Which Pool ???

[xstr8guy]

BTCGuild seems to be the most stable pool. Yes it has fees but you'll probably make up the difference because of the uptime.  In my opinion, no one gets close to matching BTCG's customer support.

Eligius is good too.  No fees but you're a worthless mooch if you don't tip at least 1%.  I'm not a big fan of the stats though and payouts are often slow because of bugs in their payout system.  You'll get paid but sometimes they run waaaay behind.

GHash.io is a monster and controls a huge percentage of the network hashrate.  No fees and you don't have to feel guilty about it since they make a fudge-load of money from selling hashing power.  It does seem like they have a bit of downtime nearly everyday though.

I split my hash between all three... roughly 1/3 each.  It reduces variance because combined they represent over 50% of the network hashrate.  I like the thought of getting a piece of 50%+ of every blocked generated.  

[roy7]

Oh come on, don't suggest people use the 2 largest pools in the world.

[xstr8guy]

Oh come on, don't suggest people use the 2 largest pools in the world.

If you have some serious hash power the variance will kill you in the small pools.  And most of the big pools (BTCG and Eligius) are big for a reason... they're the best.

GHash is a different story though and I wouldn't recommend them if I didn't think it made financial sense at least in the short term.  And they are facing some serious competition from massive private pools so they won't be a threat for much longer.

[smooth]

Splitting hash is a good idea if you want to reduce variance but also help decentralization using p2pool or at least smaller pools. I set up a mining farm for a friend with some hash on p2pool and some on eligius. If there were a need to further reduce variance, I would add ghash. I would avoid the pools with fees as it is not necessary to pay fees for good pools in the current market.

[lightfoot]

Question: Does a large pool have any advantage over a smaller pool (or a single person with a block erupter) for finding the next block?

Reason I am asking is this: If a large pool has found the last block, they know this before anyone else, and can start working on the next block. Meantime the guy with the BE does not know this and is whacking away on an invalid block. Small advantage, but with blocks being generated every 10 minutes it could add up.

Or is this totally wrong?

C

[roy7]

lightfoot, the problem of withholding blocks is that if someone else finds a block before you submit it to the network, yours can get orphaned and then you are left with nothing. Risking 25 BTC to try and get a 'head start' on another block wouldn't seem to make sense. And it's all random anyway, partial work means nothing. Your odds of finding a 2nd block are the same whether you report the prior block or hold on to it. I think all you do is hurt yourself.

[faxfan2002]

With 2 antminer S1's I've used Slush's, Eligius, BTCguild - BTC & Eligius give a very consistent payout, however I've found Slush's pool to give out the same payment but more often than not a little bit more. I think it's all down to share of the pool, obviously with a smaller pool you get a little bit more of the share.

[BitcoinTraders]

Which are the best pools?

[Trongersoll]

Which are the best pools?

All of them.

[smooth]

lightfoot, the problem of withholding blocks is that if someone else finds a block before you submit it to the network, yours can get orphaned and then you are left with nothing. Risking 25 BTC to try and get a 'head start' on another block wouldn't seem to make sense. And it's all random anyway, partial work means nothing. Your odds of finding a 2nd block are the same whether you report the prior block or hold on to it. I think all you do is hurt yourself.

It's called selfish mining and some guys from Cornell (I think) did a paper on it. You can come out ahead if you are 33% of the network.

[Hunterbunter]

lightfoot, the problem of withholding blocks is that if someone else finds a block before you submit it to the network, yours can get orphaned and then you are left with nothing. Risking 25 BTC to try and get a 'head start' on another block wouldn't seem to make sense. And it's all random anyway, partial work means nothing. Your odds of finding a 2nd block are the same whether you report the prior block or hold on to it. I think all you do is hurt yourself.

It's called selfish mining and some guys from Cornell (I think) did a paper on it. You can come out ahead if you are 33% of the network.

Yeah, also another reason why p2pool nodes need more love. 33% is very close to 50%

[lightfoot]

Yeah, also another reason why p2pool nodes need more love. 33% is very close to 50%

Ok, I don't want to be a chump (did that with Catcoin) but I wouldn't mind throwing a few hundred gh somewhere. What's a good solid reliable p2p node I can point some stuff at while I sit down and compile the software on my NeXTStation Turbo?

I'll watch the payout for a few days.

C

[smooth]

Yeah, also another reason why p2pool nodes need more love. 33% is very close to 50%

Ok, I don't want to be a chump (did that with Catcoin) but I wouldn't mind throwing a few hundred gh somewhere. What's a good solid reliable p2p node I can point some stuff at while I sit down and compile the software on my NeXTStation Turbo?

I'll let others suggest nodes, but I will say you need more than a few days to evaluate p2pool. It is a small pool (about 0.5% of the whole network last I checked) so it only gets a block every day or two. To give it a fair eval you should be willing to wait a few weeks at least. This isn't really any different from other small pools if you have a few hundred GH. If you want to help decentralize things and not run on a huge pool that finds blocks every hour you have to be willing to accept more variance on found blocks.

