[QUOTE=James Heinrich;555666]By my calculations a single curve of those bounds should be in the order of 1-2 GHz-days. It's all the [i]other[/i] curves that didn't find a factor that make it such a long/hard process. :smile:[/QUOTE]

True. I wonder why that effort was not reported to Primenet and only one curve is mentioned.

[QUOTE=lycorn;555709]True. I wonder why that effort was not reported to Primenet and only one curve is mentioned.[/QUOTE]On this current effort I'm not sure. It's not impossible that he just got lucky and found the factor in the first curve he attempted at those bounds. Ryan has previously run a large number of curves on M2137, most recently reported on 2019-11-19:[code]M2137 completed 8596 ECM curves, B1=110000000, B2=776278396540
M2137 completed 19664 ECM curves, B1=260000000, B2=3178559884516
M2137 completed 348 ECM curves, B1=850000000, B2=14899382397918[/code]Remember that NF-ECM results are hidden on the exponent report page by default, you need to check the checkbox if you want to [url=https://www.mersenne.org/report_exponent/?exp_lo=2137&exp_hi=&full=1&ecmhist=1]see them[/url].

[QUOTE=lycorn;555709]True. I wonder why that effort was not reported to Primenet and only one curve is mentioned.[/QUOTE]

Probably an artifact of his batch methods whereby prime95 does stage 1 and GMP-ECM does stage 2. This big factor was found in stage 1!

[QUOTE=Prime95;555723]Probably an artifact of his batch methods whereby prime95 does stage 1 and GMP-ECM does stage 2. This big factor was found in stage 1![/QUOTE]

I actually use just GMP-ECM, but with gwnum compiled/linked in, which makes stage 1 faster for b^n-1 numbers...

Actually, this factor was found in stage 2:

[CODE]GMP-ECM 7.0.5-dev [configured with GMP 6.2.0, GWNUM 29.8, --enable-asm-redc, --enable-assert] [ECM]
Due to incompatible licenses, this binary file must not be distributed.
Input number is 2^2137-1 (644 digits)
Using B1=2900000000, B2=81712898767516, polynomial Dickson(30), sigma=0:8561132775016161148
Step 1 took 20559097ms
Step 2 took 12655697ms
********** Factor found in step 2:
434527865148151913428610180914321766584011558417928142522774921
Found prime factor of 63 digits:
434527865148151913428610180914321766584011558417928142522774921
Composite cofactor (2^2137-1)/434527865148151913428610180914321766584011558417928142522774921 has 581 digits[/CODE]

[QUOTE=ryanp;555737]Actually, this factor was found in stage 2[/QUOTE]How do you generate the result line you submitted from that? Do you just hand-craft those when you find a factor?[quote]M2137 has a factor: 434527865148151913428610180914321766584011558417928142522774921 (ECM curve 1, B1=2900000000)[/quote]I have adjusted the mersenne.ca record to include B2 and Sigma, perhaps if there are future similar factors found you could ensure the B2 and Sigma are included in the submitted result line, like this:[quote]M2137 has a factor: 434527865148151913428610180914321766584011558417928142522774921 (ECM curve 1, B1=2900000000, B2=81712898767516, Sigma=8561132775016161148)[/quote]

So it actually looks like only one curve was run.
It´s an amazing strike of luck to find a factor this size on curve #1...

Another member of 9-factor community
 

[M]M9532331[/M] has a 87.885 bit factor: [url=https://www.mersenne.ca/M9532331]285749678795460545168553689[/url]

I will try to push it to 10 factors.

[QUOTE=lycorn;555796]So it actually looks like only one curve was run.
It´s an amazing strike of luck to find a factor this size on curve #1...[/QUOTE]

No, there were many more curves run. I just report them separately (offline) to George, from time to time...

[QUOTE=James Heinrich;555756]perhaps if there are future similar factors found you could ensure the B2 and Sigma are included in the submitted result line, like this:[/QUOTE]Thank you Ryan! :smile:
[quote][m]M1999[/m] has a factor: 7452018296729329082588085050101877014364039300287 (ECM curve 1, B1=850000000, B2=15892628251516, Sigma=3085684570545921953)[/quote]

[QUOTE=James Heinrich;555804]Thank you Ryan! :smile:
[QUOTE][M]1999[/M] has a factor:[/QUOTE][/QUOTE]
So, we have factors now totaling ~1/3 the total number of digits of the number (197/602). Very nice.

and factordb tell us it is FF [url]http://factordb.com/index.php?query=M1999[/url]