MasterPrime 2006 Project - Page 29

jinydu · 2006-12-31, 13:06

Quote:

Originally Posted by Andi47

I'll do the p-1 on a Pentium 4 3400 MHz (which is running approx. 8-12 hours / day) - reserved 1800 MB RAM for P95

How long do you think it will take to finish the factoring (assuming that no factor is found)?

I have a computer that might be suitable for a LL test: My sister's desktop with a Core 2 Duo T7400. Admittedly though, it is only on about 14 hours a day, and when it is on, it is slowed down by my sister's listening to music.

StarQwest: Do you know how I can get an estimate for how long the test would take without abandoning the current tests?

Andi47 · 2006-12-31, 13:37

Quote:

Originally Posted by jinydu

How long do you think it will take to finish the factoring (assuming that no factor is found)?

That depends on how much I use my PC for gaming.
I have entered "4" in my "hours this program will run per day" which is the *very* minimum time P95 is running per day. With this assumption Prime95 estimates that the P-1 factoring will be finished around Jan 20th.

Hence P95 runs more than 4 hours / day at most days, and I will be back to work on Jan 8th (and then P95 will run approx. 12-14 h / day), I guess P-1 will be finished a few days earlier. I will keep you posted.

Edit: After an hour of self test, P95 has now started P-1:

B1 = 1095000
B2 = 29838750
Chance of finding a factor = 5,82%

StarQwest · 2006-12-31, 13:48

Quote:

Originally Posted by jinydu

StarQwest: Do you know how I can get an estimate for how long the test would take without abandoning the current tests?

It will take the same amount of time as M100,000,007, which is 79 CPU years. This is equivalent to ten tests in the M37,000,000 range, so it will take ten times longer than one of these. I would estimate five months on a Core2 Duo processor running 24 hours a day, or 9 months running 12 - 14 hours a day. You should average about 0.125 seconds per iteration.

jinydu · 2006-12-31, 14:12

Well I would have to make a final decision on whether or not to run the test soon. I will be going back to university on Jan. 6, and will not return until summer. Once I leave, I will not be able to look over the computer in person.

Using the Time option in the Advanced menu, I ran 500 iterations of M99,999,999 and got an average iteration time of about 0.274s. Notably, this was with my sister's music programs on, but inactive. This is not much better than StarQwest's iteration time. Taking into account that my sister's computer isn't on anywhere near 24 hours a day, the test could take well over a year.

An alternative would be to test M100,000,039 on my dad's office computer, a 3.59GHz Pentium 4. But I would rather not do this; since the Pentium 4 is a single core, my rise through the rankings would stagnate (or even fall) during the test.

If only I had a mostly idle Core 2 Duo that was on 24 hours a day...

Yxine · 2007-01-01, 09:07

M100.000.007 is 66.4% doublechecked. Happy new year!

Andi47 · 2007-01-01, 09:18

P-1 stage 1 on M100.000.039 is 11,44% complete; estimated completion date is Jan. 21st.

rx7350 · 2007-01-01, 17:34

[QUOTE= From: Andi47: If only I had a mostly idle Core 2 Duo that was on 24 hours a day...[/QUOTE]

I just built two Core2 Duo pcs, one of them (E6400 oc'ed 30%) to run Prime95 exclusively, so I'm offering to LL test 100000039 if need be.

Andi47 · 2007-01-01, 20:16

Quote:

Originally Posted by rx7350

[ quote= From: Andi47: If only I had a mostly idle Core 2 Duo that was on 24 hours a day...[ /QUOTE]

This quote was not from me, but from jinydu.

Quote:

I just built two Core2 Duo pcs, one of them (E6400 oc'ed 30%) to run Prime95 exclusively, so I'm offering to LL test 100000039 if need be.

30% oc'ed? I would rather not use this node to LL an exponent of THAT size. (are you sure that it runs 100,0000% stable?)

P.S.: P-1 stage 1 is at 20,06%.

Yxine · 2007-01-02, 09:03

I think I know why 100.000.007 is shown to be double-checked... But this results are not right anyway... George Waltman also seems to be anavailbable, coz I cinatcted him a week ago.

jinydu · 2007-01-02, 13:22

Quote:

Originally Posted by Yxine

I think I know why 100.000.007 is shown to be double-checked

So what do you think the reason is?

moo · 2007-01-03, 06:36

what about mp2006 the program or has that died at rc1...

2006-12-31, 14:12	#312
jinydu Dec 2003 Hopefully Near M48 2×3×293 Posts	Well I would have to make a final decision on whether or not to run the test soon. I will be going back to university on Jan. 6, and will not return until summer. Once I leave, I will not be able to look over the computer in person. Using the Time option in the Advanced menu, I ran 500 iterations of M99,999,999 and got an average iteration time of about 0.274s. Notably, this was with my sister's music programs on, but inactive. This is not much better than StarQwest's iteration time. Taking into account that my sister's computer isn't on anywhere near 24 hours a day, the test could take well over a year. An alternative would be to test M100,000,039 on my dad's office computer, a 3.59GHz Pentium 4. But I would rather not do this; since the Pentium 4 is a single core, my rise through the rankings would stagnate (or even fall) during the test. If only I had a mostly idle Core 2 Duo that was on 24 hours a day... Last fiddled with by jinydu on 2006-12-31 at 14:14

2007-01-01, 17:34	#315
rx7350 Feb 2006 AR, US 2⁴×3² Posts	[QUOTE= From: Andi47: If only I had a mostly idle Core 2 Duo that was on 24 hours a day...[/QUOTE] I just built two Core2 Duo pcs, one of them (E6400 oc'ed 30%) to run Prime95 exclusively, so I'm offering to LL test 100000039 if need be. Last fiddled with by rx7350 on 2007-01-01 at 17:36

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
k*2^n-1 primes in 2006	Kosmaj	Riesel Prime Search	2	2007-01-02 04:31
Supercomputing 2006	ixfd64	Lounge	1	2006-11-30 23:34
2006	R. Gerbicz	Puzzles	45	2006-09-19 19:32
2006 Millennium Tech. Prize.	mfgoode	Science & Technology	0	2006-06-17 16:44
March Madness 2006	Prime95	Lounge	20	2006-03-21 04:35

2007-01-01, 09:07	#313
Yxine "Aleksandr" Nov 2005 Russia 90₁₀ Posts	M100.000.007 is 66.4% doublechecked. Happy new year!

2007-01-01, 09:18	#314
Andi47 Oct 2004 Austria 2·17·73 Posts	P-1 stage 1 on M100.000.039 is 11,44% complete; estimated completion date is Jan. 21st.

2007-01-02, 09:03	#317
Yxine "Aleksandr" Nov 2005 Russia 90₁₀ Posts	I think I know why 100.000.007 is shown to be double-checked... But this results are not right anyway... George Waltman also seems to be anavailbable, coz I cinatcted him a week ago.

2007-01-03, 06:36	#319
moo Jul 2004 Nowhere 809 Posts	what about mp2006 the program or has that died at rc1...