[lightfoot]

I'll let others suggest nodes, but I will say you need more than a few days to evaluate p2pool. It is a small pool (about 0.5% of the whole network last I checked) so it only gets a block every day or two. To give it a fair eval you should be willing to wait a few weeks at least. This isn't really any different from other small pools if you have a few hundred GH. If you want to help decentralize things and not run on a huge pool that finds blocks every hour you have to be willing to accept more variance on found blocks.

Variance is fine, as long as the totals even out over a reasonable period of time. Reasonable being the 2 week block retarget period. One thought in my head though is a bad streak of luck now wouldn't be erased by an equally good stream of luck "later" because difficulty keeps creeping up. Does that skew the equations as well?

For example, if I mine a Th at Eclipse then over the past two days I have made .045btc per day due to the crap luck blocks taking 24 and 26 hours to drop. If everything was normal I would get .12btc a day, so I have .09 instead of .24. Now if everything went back to perfect normal but in two weeks I got a string of great luck to cancel out the crappy (say two days of 6 blocks a day), those six blocks would only pay .20 btc so woudl I be out of (heh heh) luck due to the underlying difficulty pop?

Never dull, thinking about this stuff. By converse though a round of good luck now is worth a lot more later. Ug....

C

[smooth]

Variance is fine, as long as the totals even out over a reasonable period of time. Reasonable being the 2 week block retarget period. One thought in my head though is a bad streak of luck now wouldn't be erased by an equally good stream of luck "later" because difficulty keeps creeping up. Does that skew the equations as well?

This is true. Of course it could go the other way. That is, if you get lucky at first and then equally unlucky later. Overall in that case you come out ahead (as you mentioned later).

In reality though, if you are mining for the long term you will be increasing your hash rate over time. If you grow your hash rate at the same rate as the network, equal (good or bad) luck later will balance out with the equal (bad or good) luck earlier.

Keep in mind though, there is no guarantee luck will ever equal out in either direction. Some people are just luckier than others over a lifetime.

[lightfoot]

In reality though, if you are mining for the long term you will be increasing your hash rate over time. If you grow your hash rate at the same rate as the network, equal (good or bad) luck later will balance out with the equal (bad or good) luck earlier.

Keep in mind though, there is no guarantee luck will ever equal out in either direction. Some people are just luckier than others over a lifetime.

Luck lasts a lifetime if you die young.

That's another good point: The problem with being in mining for me is that it's not much of a profitable venture. Actually it is explicitly impossible to make profits (long term profits) in mining because it's a perfectly elastic market or whatnot. That's a part of the reason I haven't jumped to Ants or 28nm or whatever. I'm mostly in it for the fun, learning, and to keep my house warm in the winter. :-)

There is another way to look at luck; bad luck mining could be absorbing other bad things that could be happening in my life. Losing job, kid, car, things like that are serious bad luck. If my mining is a "luck sink" for the bad joss to go to, that's oddly enough ok for me. But that is one of those woo-woo spiritual things, each roll of the dice is independent of ever other roll...

One odd thing about Eligius though: When I mined there in the beginning during the summer/fall last year they had a horrible streak of luck for the few days I was on. I think I had 50% shelved shares and I thought that sucked. So I switched to eclipse, but left eligius as my backup. And over the next few months I realized I got every one of those shares back at the value from September. Oddly enough eligius' payment system seems to transcend the concept of time coupled with difficulty. Which is very, very interesting.

C
C

[smooth]

That's another good point: The problem with being in mining for me is that it's not much of a profitable venture. Actually it is explicitly impossible to make profits (long term profits) in mining because it's a perfectly elastic market or whatnot. That's a part of the reason I haven't jumped to Ants or 28nm or whatever. I'm mostly in it for the fun, learning, and to keep my house warm in the winter. :-)

In that case you should definitely play with p2pool. You shouldn't get clobbered even if you do have some bad luck (or maybe it goes the other way and you get a windfall).

One odd thing about Eligius though: When I mined there in the beginning during the summer/fall last year they had a horrible streak of luck for the few days I was on. I think I had 50% shelved shares and I thought that sucked. So I switched to eclipse, but left eligius as my backup. And over the next few months I realized I got every one of those shares back at the value from September. Oddly enough eligius' payment system seems to transcend the concept of time coupled with difficulty. Which is very, very interesting.

Eligius had very good luck for a while. I'm not sure of the exact timeframe but I remember being at 99%+ on there for quite a while.

[justbtcme]

 ghash.io is the way to go !

[OgNasty]

ghash.io is the way to go !

I hope you're kidding. ghash.io has a dangerous amount of network share right now. It is the worst pool you could choose to mine on at the moment.

[neot]

BTC Guild.  IMHO