Please get these DONE!

outlnder · 2002-11-24, 08:28

I hate to be pushy but we need to get the following exponents completed.

5621653 D 63 3448704 622.5 -50.6 9.4 19-Sep-02 18:10 11-Mar-01 21:30 rexmaughan gntmau4
5837947 D 63 4345600 792.6 -0.3 59.8 22-Nov-02 02:09 22-Sep-00 18:26 rexmaughan gntmau2
6070433 D 63 5797385 541.4 -23.3 36.7 30-Oct-02 00:46 31-May-01 23:30 dobrien Compaq
6095153 D 63 4343423 415.0 -13.8 46.2 06-Nov-02 12:20 05-Oct-01 07:31 wonsuck wonsuck
6120227 D* 63 2845334 279.1 12.4 56.4 19-Nov-02 17:28 18-Feb-02 06:17 TempleU-DI CFC9DCD8F
6223561 D 63 5243647 718.6 -53.5 6.5 27-Sep-02 19:20 05-Dec-00 18:53 sqf pentecost
6233411 D 63 4832416 583.7 14.3 72.3 08-Nov-02 16:11 19-Apr-01 16:03 aga aga13
6272333 D 63 6157313 710.8 0.3 60.3 21-Nov-02 14:39 13-Dec-00 13:34 ckrusen library
6348107 D 63 6065846 670.1 8.5 68.5 15-Nov-02 19:30 23-Jan-01 05:40 matheux 03
6454769 D 63 3851544 646.5 -16.8 43.3 27-Oct-02 14:52 15-Feb-01 21:48 S02664 P3400Mz
6484129 D 64 3581861 280.1 -24.6 32.4 29-Sep-02 18:14 17-Feb-02 06:45 S55781 C8794D514

If you own one and can't get it done, post here and let someone else take it. These are the only outstanding exponents below 6.5M.

smh · 2002-11-24, 10:03

All numbers seem to be in progress and over half way done (except for 1). Some have been updated this month (isn't the default connection time 28 days?) and all the others less then two months ago. 2 numbers are about to expire within 10 days.

I guess some of these are pc's that are not on 24/7. But i admit, 2 years to complete one exponent is a bit long

Deamiter · 2002-11-24, 16:01

Doesn't seem to me that this is really all that important. I mean, it's not as if we're competing with some other organization to verify M38... I spend my time worrying about how to get my boxes to run a little faster or how to add to my computing power rather than worrying about some 386 that's taking 3+years to complete a DC. Just my $0.02

wfgarnett3 · 2002-11-25, 06:56

Yeah, I anxiously await to check the 60 some numbers left to conclusively verify that Mersenne #38 is in fact #38. I guess some people could manually test them and submit the results, to get the double checks done quicker.

Joe O · 2002-11-25, 14:28

What you are suggesting is called "poaching". It is highly frowned upon. These people have put in a lot of hard work on their exponents. Your suggestion would waste their efforts, and might even "piss them off".

eepiccolo · 2002-12-26, 16:36

I've looked at the assignments report, and it looks like some computers have exponents reserved for the NEXT TEN YEARS.

So here's a question. Does it make sense to allow people to reserve exponents when the expected start date of the exponent (which would be the expected end date of the currently assigned exponents) is farther away than the next required check-in time? To me, this does not make sense, for you could just reserve the exponent at your next check-in. It would be practically imposible to start another exponent before that time. Perhaps someone else has some insight on this?

Xyzzy · 2002-12-27, 00:22

They could be using the same name for multiple computers...

eepiccolo · 2002-12-27, 13:43

Quote:

Originally Posted by Xyzzy

They could be using the same name for multiple computers...

I see how naming different computers the same name could make it difficult. But perhaps a solution lies in a modification to Prime95. The program could talk to the server, see what the overdue check-in is set to, and not allow the computer to reserve any exponents if the currently assigned exponents have a farther away expected finished time than the overdue check-in time.

My discussion is based on my humble opinion that there is no point to reserving exponents that far in advance. I'd like to see if there are others who aggree with this opinion, or if there are some that think we should be allowed to reserve exponents in as far in advance as we want.

Of course, the solution I just put forth two paragraphs ago probably would not help for manual reservation. But hey, idea :idea: . If a person manually reserves exponents, could they be required to enter a computer ID? And if they really are getting exponents for different computers, they'll just have to go through the "trouble" of naming their computers different names if they want their exponents.

So, any ideas or opinions out there?

Xyzzy · 2002-12-27, 13:59

Well, for network enabled computers you could use the MAC address to uniquely identify them...

But the current system is fine as I see it... If someone reserves a pile of exponents I don't have too much of a problem with it... There is an infinite supply of them anyways...

eepiccolo · 2002-12-27, 14:17

Quote:

Originally Posted by Xyzzy

If someone reserves a pile of exponents I don't have too much of a problem with it... There is an infinite supply of them anyways...

Ah, infinity, how handy you are. This is very true, so I guess people don't really need to get bent out of shape, then. Now, I see how one could get anxious about verifying M38, but in the big picture, we've found 5 numbers in six years, but the other 34 practically took hundreds of years to find. So it seems that GIMPS is doing pretty darn well.

dswanson · 2002-12-27, 18:10

Quote:

Originally Posted by eepiccolo

My discussion is based on my humble opinion that there is no point to reserving exponents that far in advance.

