mersenneforum.org What are your per-iteration times?
 Register FAQ Search Today's Posts Mark Forums Read

 2011-07-03, 00:19 #1 LiquidNitrogen     Jun 2011 Henlopen Acres, Delaware 7·19 Posts What are your per-iteration times? I am just curious about what some of your per-iteration times might be. M 46,878,773 = 0.024 for me.
2011-07-03, 00:41   #2
science_man_88

"Forget I exist"
Jul 2009
Dumbassville

26×131 Posts

Quote:
 Originally Posted by LiquidNitrogen I am just curious about what some of your per-iteration times might be. M 46,878,773 = 0.024 for me.
I'm doing 2 double checks of ll I think and my minimum I can see so far is .078 I'm not sure if they are double checks or not.

Last fiddled with by science_man_88 on 2011-07-03 at 00:42

 2011-07-03, 02:33 #3 Christenson     Dec 2010 Monticello 5×359 Posts 30-40msec on M24,262,141....Liquid Nitrogen has a relatively "hot" machine.....
 2011-07-03, 06:07 #4 imwithid     Apr 2009 Venice, Chased by Jaws 3·29 Posts My machines are a bit old, but out of curiosity for what others are able to output, I'll post what I can. I hope this is not simply a low brow "mine is bigger than yours" thread. In today's age of computers, your unit will be blown away eventually by what I replace it with when the time comes. I had four machines (now three, since two weeks ago - one TF and now P-1 for GIMPS, the rest SoB). Three are Intel Core 2 class. Depending on the system's resource allocation and relative "cleanliness", I get: Core2 6300 @ 1.86: LL ~0.059 19913348, 19913348 (moved now to GIMPS P-1 & LL - 47907107 - timings, soon to come) Core2 4300 @ 1.80: LL ~0.080 19880194. Not currently running (too hot, even with a laptop fan cooler - used as a heater during winter -- thanks HP!): AMD Mobile K10 X2 RM-72: LL ~.110 19913348.
 2011-07-03, 08:21 #5 lycorn     Sep 2002 Oeiras, Portugal 23·181 Posts M28865407 (1536K FFT) -> 0.010 - 0.011 msec/iter (using two cores of a slightly OCed i5-750 - timing fluctuations are due to variations in the type of work done by the other two cores: e.g., if I use both of them to P-1 testing, I get 0.011 msec/iter; if I switch one of them to TF I normally get 0.010.
2011-07-03, 15:08   #6
LiquidNitrogen

Jun 2011
Henlopen Acres, Delaware

2058 Posts

Quote:
 Originally Posted by Christenson 30-40msec on M24,262,141....Liquid Nitrogen has a relatively "hot" machine.....
Well it runs mostly at about 58 C, so maybe not too hot.

:)

It's an overclocked i7-870 @ 3.9 GHz. This is my "spare system," it has single core speed that is slightly faster than a stock Gulftown (i7-980X @ 3.33 GHZ)

2011-07-03, 15:13   #7
LiquidNitrogen

Jun 2011
Henlopen Acres, Delaware

100001012 Posts

Quote:
 Originally Posted by imwithid I hope this is not simply a low brow "mine is bigger than yours" thread.
No, I was just curious. I noticed for smaller numbers the time is much faster, so I am just trying to get a sense of how fast/slow other people are iterating for numbers of various sizes.

Quote:
 Originally Posted by imwithid In today's age of computers, your unit will be blown away eventually by what I replace it with when the time comes.
Ummm.... I don't think so. I own a 5.27 GHz i7-980X that is cooled to -40 using a Single Stage Vapor Phase Change unit. Message me here privately and I will send a link to its YouTube video.

It will be faster than whatever you can buy until the year 2018 or maybe longer.

:)

 2011-07-03, 16:36 #8 ATH Einyen     Dec 2003 Denmark 57278 Posts Use "PercentPrecision=6" in prime.txt then you get maximum precision and decimals on the milliseconds.
2011-07-09, 19:36   #9
LiquidNitrogen

Jun 2011
Henlopen Acres, Delaware

7×19 Posts

Quote:
 Originally Posted by ATH Use "PercentPrecision=6" in prime.txt then you get maximum precision and decimals on the milliseconds.
I did do this, and the percent precision is now 6 decimal places, but not the per iteration time.

 2011-07-09, 20:58 #10 Batalov     "Serge" Mar 2008 Phi(4,2^7658614+1)/2 2×4,663 Posts TimingOutput=4 in prime.txt
 2011-07-10, 12:45 #11 ATH Einyen     Dec 2003 Denmark BD716 Posts Yeah sorry I forgot that there are 2 settings for precision, one for progress and one for timings. I have had those on since I started GIMPS. I should have checked undoc.txt before posting: Code: You can adjust how many decimal digits of precision are displayed in lines output to the screen. In prime.txt enter a value between 0 and 6: PercentPrecision=n You can alter the way the program outputs timings. In prime.txt set: TimingOutput=n Where n=1 for seconds with 3 digits of precision, n=2 for milliseconds with 1 digit of precision, n=3 for milliseconds with 2 digits of precision, n=4 for milliseconds with 3 digit of precision. Last fiddled with by ATH on 2011-07-10 at 12:52

 Similar Threads Thread Thread Starter Forum Replies Last Post Jud McCranie Information & Answers 53 2013-08-17 19:09 rx7350 Hardware 12 2006-05-08 21:54 SlashDude 15k Search 0 2004-01-28 05:47 PLeopard Hardware 9 2003-10-29 05:48 smoffat Software 13 2003-10-22 22:50

All times are UTC. The time now is 19:22.

Fri Feb 26 19:22:20 UTC 2021 up 85 days, 15:33, 0 users, load averages: 2.11, 1.87, 2.08