mersenneforum.org > Data Deep dive TF
 Register FAQ Search Today's Posts Mark Forums Read

 2019-03-30, 15:46 #23 dcheuk     Jan 2019 Iowa, US 3·71 Posts Is there any way to get assignments as well as ask mfaktc to keep trial factoring until 72bit or until factor found? It gets tedious after a while trying to copying and pasting 1000 lines of assignments every day onto worktodo.txt. Thanks much appreciated.
2019-03-30, 16:04   #24
petrw1
1976 Toyota Corona years forever!

"Wayne"
Nov 2006

2·5·7·61 Posts

Quote:
 Originally Posted by dcheuk Is there any way to get assignments as well as ask mfaktc to keep trial factoring until 72bit or until factor found? It gets tedious after a while trying to copying and pasting 1000 lines of assignments every day onto worktodo.txt. Thanks much appreciated.
Unless I misunderstand you:

1. Assuming you are using misfit there is an option under
Operations...
Configuration Editor...
to request the final bit level.

2. You can edit you worktodo.txt file so the assignments look like:
Factor=xxx,<exponent>,xx,72

3. In mfaktc.cfg
StopAfterFactor=1

2019-03-30, 17:08   #25
kriesel

"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

66018 Posts

Quote:
 Originally Posted by dcheuk Is there any way to get assignments as well as ask mfaktc to keep trial factoring until 72bit or until factor found? It gets tedious after a while trying to copying and pasting 1000 lines of assignments every day onto worktodo.txt. Thanks much appreciated.
https://www.mersenne.org/manual_gpu_assignment/
see "optional bit level to factor to"
and also "work preference" (highest bit levels or lowest exponents can help)
TF computing effort is roughly proportional to 2bitlevel/exponent.
For the same bit level, TF on a 93M exponent will run about 10x as long as 930M. It's also useful much sooner (decades or centuries) for the P-1 and primality testing wavefront.

in mfaktc.ini, set this section as you like:
Code:
# Allow to split an assignment into multiple bit ranges.
# 0 = disabled
# 1 = enabled
# Enabled Stages make only sense when StopAfterFactor is 1 or 2.
# Do not change this in the middle of a run which spans over multiple
# bitlevels, in this case mfaktc will ignore the checkpoint file and
# restarts from the beginning.
#
# Default: Stages=1

Stages=1

# possible values for StopAfterFactor:
# 0: Do not stop the current assignment after a factor was found.
# 1: When a factor was found for the current assignment stop after the
#    current bitlevel. This makes only sense when Stages is enabled.
# 2: When a factor was found for the current assignment stop after the
#    current class.
#
# Default: StopAfterFactor=1

StopAfterFactor=1
You may want StopAfterFactor=2, for the earliest exit upon factor found. Computing credit is not given for the following classes in a bit level when StopAfterFactor=1 and a factor is found.

Last fiddled with by kriesel on 2019-03-30 at 17:15

2019-03-30, 17:51   #26
dcheuk

Jan 2019
Iowa, US

D516 Posts

Quote:
 Originally Posted by petrw1 Unless I misunderstand you: 1. Assuming you are using misfit there is an option under Operations... Configuration Editor... to request the final bit level. 2. You can edit you worktodo.txt file so the assignments look like: Factor=xxx,,xx,72 3. In mfaktc.cfg StopAfterFactor=1
Thanks for pointing it up! I haven't heard of the misfit until now.

