[QUOTE=jyb;431850]As part of preparing an ECMnet server to coordinate work on these numbers, I'm trying to gather information on ECM work done up to now. Can anybody who has done any non-trivial work since Paul posted the extensions on April 1 please let me know, either here or by PM?

I'm looking for which composites you attacked, the B1 value, and a rough number of curves. Any work done on numbers which are now fully factored need not be included.

I'm particularly interested in hearing from the following people, since you'll all done enough work to actually find factors via ECM:

C Pinho
L Morelli
S Batalov
G Childers
D Domanov

Thanks![/QUOTE]

I ran about 25/30 batches of 512 curves in parallel with GPU-ECM, I think it was B1=1,000,000 or 11,000,000, of the big ones, and found one or two factors. I didn't gather the results, sorry...

Luigi

5^437-4^437

[CODE]Using B1=11000000, B2=35133391030, polynomial Dickson(12), sigma=1889940718
Step 1 took 133103ms
Step 2 took 40753ms
********** Factor found in step 2: 64614673910347063423627202909052942802061
Found probable prime factor of 41 digits: 64614673910347063423627202909052942802061
Composite cofactor <...> has 229 digits
[/CODE]

5^403-4^403

[CODE]Using B1=11000000, B2=35133391030, polynomial Dickson(12), sigma=302519623
Step 1 took 95581ms
Step 2 took 30500ms
********** Factor found in step 2: 6379707193081308369849821852840781520711
Found probable prime factor of 40 digits: 6379707193081308369849821852840781520711
Composite cofactor <...> has 158 digits
[/CODE]

ECMNET server set up
 

Following a great deal of work by jyb, we've been able to set up an ECMNET server loaded with the extension tables. Tasks are presently being allocated at the B1=11M (or t45) level. Please note this is a V2 server and it will not talk properly to V3 clients.

The server address and port number is 109.109.168.71:8195 and please be careful to choose the non-standard port. The standard 8194 connects to a GCW server.

One caveat: factors have been arriving in profusion recently --- 114 in the last ten days alone --- and I may not be able to correct the server's database more than every day or few. Consequently, there is a risk of the server allocating already factored numbers. I'll try to minimize that risk.

Also: the servers fail to send out email when factors come in. Something changed at my ISP a couple of years back and I've not yet managed to compensate.

Paul

Thank you Paul. 7 cores pointed to the server.

3^564+2^564

Nice split:
[CODE]P54 = 450383239921051560226333500267533413789695831405334657
P54 = 524201780540025413375887737691449576421975079035606753
P58 = 6965198410039886089733861470998879126307716578291765438689[/CODE]

[2016-04-22 18:02:03 GMT] a: Factor found! 11+2_261 / (prime) 10464235002192262098539087136817977907 B1: 11000000 sigma: 1 (found in step 2)
[2016-04-22 18:02:03 GMT] a: Co-factor: 11+2_261 / (Prime) 7104976982075547754907721920149770526378916139391985702098612373306192300507545659109988876000701995289576847862345767929

[QUOTE=pinhodecarlos;432254][2016-04-22 18:02:03 GMT] a: Factor found! 11+2_261 / (prime) 10464235002192262098539087136817977907 B1: 11000000 sigma: 1 (found in step 2)
[2016-04-22 18:02:03 GMT] a: Co-factor: 11+2_261 / (Prime) 7104976982075547754907721920149770526378916139391985702098612373306192300507545659109988876000701995289576847862345767929[/QUOTE]
Confirmed.

[code][2016-04-22 17:59:13 GMT] Connection from 86.7.53.62
[2016-04-22 17:59:13 GMT] pinhodecarlos@yahoo.com (home_05) at 86.7.53.62: Received 11+2_261 : 24 ECM curves with B1=11000000
[2016-04-22 17:59:13 GMT] 11+2_261: prime factor returned by pinhodecarlos@yahoo.com (home_05)! Factor=10464235002192262098539087136817977907 Method=ECM B1=11000000 Sigma=1
[2016-04-22 17:59:13 GMT] 11+2_261: Prime factor returned by pinhodecarlos@yahoo.com (home_05)! Factor=7104976982075547754907721920149770526378916139391985702098612373306192300507545659109988876000701995289576847862345767929 Method=ECM B1=11000000 Sigma=1
[2016-04-22 17:59:13 GMT] SMTP server error. Expected 220, got 0
[2016-04-22 17:59:13 GMT] Total Time: 9:53:18 Curves: 5784 Total Work: 1539559 Total Factors Found: 1
[/code]

according to the server output. Note that automagic email failed :sad:

It seems that the ecmserver / ecmclient parser codes should be made aware of the new formats of reporting sigma (those with ":"); it appears that the parser discarded the value behind the colon. Both client and server side show simply "1".

The core finder was fired 1h30 ago. Lucky.

PS(Found again on another core...lol)

[QUOTE=Batalov;432257]It seems that the ecmserver / ecmclient parser codes should be made aware of the new formats of reporting sigma (those with ":"); it appears that the parser discarded the value behind the colon. Both client and server side show simply "1".[/QUOTE]

Indeed. Can you show me a sample line that includes the new format?

Changing the code for both client and server is easy enough to do, but deploying it is harder. Everyone would have to update their client code to see any change. Changing it on the server alone won't do anything, since the server just receives the sigma value from the client.