Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 446. (Read 2591964 times)

[edited more than once]

You're right, I think there was some bias in there.  I'm not new to p2pool, nor new to mining.  I've always been a small miner in p2pool, and the huge variance I experience hurts.  And the 16% stale rate, on average, doesn't help matters.

I thought a bit about why I stay away from p2pool and why I suggest other "small" miners do the same.

It's the variance and the continued massive increases in difficulty.

Over the roughly 2 week period between BTC difficulty increases, if you average a share every 3 days, that means there will be times when it's longer, and there will be times that it's shorter.  Because of stale shares, there will be more longer periods than there will be shorter periods.  If BTC difficulty never increased, it would even out in the end compared to everyone else in p2pool.

The rub is in the difficulty increase.  In that 2 week period I should get 4 shares on average, assuming share difficulty doesn't increase.  But it is... so if I luck out and get 4 shares, or more, then great.  But if I don't, I will do better on a standard pool with little variance (like BTC Guild).  

Maybe my math is wrong.  The resident math expert will likely chime in here correcting me if I'm off.  It may just be a feeling.  Regardless, it seems to me because of the endless drastic difficulty increases, when I come out behind in that difficulty period, I will never catch up in the next difficulty period.  And there will be times when I come out behind... so therefore I'm better off in a standard pool.

Again, this applies to small miners.  With share difficulty approaching 1m, I can't imagine pointing my measly 38 gh/s at p2pool.  The variance would be awful, and before I know it, the BTC difficulty will increase again.

M

Good Pool 

Are you using any nonstandard configuration or settings?

I'm planning to give p2pool another shot on my AntMiners next week. The hardware for the node is on order right now.

How about you drop the obvious bias? "Will" be dry periods, and "probably" some days its shorter?  Seriously?

There WILL be longer dry periods and there WILL be shorter dry periods. That is the nature of variance. But when your total hash rate translates to $2/day the absolute variance will not be much. You will NEVER underperform by more than $2/day.


If you have a "high" reject rate then something is malfunctioning or misconfigured.  A few percent or so rejects is normal and anything up to about 10% probably does not hurt your earnings relative to another pool.


Then you would be wrong. If your expected earnings are $2/day and you have multiple days of getting $0 then you WILL also have days of getting $6 or more. Mathematical certainty.

Over periods of weeks you will very likely earn more than regular pools even with 12 GH. If you want to earn less, go right ahead, but stop spreading biased misinformation.

it was allways the same with him/it and it will always be stupidity meets ignorance 

You should probably stop feeding the troll.  He'll never use p2pool long enough or with any objectivity.  He just knows he doesn't like it without having a valid reason.
Btw, over the past 3 days I've gotten 0.085+ bitcoins with only 30Ghs.  The 4 days prior to that I got 0 however.  So yeah, variance...  The good news is that over time I've always come out better than other pools.  You just have to be patient.  I completely understand though that people with low hashrates might prefer a different pool.  But really everyone with hundreds of GHs should consider trying p2pool for a couple of weeks.

Payout is over 3 blocks, so takes 1.5 days to payout. 

So, just ran my 120GH on p2pool for 24 hours.  Payout stands at 0.004BTC.  That's shit.  Barely pays for the power used running a PC all day.

Back to a real pool.

P2pool pays out over 3 blocks on average.  The pool is finding around 2 blocks per day.  This means if you find a share, you will get 3 payouts spaced every 12 hours (on average).

If you find a share once every 3 days, it remains active for 1.5 days, so you have an active share for 50% of the time.

On average it'll take you about 3 days to find a share at current sharechain difficulty (770k).  That means there will be dry periods much longer, and probably some days it's shorter.  Another factor regularly overlooked with p2pool is the high reject rate.  I know I was extremely disappointed when those 3 or 4 days would go by and I'd finally get that cherished share ... only to have it rejected.

I tend to think you won't be getting $5 or $10 every few days with 12gh/s. 

M

There is no reason to assume that the satisfaction of seeing a $2/day reward is greater than the satisfaction of seeing a reward of $5 or $10 every few days. This is hardly solo mining where you are trying to get very lucky and score $20K but most will never get anything (although I'd add that many very small miners might well enjoy that too -- after all lots and lots of people play the lottery and unlike mining the vig on that is atrocious). Every miner with say 12 GH on p2pool will get fairly consistent rewards every week at least, just not every day.

I will say this though, the cost of running a full bitcoin node and p2pool node in terms of bandwidth and effort are significant. If you are only pulling own a few dollars per day it probably isn't worth it.

my 4GH/s miners get an accept on my local node just as often as my 60Gh/s miners do. BFL's in this case. The point is your chances of getting a share are equal proportionally.. AS FAR AS I CAN TELL .. YMMV.

The satisfaction of seeing a reward, even small, is what makes mining worth while.  With a share difficulty now at 888k, your odds of getting a share with 12gh/s is very very low.

If you want steady, although small income, point it to a large pool like BTCGuild.

M

This has also long been my opinion. I absolutely cannot understand what people are thinking, my best guess is that they're honestly confused about expected returns.

What doubly boggles me is that some of the same people who say they cannot tolerate variance turn around and gamble (with negative expectation).

"Wildly extreme" meaning your current estimated $2/day earnings will end up being $0 on a lot of days (and well above $2 on some other days)? How does that even matter to anyone?

I would not put that little hashrate on p2pool.  You'll variance will be wildly extreme.

M

It is hard for me to imagine anyone with 12 GH needs to use it to generate consistent cash. At that point your expected income is around $2 per day. If that goes to $0 for a few days or longer (balanced out by some higher paying days), who would even notice?

If you are running the miner 24/7 and plan to just leave it run for weeks/months, then it will average out in the end.

Some people need a steady stream of cash, as they spend the money as it comes in.  They need lower variance.

How much are you going to get no matter how you mine?

It will be a little less consistent with p2pool but your expected payout is higher.

BTW, one share every four days does not sound right to me, but I'm not positive.  I did some p2pool tests not too long ago with 70 GH and I was getting consistent shares.

So.....with my puny 12 Gh/s rig I might as well forget P2Pool right?  Expect share is somewhere around 4 days right now and payout per block is BTC0.00143.  So that's something like BTC0.00143 every 4 days?  Am I reading this right?