I can think of one reason. I have a computer at home that I use at 6:00 UTC (which is 10:00 in the evening local time) to collect small exponents just after they expire. I then farm these exponents out to a bunch of computers at work over the next couple of days. So my home computer tends to have a list of 10-15 unworked exponents for hours to days, until I get them transferred.

But I keep the required check-in time on all of my computers, including my home computer, set to 1 day so I can keep tabs on progress and watch out for any dead machines. Disallowing reservation of exponents beyond the 1 day check-in interval would be a nuisance. I can think of a work-around (set the check-in interval to 90 days, reserve new exponents, set the interval back to 1 day), but as I said it would be a nuisance.

2002-11-24, 08:28	#1
outlnder Aug 2002 2×3×53 Posts	Please get these DONE! I hate to be pushy but we need to get the following exponents completed. 5621653 D 63 3448704 622.5 -50.6 9.4 19-Sep-02 18:10 11-Mar-01 21:30 rexmaughan gntmau4 5837947 D 63 4345600 792.6 -0.3 59.8 22-Nov-02 02:09 22-Sep-00 18:26 rexmaughan gntmau2 6070433 D 63 5797385 541.4 -23.3 36.7 30-Oct-02 00:46 31-May-01 23:30 dobrien Compaq 6095153 D 63 4343423 415.0 -13.8 46.2 06-Nov-02 12:20 05-Oct-01 07:31 wonsuck wonsuck 6120227 D* 63 2845334 279.1 12.4 56.4 19-Nov-02 17:28 18-Feb-02 06:17 TempleU-DI CFC9DCD8F 6223561 D 63 5243647 718.6 -53.5 6.5 27-Sep-02 19:20 05-Dec-00 18:53 sqf pentecost 6233411 D 63 4832416 583.7 14.3 72.3 08-Nov-02 16:11 19-Apr-01 16:03 aga aga13 6272333 D 63 6157313 710.8 0.3 60.3 21-Nov-02 14:39 13-Dec-00 13:34 ckrusen library 6348107 D 63 6065846 670.1 8.5 68.5 15-Nov-02 19:30 23-Jan-01 05:40 matheux 03 6454769 D 63 3851544 646.5 -16.8 43.3 27-Oct-02 14:52 15-Feb-01 21:48 S02664 P3400Mz 6484129 D 64 3581861 280.1 -24.6 32.4 29-Sep-02 18:14 17-Feb-02 06:45 S55781 C8794D514 If you own one and can't get it done, post here and let someone else take it. These are the only outstanding exponents below 6.5M.

2002-11-25, 06:56	#4
wfgarnett3 "William Garnett III" Oct 2002 Bensalem, PA 1010110₂ Posts	Mersenne 38 Yeah, I anxiously await to check the 60 some numbers left to conclusively verify that Mersenne #38 is in fact #38. I guess some people could manually test them and submit the results, to get the double checks done quicker.

2002-11-25, 14:28	#5
Joe O Aug 2002 3·5²·7 Posts	wfgarnett3 What you are suggesting is called "poaching". It is highly frowned upon. These people have put in a lot of hard work on their exponents. Your suggestion would waste their efforts, and might even "piss them off".

2002-11-24, 10:03	#2
smh "Sander" Oct 2002 52.345322,5.52471 2245₈ Posts	All numbers seem to be in progress and over half way done (except for 1). Some have been updated this month (isn't the default connection time 28 days?) and all the others less then two months ago. 2 numbers are about to expire within 10 days. I guess some of these are pc's that are not on 24/7. But i admit, 2 years to complete one exponent is a bit long

2002-11-24, 16:01	#3
Deamiter Sep 2002 3²×13 Posts	Doesn't seem to me that this is really all that important. I mean, it's not as if we're competing with some other organization to verify M38... I spend my time worrying about how to get my boxes to run a little faster or how to add to my computing power rather than worrying about some 386 that's taking 3+years to complete a DC. Just my $0.02

2002-12-26, 16:36	#6
eepiccolo Dec 2002 Frederick County, MD 2·5·37 Posts	I've looked at the assignments report, and it looks like some computers have exponents reserved for the NEXT TEN YEARS. So here's a question. Does it make sense to allow people to reserve exponents when the expected start date of the exponent (which would be the expected end date of the currently assigned exponents) is farther away than the next required check-in time? To me, this does not make sense, for you could just reserve the exponent at your next check-in. It would be practically imposible to start another exponent before that time. Perhaps someone else has some insight on this?

2002-12-27, 00:22	#7
Xyzzy "Mike" Aug 2002 20040₈ Posts	They could be using the same name for multiple computers...

2002-12-27, 13:59	#9
Xyzzy "Mike" Aug 2002 10000000100000₂ Posts	Well, for network enabled computers you could use the MAC address to uniquely identify them... But the current system is fine as I see it... If someone reserves a pile of exponents I don't have too much of a problem with it... There is an infinite supply of them anyways...