-01-14, 14:35

The PRPNet 4.1.2 client again deletes results the were not successfully reported...

...so they remain in the pending list until they time out.

Oddball · 2011-01-14, 18:38

Congrats to Flatlander, who found the first prime in the rally!

2011-01-09, 08:20	#1
Oddball May 2010 499 Posts	TPS Discovery Rally: January 14-16 On January 15 2007, TPS discovered its first record-breaking twin: 20036636132^195000+/1. Four years later, we're holding a weekend rally to commemorate that event. It starts on 0:00 CST January 14 (6:00 GMT), and it ends exactly three days later. The main theme for this rally will be PRPnet vs. LLRnet. Pick a server you are more comfortable with, and try to stick with that server for the duration of the rally. There are also several side competitions: first person to discover a prime, person with the most candidates tested during the rally, and person with the most primes discovered during the rally. Of course, the rally also features a contest to be the first person to discover a 145,000+ digit twin, and the prize for winning that is a spot at the top of this list: http://primes.utm.edu/top20/page.php?id=1 Nobody has to be part of a team to participate in the rally, but if you want to form one, post here and say what team you're on. The individual stats for both servers are updated every 15 minutes here: http://noprimeleftbehind.net/tps/ , and they (along with the team scores) will be posted on this thread at the end of each day. One k/n pair takes a single core roughly 5 minutes to test. Here is a guide to PRPnet and LLRnet for newcomers: http://www.mersenneforum.org/showthread.php?t=13805 PRPnet config info.: server = "www.noprimeleftbehind.net" port = 12000 LLRnet config info.: server = "www.noprimeleftbehind.net" port = 12050 Total rally stats: Code: User k/n pairs tested Primes found sm5ymt 21529 3 mdettweiler 7884 1 Flatlander 7128 3 Oddball 5317 0 Ralf_Recker 4010 3 Mini-Geek 3236 1 cipher 2962 0 MiHost 1046 0 Robert47 66 0 acer 1 0 Team k/n pairs tested Primes found Everyone else 34931 6 Raiders of the Lost Primes 18248 5 Server k/n pairs tested Primes found PRPnet 39688 8 LLRnet 13491 3 Total: 53179 pairs tested, 11 primes found, 0 twins found Rate: 1 k/n pair every 4.87 seconds Density: 1 prime every 4834 candidates January 14 stats: Code: User k/n pairs tested Flatlander 2621 mdettweiler 2492 Oddball 1780 cipher 1228 Mini-Geek 1192 Ralf_Recker 967 Team k/n pairs tested Raiders of the Lost Primes 6305 Everyone else 3975 Server k/n pairs tested PRPnet 5879 LLRnet 4401 Total: 10280 pairs tested Rate: 1 candidate every 8.40 seconds January 15 stats: Code: User k/n pairs tested sm5ymt 2684 mdettweiler 2680 Flatlander 2638 Oddball 1766 Ralf_Recker 1537 Mini-Geek 1188 cipher 444 Robert47 1 Team k/n pairs tested Raiders of the Lost Primes 6506 Everyone else 6432 Server k/n pairs tested PRPnet 8534 LLRnet 4404 Total: 12938 pairs tested Rate: 1 candidate every 6.68 seconds January 16 stats: Code: User k/n pairs tested sm5ymt 18845 mdettweiler 2712 Flatlander 1869 Oddball 1771 Ralf_Recker 1506 cipher 1290 MiHost 1046 Mini-Geek 856 Robert47 65 acer 1 Team k/n pairs tested Everyone else 24524 Raiders of the Lost Primes 5437 Server k/n pairs tested PRPnet 25275 LLRnet 4686 Total: 29961 pairs tested Rate: 1 candidate every 2.88 seconds Primes found during the rally: Code: 312392^484153-1 (Flatlander) 70772^482292-1 (Ralf_Recker) 106412^482399-1 (mdettweiler) 307052^482399-1 (Mini-Geek) 827192^482469-1 (Ralf_Recker) 521252^482492-1 (sm5ymt) 88472^484394-1 (Flatlander) 780812^482529-1 (sm5ymt) 75332^484406-1 (Flatlander) 679172^482936-1 (Ralf_Recker) 100712^482995-1 (sm5ymt) Twins found during the rally: Code: None Happy prime hunting! Last fiddled with by Oddball on 2011-01-17 at 18:59

2011-01-09, 18:45	#3
mdettweiler A Sunny Moo Aug 2007 USA 2·47·67 Posts	I'm in. I should be able to participate with at least one quad (hopefully two) plus I'll see how much of my dualcore I can free up in time for the rally. I'll be on PRPnet, FYI. Edit: ooh, this reminds me, I need to get some more work loaded into the servers (PRPnet in particular) before the rally. I just spotted gribozavr's latest small k file (January 7) in the sieving thread...I'll download that and get the servers loaded up shortly. Last fiddled with by mdettweiler on 2011-01-09 at 18:47

2011-01-09, 22:30	#4
Flatlander I quite division it "Chris" Feb 2005 England 31·67 Posts	Sounds fun. I'll check my diary ... yep, empty for 2011. (I exaggerate, I have to worm the dog in March.) I'm having issues with my quad but will try to be ready for the 15th. edit:14th. Last fiddled with by Flatlander on 2011-01-09 at 22:38

2011-01-12, 04:14	#5
Oddball May 2010 499 Posts	Two days to go until the rally starts! Max, could you replace the n=484105-484899 candidates in the LLRnet server with the n=484105-484899 candidates from the new sieve file? The new file has a much higher sieve depth. Last fiddled with by Oddball on 2011-01-12 at 04:20

2011-01-13, 21:14	#7
mdettweiler A Sunny Moo Aug 2007 USA 14232₈ Posts	One quad has been added to PRPnet for the rally. The other quad and dualcore will follow sometime later today. Last fiddled with by mdettweiler on 2011-01-13 at 21:14 Reason: nah, that wording was kind of ambiguous

2011-01-09, 16:40	#2
cipher Feb 2007 211 Posts	Awesome Idea!!

2011-01-14, 14:35	#10
Ralf Recker Oct 2010 10111111₂ Posts	The PRPNet 4.1.2 client again deletes results the were not successfully reported... ...so they remain in the pending list until they time out. Last fiddled with by Ralf Recker on 2011-01-14 at 14:37

2011-01-14, 18:38	#11
Oddball May 2010 499 Posts	Congrats to Flatlander, who found the first prime in the rally!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
January 2018	Xyzzy	Puzzles	0	2018-01-02 03:09
January 2017	Xyzzy	Puzzles	4	2017-02-22 21:34
January 2015	Xyzzy	Puzzles	1	2015-02-02 17:17
LLRnet/PRPnet rally January 3rd-10th	mdettweiler	No Prime Left Behind	48	2011-01-12 10:14
My new discovery!!	Merfighters	Lounge	3	2010-04-07 11:46