Quote:
 Originally Posted by kriesel https://www.mersenne.org/manual_gpu_assignment/ see "optional bit level to factor to" and also "work preference" (highest bit levels or lowest exponents can help) TF computing effort is roughly proportional to 2bitlevel/exponent. For the same bit level, TF on a 93M exponent will run about 10x as long as 930M. It's also useful much sooner (decades or centuries) for the P-1 and primality testing wavefront. in mfaktc.ini, set this section as you like: Code: # Allow to split an assignment into multiple bit ranges. # 0 = disabled # 1 = enabled # Enabled Stages make only sense when StopAfterFactor is 1 or 2. # Do not change this in the middle of a run which spans over multiple # bitlevels, in this case mfaktc will ignore the checkpoint file and # restarts from the beginning. # # Default: Stages=1 Stages=1 # possible values for StopAfterFactor: # 0: Do not stop the current assignment after a factor was found. # 1: When a factor was found for the current assignment stop after the # current bitlevel. This makes only sense when Stages is enabled. # 2: When a factor was found for the current assignment stop after the # current class. # # Default: StopAfterFactor=1 StopAfterFactor=1 You may want StopAfterFactor=2, for the earliest exit upon factor found. Computing credit is not given for the following classes in a bit level when StopAfterFactor=1 and a factor is found.
Again, thanks for the detailed explanation and response. I appreciate your time.

Also, thanks for sharing the link.

I feel ashamed I have been running mfaktc blindly the entire time without reading anything i.e. readme.txt.

This would make exponents that has no factors up to 72bit available quickier for p-1 testing (or eliminate them quicker thru factoring up to 72bit), meanwhile saving me some time for copying and pasting assignments/results.

Thanks again. Gonna be doing things this way after finishing my current thousands of assignments lol.

Last fiddled with by dcheuk on 2019-03-30 at 17:52

2019-03-30, 18:55   #27
kriesel

"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

3,457 Posts

Quote:
 Originally Posted by dcheuk Thanks for pointing it up! I haven't heard of the misfit until now. Again, thanks for the detailed explanation and response. I appreciate your time. Also, thanks for sharing the link. I feel ashamed I have been running mfaktc blindly the entire time without reading anything i.e. readme.txt. This would make exponents that has no factors up to 72bit available quickier for p-1 testing (or eliminate them quicker thru factoring up to 72bit), meanwhile saving me some time for copying and pasting assignments/results. Thanks again. Gonna be doing things this way after finishing my current thousands of assignments lol.
Why not go to 76? That's how high gpu TF gets assigned for current ~93M TF assignments.

There's an expression, ignorance is highly curable. Read up some. The readme files and the comments in ini files are pretty informative.
There's also lots of reference info I've assembled at https://www.mersenneforum.org/forumdisplay.php?f=154. Each reference thread's first post contains an index. (Discussion threads do not.) I try to provide there info that is not duplicated in the readme or ini files.

Last fiddled with by kriesel on 2019-03-30 at 18:57

2019-04-01, 02:45   #28
dcheuk

Jan 2019
Iowa, US

3×71 Posts

Quote:
 Originally Posted by kriesel Why not go to 76? That's how high gpu TF gets assigned for current ~93M TF assignments. There's an expression, ignorance is highly curable. Read up some. The readme files and the comments in ini files are pretty informative. There's also lots of reference info I've assembled at https://www.mersenneforum.org/forumdisplay.php?f=154. Each reference thread's first post contains an index. (Discussion threads do not.) I try to provide there info that is not duplicated in the readme or ini files.
I tried to request one assignment with optional bit level to 76. The page took a long time to respond (still loading).

Did I do anything wrong lol?

EDIT: Tried leaving everything blank. Both tries returned the below error code.

Code:
Error code: 40
Error text: No assignment available for GPU trial factoring, cpu_id: xxxxxxx, user_id = xxxxxx

Last fiddled with by dcheuk on 2019-04-01 at 02:50

 Similar Threads Thread Thread Starter Forum Replies Last Post cheesehead Science & Technology 47 2014-12-14 13:45 diep Math 5 2012-10-05 17:44 MercPrime Software 22 2009-01-13 20:10 lavalamp Open Projects 53 2008-12-01 03:59 ixfd64 Lounge 5 2005-07-06 13:46

All times are UTC. The time now is 22:51.

Sat Apr 4 22:51:40 UTC 2020 up 10 days, 20:24, 0 users, load averages: 1.48, 1.24, 1.33