24 hour rally - Page 9

BlisteringSheep · 2008-02-09, 23:04

Quote:

Originally Posted by em99010pepe

It's a dynamic IP and it changed 15 min after the start of the rally, sorry, I can't control that.

No worries, just explaining why it looked like I grabbed the first result & then quit. If I had just stayed downstairs another 10 minutes I would have seen it when it happened. That'll teach me!

em99010pepe · 2008-02-09, 23:05

Quote:

Originally Posted by BlisteringSheep

I will continue these four machines at least through tomorrow. I did bump my caches up in the interim, but won't abandon any work.

When I get my new dual E5440 2.83ghz put together (hope to have all of the parts by next weekend), I may put some of it on NPLB.

I suppose by then to have Top 5000 pairs in the server. I'll keep here my machines until the server gets almost dry, then I still have to finish my manual reservations and once for all to NPLB have 320K-333.2K complete.

I'm going to sleep.

kar_bon · 2008-02-09, 23:05

i bet 28.000 k/n-pairs at the race!

em99010pepe · 2008-02-09, 23:07

I'll go for 25 000 k/n pairs. More informations for the next bet, XS team increased their output, Free-DC too.

kar_bon · 2008-02-09, 23:50

bets so far:
ET_: 35.000
kar_bon: 28.000
em99010pepe: 25.000
gd_barnes: 31.000

gd_barnes · 2008-02-10, 00:46

According to the tests that I am recieving from the server using GMT times:

time / n-value / n-completed in hour
19:00 / 327987 / 0
20:00 / 328047 / 60
21:00 / 328114 / 67
22:00 / 328205 / 91 (This is where I got 4 more cores running)
23:00 / 328302 / 97
0:00 / 328402 / 100

Then between 0:07 and 0:10, some people reserved a whole bunch (probably partially still in their queue). The n-value jumped from 328413 to 328454. Since then, we've been ripping off about 6-7 / 3 mins. = 2n / min = 120n / hour.

So assuming 2300 candidates per 100n = 11.5 tests / n:

First 5 hours: (328402 - 327987) = 415n = 4773 tests
4773 tests / 5 hours = 954 tests / hour
415n / 5 hours = 83n / hour

Assuming that the heavy hitters remain we have:
First 5 hours: 4773 tests
Last 19 hours: 120n / hour = 2280n * 11.5 tests/n = 26220 tests

And finally:
Test prediction: 4773 + 26220 =
30993 tests

Ending n-value: n-value @ 0:00 GMT = 328402 + 120n/hour * 19 remaining hours =
n=330682

It looks like we have enough reserved for LLRnet but it won't take along to finish it off after the rally.

EDIT: It looks like my 7 cores are averaging about 90-95 tests / hour after they all got running around 21:15.

Gary

em99010pepe · 2008-02-10, 11:31

Just a quick note, if the client sleeps please stop it and increase the wu cache....

ET_ · 2008-02-10, 11:34

Quote:

Originally Posted by em99010pepe

Just a quick note, if the client sleeps please stop it and increase the wu cache....

Done already.

Also remember to disable automatic updates of Microsoft and Java.
I found my mobile rebooted this morning

Luigi

em99010pepe · 2008-02-10, 11:39

Quote:

Originally Posted by ET_

Done already.

Also remember to disable automatic updates of Microsoft and Java.
I found my mobile rebooted this morning

Luigi

All that crap is off...only manual updates for me!

em99010pepe · 2008-02-10, 13:43

The server is already sending pairs from range 330K.

Mini-Geek · 2008-02-10, 13:48

Quote:

Originally Posted by em99010pepe

The server is already sending pairs from range 330K.

That's approx. 1/2 of the 320K-333.2K range reserved. The rally has been going on for about 18:45 (h:m). It would require over 37 hours at this speed to clear the range.

2008-02-09, 23:05	#91
kar_bon Mar 2006 Germany 2³×3×11² Posts	another bet i bet 28.000 k/n-pairs at the race!

2008-02-09, 23:07	#92
em99010pepe Sep 2004 B0E₁₆ Posts	I'll go for 25 000 k/n pairs. More informations for the next bet, XS team increased their output, Free-DC too. Last fiddled with by em99010pepe on 2008-02-09 at 23:09

2008-02-09, 23:50	#93
kar_bon Mar 2006 Germany 2904₁₀ Posts	24h race bets bets so far: ET_: 35.000 kar_bon: 28.000 em99010pepe: 25.000 gd_barnes: 31.000 Last fiddled with by kar_bon on 2008-02-10 at 16:48 Reason: Gary's bet

2008-02-10, 00:46	#94
gd_barnes May 2007 Kansas; USA 3³×5×7×11 Posts	According to the tests that I am recieving from the server using GMT times: time / n-value / n-completed in hour 19:00 / 327987 / 0 20:00 / 328047 / 60 21:00 / 328114 / 67 22:00 / 328205 / 91 (This is where I got 4 more cores running) 23:00 / 328302 / 97 0:00 / 328402 / 100 Then between 0:07 and 0:10, some people reserved a whole bunch (probably partially still in their queue). The n-value jumped from 328413 to 328454. Since then, we've been ripping off about 6-7 / 3 mins. = 2n / min = 120n / hour. So assuming 2300 candidates per 100n = 11.5 tests / n: First 5 hours: (328402 - 327987) = 415n = 4773 tests 4773 tests / 5 hours = 954 tests / hour 415n / 5 hours = 83n / hour Assuming that the heavy hitters remain we have: First 5 hours: 4773 tests Last 19 hours: 120n / hour = 2280n * 11.5 tests/n = 26220 tests And finally: Test prediction: 4773 + 26220 = 30993 tests *Ending n-value: n-value @ 0:00 GMT = 328402 + 120n/hour 19 remaining hours = n=330682** It looks like we have enough reserved for LLRnet but it won't take along to finish it off after the rally. EDIT: It looks like my 7 cores are averaging about 90-95 tests / hour after they all got running around 21:15. Gary Last fiddled with by gd_barnes on 2008-02-10 at 00:58

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
24 hour news	davieddy	Soap Box	4	2011-12-19 19:35
mfaktc slows on WinXP after about an hour	Christenson	GPU Computing	5	2011-05-27 21:47
Prime 95 shuts down itself after exactly 9:22 hour	Unregistered	Information & Answers	13	2009-09-02 05:11
! hour limit on editing	davieddy	Lounge	7	2009-09-01 15:57
1 buck an hour	crash893	Hardware	6	2009-06-18 01:45

2008-02-10, 11:31	#95
em99010pepe Sep 2004 2×5×283 Posts	Just a quick note, if the client sleeps please stop it and increase the wu cache....

2008-02-10, 13:43	#98
em99010pepe Sep 2004 2·5·283 Posts	The server is already sending pairs from range 330K.