mersenneforum.org TF times
 Register FAQ Search Today's Posts Mark Forums Read

 2012-04-13, 13:16 #1 Unregistered   2×3×61 Posts TF times Hi, Why is it that TF M805085069 from 65 to 66 bits takes 4 minutes but TF M101108803 from 65 to 66 bits takes half an hour on the same CPU?
 2012-04-13, 14:06 #2 gjmccrac     Aug 2009 Ontario, Canada 100000012 Posts As a number gets larger there are fewer possible factors to test in a range. A potential factor has the form 2*p*k+1. This means when comparing p=805085069 and p=101108803 you would expect that there are approx 7.9 times (805085069/101108803) more factors to test in the 100M number vs the 800M number. This means the time would be approx 7.9 times longer. 4 minutes * 7.9 is approx 31 minutes Grant.
2012-04-13, 14:06   #3
bcp19

Oct 2011

7×97 Posts

Quote:
 Originally Posted by Unregistered Hi, Why is it that TF M805085069 from 65 to 66 bits takes 4 minutes but TF M101108803 from 65 to 66 bits takes half an hour on the same CPU?
It sounds like there is a process running in the background causing a slowdown during your testing, as the time to run a 100M exp to a given bit level will always take less time than 80M exponent.

 2012-04-13, 14:13 #4 Unregistered   3×7×347 Posts Thanks a lot gjmccrac, this explains it.
2012-04-13, 14:53   #5
petrw1
1976 Toyota Corona years forever!

"Wayne"
Nov 2006

41×107 Posts

Quote:
 Originally Posted by bcp19 It sounds like there is a process running in the background causing a slowdown during your testing, as the time to run a 100M exp to a given bit level will always take less time than 80M exponent.
800M

2012-04-13, 19:23   #6
science_man_88

"Forget I exist"
Jul 2009
Dumbassville

20C016 Posts

Quote:
 Originally Posted by gjmccrac As a number gets larger there are fewer possible factors to test in a range. A potential factor has the form 2*p*k+1. This means when comparing p=805085069 and p=101108803 you would expect that there are approx 7.9 times (805085069/101108803) more factors to test in the 100M number vs the 800M number. This means the time would be approx 7.9 times longer. 4 minutes * 7.9 is approx 31 minutes Grant.
interesting that means the first one doable in under a second for that bit range should be around 192B

2012-04-13, 19:40   #7
bcp19

Oct 2011

7·97 Posts

Quote:
 Originally Posted by petrw1 800M
I should have seen that :/ Sometimes the eyes just don't work right.

 2012-04-14, 11:45 #8 Chuck     May 2011 Orange Park, FL 86510 Posts It's hard to read these; it would help to put commas in. M805,085,069

 Similar Threads Thread Thread Starter Forum Replies Last Post carpetpool Information & Answers 12 2018-03-24 08:48 LiquidNitrogen Hardware 22 2011-07-12 23:15 roger Information & Answers 1 2007-04-25 14:35 masser Sierpinski/Riesel Base 5 6 2007-03-05 18:17 lpmurray Hardware 2 2007-02-17 19:53

All times are UTC. The time now is 05:18.

Sat Sep 26 05:18:41 UTC 2020 up 16 days, 2:29, 0 users, load averages: 2.24, 2.01, 1.77