mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   PrimeNet (https://www.mersenneforum.org/forumdisplay.php?f=11)
-   -   Resuming L-L with a clean hard drive (https://www.mersenneforum.org/showthread.php?t=3492)

Unregistered 2004-12-31 23:51

Resuming L-L with a clean hard drive
 
I was peacably testing 2070457. According the assignment file, Gimps checked my progress about 4:10 yesterday morning and found I was at iteration 2 million +. When I came to the computer after breakfast, my hard drive was dead (I trust entirely coincidentally!). Anyway I have a new hard drive installed.
How do resume testing my number, preferably not at the beginning?
Thanks!

Xyzzy 2005-01-01 00:46

I hope you have a backup!

This is one reason I use "InterimFiles" every 1e6 iterations... I then back up the interim files periodically...

[quote]You can have the program generate save files every n iterations. The files will have a .XXX extension where XXX equals the current iteration divided by n. In prime.ini enter:

InterimFiles=n[/quote]
If you have a backup there will be a regular save file and a backup save file as well...

Worst case you can restart your work by recreating the worktodo.ini file...

Xyzzy 2005-01-01 00:59

Assuming this is you:

[code]27070457 68 2056703 11.3 37.2 86.2 30-Dec-04 04:48 20-Dec-04 17:22 danjmi CA3C1D458[/code]
Then (I think!) you want this in your worktodo.ini file:

[code]Test=27070457,68,0[/code]

markr 2005-01-01 01:25

Mmm, the boundary between 67 & 68 is 28130000, so the line in your worktodo.ini should be [CODE]Test=27070457,67,0[/CODE]

Mike 2005-01-01 04:24

Not necessarly... The exponent might have been factored up to 2^68, therefore it's better to stick to what Primenet is telling us, which is 68.

markr 2005-01-01 05:46

Unfortunately Primenet is misleading us. The user report or status.txt "fact bits" is one higher than the TF level if a p-1 test has been done. I'm pretty sure I read some post saying that it actually adds 0.5 internally, and then the report rounds up... Anyway, for the machine I'm on now, the worktodo.ini has

[FONT=Courier New]Test=17813597,65,1[/FONT]

but my user report has

[FONT=Courier New]17813597 66 8958364 67.9 12.0 72.0 29-Dec-04 04:27 22-Oct-04 06:14 new[/FONT]

Also, the latest nofactor file has no 27M exponents TF'd above 2^67, although it's not as current as a primenet report.

It won't affect anything anyway, since the p-1 step has been done.

garo 2005-01-05 14:16

So the correct worktodo entry is:
Test=27070457,67,1


All times are UTC. The time now is 11:22.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.