Automatic submit results + fetch assignments for mfaktc?
 

I'm running it on my GTX1080 when i'm not using it, getting around 980ghz/days at the moment, which means 12 assignments takes about an hour

Is there anything I can do besides request a whole bunch that'll let it run automatically?

Are you using Windows or Linux?

[QUOTE=DuskFalls;454108]I'm running it on my GTX1080 when i'm not using it, getting around 980ghz/days at the moment, which means 12 assignments takes about an hour

Is there anything I can do besides request a whole bunch that'll let it run automatically?[/QUOTE]
If you are running Windows, [URL="http://www.mersenneforum.org/showthread.php?t=17915"]Misfit [/URL]will get and submit assignments. There are other means in Linux, but I am not as familiar with them.

For Linux, there's the script [url=https://github.com/MarkRose/primetools/blob/master/mfloop.py]I use[/url].

[QUOTE=DuskFalls;454108]I'm running it on my GTX1080 when i'm not using it, getting around 980ghz/days at the moment, which means 12 assignments takes about an hour

Is there anything I can do besides request a whole bunch that'll let it run automatically?[/QUOTE]

If sticking with manual methods, you could follow the suggestion in the Mfaktc readme and increase the bit depth to full prime95 depth. Or go even higher; James Heinrich has lots of CUDALucas/Mfaktc factor depth curves versus exponent and per GPU model. [URL]http://www.mersenne.ca/cudalucas.php?model=683[/URL] gtx1070
Note though that the entire stack of bit levels per exponent should be reported together to avoid reassignment and duplication of effort.

Or let Mfaktc run out occasionally, and have CUDALucas chew on some LL test or double check exponent a while. Use a batch file to alternate them. A 90-100M exponent should keep a GTX1080 occupied for about a week.

auto work get or submit results
 

See also the pdf attached at
[URL]http://www.mersenneforum.org/showpost.php?p=465293&postcount=24[/URL]