TPS Debut Rally: November 15-16

[Neo]

Server is getting taxed.... (I think)
Getting error 11044.. waiting three mins now for reconnect..

Neo

[mdettweiler]

That's strange...the server should definitely not be overloaded during this rally, what with the candidates being megabit-sized. We handled much higher loads during our last rally at NPLB on smaller numbers (>30,000 candidates/day processed, versus a little under 3,000 candidates processed so far today in this rally) and didn't have any server problems, so it would be decidedly odd if the server was buckling under the load now.

I'm not sure what error 11044 means; I haven't seen it on my own clients, but then again mine haven't needed to talk to the server in over 50 minutes, so if this is a recent phenomenon I wouldn't be seeing it yet.

[odicin]

First prime found on this port ;)

user=odicin
[2011-11-16 06:44:17]
20667679305*2^1000000-1 is prime! Time : 0.0 sec.

Your client also tested 20667679305*2^1000000+1 to determine if it is a prime twin. Unfortunately, it is neither PRP nor prime.

Regards Odi

[axn]

Originally Posted by odicin

First prime found on this port ;)

user=odicin
[2011-11-16 06:44:17]
20667679305*2^1000000-1 is prime! Time : 0.0 sec.

Your client also tested 20667679305*2^1000000+1 to determine if it is a prime twin. Unfortunately, it is neither PRP nor prime.

Regards Odi

That is a reportable prime in the Top 5000 table. Congratulations.

@Oddball: I guess you'll need to issue guidelines for reporting the primes!

[Lennart]

Originally Posted by axn

That is a reportable prime in the Top 5000 table. Congratulations.

@Oddball: I guess you'll need to issue guidelines for reporting the primes!

Here are one more 24335320035*2^1000000-1Prime.

Yes what to use LLR Newpgen TPS ???


Lennart

[ellipse]

Originally Posted by Lennart

Yes what to use LLR Newpgen TPS ???

From the first post:

The candidates we'll be testing during the rally will have ~301,000 digits and will enter the top 5000 primes list at rank ~1000. If you find a prime, you'll need to submit it to the Prime Pages under a prover account that has your name, LLR, TPS, and NewPGen.

What's missing from that list is axn's sieve:
http://www.mersenneforum.org/showpos...0&postcount=33

Do we have to include it in the prover code?

[Oddball]

Originally Posted by axn

@Oddball: I guess you'll need to issue guidelines for reporting the primes!

what to use LLR Newpgen TPS ???

Yes, you'll be reporting it with your name, LLR, NewPGen, and TPS. Congrats, Lennart and odicin!

What's missing from that list is axn's sieve:
http://www.mersenneforum.org/showpos...0&postcount=33

Do we have to include it in the prover code?

No, because axn's sieve was not used until k>20T.

[Oddball]

The results for the rally are in. Lennart and the PrimeSearchTeam continued their TPS rally victory streak by finishing in the top spot for the 4th time in a row. Second place was more competitive: After day 1, odicin was ahead by over 300 tests, but Lumiukko caught up and passed him just before the rally finished. The final score was 2357 to 2237.

Overall, we found three megabit primes. The largest was 26983775475*2^1000000-1 and enters the top 5000 list at rank 1124:
http://primes.utm.edu/primes/page.php?id=103141

Once again, I'd like to thank the Primegrid admins, who let me start a TPS rally thread in their forums. Interested crunchers can try out their Leonids Challenge, which starts on the 18th and looks for primes of the form 3*2^n-1.

[Neo]

Originally Posted by mdettweiler

That's strange...the server should definitely not be overloaded during this rally, what with the candidates being megabit-sized. We handled much higher loads during our last rally at NPLB on smaller numbers (>30,000 candidates/day processed, versus a little under 3,000 candidates processed so far today in this rally) and didn't have any server problems, so it would be decidedly odd if the server was buckling under the load now.

I'm not sure what error 11044 means; I haven't seen it on my own clients, but then again mine haven't needed to talk to the server in over 50 minutes, so if this is a recent phenomenon I wouldn't be seeing it yet.

Probably just a large w/u request from Lennart's farm :)

Neo