mersenneforum.org Paying for DC work(good or bad idea?)
 User Name Remember Me? Password
 Register FAQ Search Today's Posts Mark Forums Read

 2006-05-26, 00:52 #1 jasong     "Jason Goatcher" Mar 2005 3·7·167 Posts Paying for DC work(good or bad idea?) Guys, I've been thinking about some of these projects, and at least for some of the ones without teams(so people don't get offended) I'm thinking it might be worthwhile to attempt to get people to crunch them with incentives. Beyond the fact of whether or not this is a good idea, I'm trying to determine how one would pay. Gift Cetificates came to mind. Also, I think Wal-Mart cards are useful online if you know the number under the scratch-off place. I'm thinking someone could obtain a Wal-Mart card(you could probably walk out of the store with an empty one if you told them it was going to be added to online). They could post the number on the Wal-Mart card that ISN'T under the scratch-off, and people could donate but not be able to steal. Every three months or so, a benchmark computer would be determined that, for one processor, would be worth $1 a week. Obviously, if you had dual-core and each core fit the benchmark, it would be worth$2/week. Maybe this would be different benchmarks for different projects.(some projects would use different benchmarks depending on the architecture). Beyond that, there's the matter of fraud, especially for ecm. I have a thought on that(which may be naive) which I intend to write in the GMP-ECM forum. thoughts?
 2006-05-26, 02:04 #2 ColdFury     Aug 2002 26·5 Posts I don't think the economics works out for either side.
 2006-05-26, 02:20 #3 ixfd64 Bemusing Prompter     "Danny" Dec 2002 California 2,411 Posts I've thought about this before, and I'm sure many others have, too. However, DC projects are supposed to be volunteer projects, and I personally don't think that money should be an incentive. However, monetary incentives could encourage more people to take interest. If a payout is to be implemented, it should be minimal. It should just cover the computing costs at the most. Also, the money for the payout should be based on donations and should only be given out if there are enough donations. I don't think GIMPS would have enough money to keep paying people for LL-tests. Although, if GIMPS does have a very large fund, then here would be reasonable payout amounts: 1. Trial-factoring: $1-2, depending on factor depth 2. double-checking:$10-15, depending on size 3. regular testing: $20-35, depending on size (Assume that each kWh costs$.15, an average desktop PC uses 300 W and is on all the time, and that an average LL test takes 21 days. Each kWh = 3.6 MJ and PC uses 300 W -> 1 kWh every 12,000 seconds. 21*86400/12,000*.15 is about $22.68.) (The figures are based on the sizes of recent exponents.) Assuming an average of$30 per test, and using a figure of 7,000 LL tests per month, that would amout to $210,000. And I don't think that GIMPS has that much money. Last fiddled with by ixfd64 on 2006-05-26 at 02:25 2006-05-26, 02:58 #4 jasong "Jason Goatcher" Mar 2005 DB316 Posts Quote:  Originally Posted by ixfd64 (Assume that each kWh costs$.15, an average desktop PC uses 300 W and is on all the time, and that an average LL test takes 21 days. Each kWh = 3.6 MJ and PC uses 300 W -> 1 kWh every 12,000 seconds. 21*86400/12,000*.15 is about $22.68.) (The figures are based on the sizes of recent exponents.) Assuming an average of$30 per test, and using a figure of 7,000 LL tests per month, that would amout to $210,000. And I don't think that GIMPS has that much money. I agree that it's a ton of money, but you've touched on something important. Namely that what we're doing as volunteers is a MAJOR favor to whatever project we're involved in. The scientific community, and I mean this in a very broad sense, should be very thankful for our contribution. Especially since very few DCers have EVER made money off their hobby.  2006-05-26, 05:19 #5 Uncwilly 6809 > 6502 """"""""""""""""""" Aug 2003 101×103 Posts 999710 Posts What if Merck wanted to pay people a quid per potential drug that a person's machine tests? Like a folding @home project. The intial download could have the shapes of certain important biological molecules. Then the server would send the formula for a new found substance. The local machine would do the folding and look for a fit with target molecules.  2006-05-26, 08:04 #6 outlnder Aug 2002 2·3·53 Posts I would be happy just getting reimbursed for the cost of the electricity. My monthly bill for DC is over$110. Last fiddled with by outlnder on 2006-05-26 at 08:04
2006-05-26, 11:13   #7
xilman
Bamboozled!

"𒉺𒌌𒇷𒆷𒀭"
May 2003
Down not across

1095010 Posts

Quote:
 Originally Posted by outlnder I would be happy just getting reimbursed for the cost of the electricity. My monthly bill for DC is over \$110.
We get AC over here. 230V, 50Hz

Paul

 2006-05-26, 16:55 #8 Fusion_power     Aug 2003 Snicker, AL 7×137 Posts Zilman, Why on earth would you be worried about Air Conditioning. And at 50 hz at that. I personally would not be interested in a DC program that paid. One of the reasons I participate in gimps is that it is free to join and free to leave with no recriminations either way a person goes. Its a matter of doing something with my computer that would otherwise just be wasted time. Fusion
 2006-05-26, 17:20 #9 Greenbank     Jul 2005 2·193 Posts 50Hz gives a nice annoying low hum from the air conditioning. You don't get that in the District of Columbia.
2006-05-26, 20:09   #10
xilman
Bamboozled!

"𒉺𒌌𒇷𒆷𒀭"
May 2003
Down not across

2·3·52·73 Posts

Quote:
 Originally Posted by Greenbank 50Hz gives a nice annoying low hum from the air conditioning. You don't get that in the District of Columbia.
Nonetheless, I worry about Anonymous Cowards.

Paul

 2006-05-27, 03:29 #11 moo     Jul 2004 Nowhere 11001010012 Posts 220 for most ac units in the us but we get 60 hz -_- and we get a nice drop in the lights when it kicks in.

 Similar Threads Thread Thread Starter Forum Replies Last Post jasong jasong 8 2017-04-07 00:23 Dubslow Forum Feedback 7 2016-12-02 14:26 RienS Hardware 17 2014-11-18 22:58 NBtarheel_33 PrimeNet 3 2012-07-28 14:51 ThomRuley Lone Mersenne Hunters 5 2003-07-15 05:51

All times are UTC. The time now is 04:57.

Mon Oct 18 04:57:29 UTC 2021 up 86 days, 23:26, 0 users, load averages: 1.06, 1.03, 1.02