mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Software (https://www.mersenneforum.org/forumdisplay.php?f=10)
-   -   P-1 B1/B2 selection with "Test=" vs "Pfactor=" (https://www.mersenneforum.org/showthread.php?t=3889)

James Heinrich 2005-03-19 20:55

P-1 B1/B2 selection with "Test=" vs "Pfactor="
 
Why is it that the B1/B2 bounds selected are so vastly different with the same exponent on the same machine if the P-1 test is invoked with these two different lines?[code]Pfactor=79299821,72,1

Test=79299821,72[/code]The Pfactor line gives[quote]B1=395000,B2=8295000, chance = 3.84%[/quote]
The Test line gives[quote]B1=835000,B2=21918750, chance = 5.13%[/quote]
Why would Prime95 select such vastly-different bounds for the two testing methods? Whether a LL is going to be performed immediately afterwards shouldn't make any difference... should it?

garo 2005-03-19 21:32

Try

Pfactor=79299821,72,0

The answer lies in whether a factor would save one test or two.

James Heinrich 2005-03-19 21:58

Indeed...
Pfactor=79299821,72,0 gives the same result as Test=79299821,72

Now I better understand the documentation:[quote]Pfactor=exponent,how_far_factored,has_been_LL_tested_once[/quote]


All times are UTC. The time now is 17:35.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.