mersenneforum.org > Data Strategic double and triple checks (PRP's and P-1's too)
 Register FAQ Search Today's Posts Mark Forums Read

 2020-08-01, 14:14 #12 ATH Einyen     Dec 2003 Denmark 53×23 Posts When there are no known factors type 5 is the same as type 1: a^(N*known_factors-1) vs a^(N-1) because known_factors=1 But you do not need to specify type, just let Prime95/mprime use the default type: PRP=1,2,84946391,-1 Gpuowl only does type1 still I believe, and unless the input format changed in the last few months it should be: PRP=,1,2,84946391,-1,75,0 Gpuowl need the ,75,0 at the end for some reason (unless this has changed lately): Code: ## worktodo.txt The lines in worktodo.txt must be of one of these forms: * 70100200 * PRP=FCECE568118E4626AB85ED36A9CC8D4F,1,2,77936867,-1,75,0
 2020-08-01, 14:16 #13 ATH Einyen     Dec 2003 Denmark 53×23 Posts This one needs a TC: 53301539 But it will be picked up automatically at any time probably.
 2020-08-01, 16:49 #14 ATH Einyen     Dec 2003 Denmark 53·23 Posts Version 30 has the new PRP-VDF where it only needs 1 test and a short proof validation instead of a double check. But these already has 2 PRP tests, so most likely another one will finish the double check. Try this: PRP=1,2,84946391,-1 PRP=1,2,84963899,-1 PRP=1,2,86291299,-1 PRP=1,2,86793229,-1 PRP=1,2,86793251,-1 Last fiddled with by ATH on 2020-08-01 at 16:52
2020-08-02, 00:36   #15
Uncwilly
6809 > 6502

"""""""""""""""""""
Aug 2003
101×103 Posts

836210 Posts

Quote:
 Originally Posted by ATH Version 30 has the new PRP-VDF where it only needs 1 test and a short proof validation instead of a double check.
Do you agree that as any new single LL runs that pop up as suspect, that we should just do a fresh PRP on v30? Roughly the same amount of time to run the test, but savings if a LL-TC were needed.

2020-08-02, 14:06   #16
ATH
Einyen

Dec 2003
Denmark

53·23 Posts

Quote:
 Originally Posted by Uncwilly Do you agree that as any new single LL runs that pop up as suspect, that we should just do a fresh PRP on v30? Roughly the same amount of time to run the test, but savings if a LL-TC were needed.
Yes, Suspect tests should be run with PRP VDF.
If there is 1 Suspect and 1 Unverified LL I guess if the person really wants they can run LL DC because in most cases probably >95% it will match the non-suspect, but PRP-VDF preferred.

2020-08-02, 15:43   #17
kruoli

"Oliver"
Sep 2017
Porta Westfalica, DE

229 Posts

Quote:
 Originally Posted by ATH This one needs a TC: 53301539 But it will be picked up automatically at any time probably.
Somehow, it got TF assigned without any need. The exponent is already factored one level farther than GPU72 standard.

2020-08-03, 01:04   #18
S485122

Sep 2006
Brussels, Belgium

1,549 Posts

Quote:
 Originally Posted by kruoli Somehow, it got TF assigned without any need. The exponent is already factored one level farther than GPU72 standard.
And it is in double-check CAT 1 range, it should not have been assigned for TF through the Manual Testing pages.

Jacob

Last fiddled with by S485122 on 2020-08-03 at 01:06 Reason: added bit about manual testing pages

 Similar Threads Thread Thread Starter Forum Replies Last Post Madpoo Marin's Mersenne-aries 1841 2019-07-16 03:30 GP2 Marin's Mersenne-aries 4 2019-03-16 15:31 patrik Data 5 2004-01-31 22:19 outlnder Lounge 4 2003-04-07 18:06 outlnder Lounge 6 2003-01-24 22:01

All times are UTC. The time now is 09:23.

Mon Aug 3 09:23:41 UTC 2020 up 17 days, 5:10, 0 users, load averages: 1.20, 1.23, 1.21