Topic: Eligius: 0% Fee BTC, 105% PPS NMC, No registration, CPPSRB - page 134. (Read 1061417 times)

Wouldn't having large asics doing higher difficulty ease the load on the server?

I understand a machine can do 1024 difficulty fast, so running 256 will just submit more shares at a more aggressive rate correct?

I was under the impression it was the opposite, that if 1024 was set and the machine could run through it it would end up submitting more shares at a higher rate than 256 from the constant new work and submitting more shares of less difficulty.

(Sorry I am a little bit confused at this point!)

It can not.  There is pretty much zero reason to set a custom difficulty.  The server already pretty aggressively chooses an appropriate difficulty.

Through my research I read that the server actually rejects the command line setting on bfgminer

Really curious if it can or can't be done, can anyone provide answer?

@wizkid057
Yeah it does, I'm sure you have explained this about 148 pages worth and it's frustrating to explain it 4 more times on this page alone to just me. Thanks for your help.

There's A LOT of misinformation on these forums(not talking about this topic btw), so hearing from the man himself is refreshing.

@baddw

My thanks to you as well sir. You read it correctly, I wrote it wrong.

Yes.  I was reading your formula incorrectly before.  I see now that you mean (25/173366316979)*(100000+50000) while I was interpreting the latter part to still be part of the denominator.

Just to be super clear here, your earnings will be (25/173366316979)*(A+S) where

A = Accepted shares for this round, subject to a maximum roughly equal to (Your Hashrate) * (Expected Time to Find a Block, i.e. the time that would constitute 100.000% Luck).  In other words, when the current round drops below 100% luck, the oldest shares from this round will start being shelved, and A will max out, and be less than the actual number of shares that you submit in this round.
S = Previously shelved shares (will be 0 if round drops below 100% luck)

To make it even simpler, A+S = "any of your shelved shares within the most recent 173366316979 shares on the shelf".  Instead of thinking of it as a 2-queue system ("current round shares" versus "shelved shares"), think of it as a unified system (which it actually is).  Every share that you submit immediately goes to the shelf, period.  The share that you submitted 5 minutes ago is higher priority than the one that you submitted 10 minutes ago.  And all the shares submitted by other users between 5 and 10 minutes ago, lie in between those shares.

While you can't necessarily know how many shares you have shelved for any given block, if you have been mining continuously and consistently, you can get an estimated idea of your earnings for any block by just gauging your hashrate versus the pool hashrate.  E.g. if you are hashing at 1TH/s and the pool is hashing at 7000TH/s then your earnings will be (on average, of course) 1/7000 of the pool earnings.  So if there are 17.3B shares paid per block, you can expect for 2.47M of them to be your shares.  And you can look at an unlucky round that's say 50% luck, i.e. 34.6B shares accepted during that round.  Then you would have submitted 4.94M shares during that round, and you would have been paid out for 2.47M of them, so you would have 2.47M remaining shelved from that round.

But all of this blockwise talk of earnings is really digging into a lot of detail that kind of glosses over the basic facts of mining.  Bitcoin mining is a stochastic process; your miners are hashing at a given rate, the pool is hashing at a given rate, and the Bitcoin network expects to find blocks at a given rate.  The best way to estimate your earnings is to look over longer periods of time to smooth over the variance that comes with good luck and bad luck.  Eligius gives an "Estimated earnings per day" on the right-hand-side of your screen, which is based on your 3-hour average hashrate.  This is a pretty solid number which is really just based on your hashrate and the current difficulty, and should hold true regardless of whether you were mining with Eligius, mining with another pool, or solo-mining.

I'm not sure what you're trying to figure out still.  The maximum earnings of a miner are (25/{difficulty} * weighted shares).  CPPSRB attempts to pay all of them and never forgets a single share.

In lucky rounds, all shares submitted since the previous block are paid at 100% PPS (the formula above).  Older shares in the share log are paid with the remaining funds in latest to oldest order until 25 BTC worth of shares are paid.

In unlucky rounds, the last {difficulty} shares are paid in latest to oldest order.  This means that some number of shares are left on the shelf to be paid later.  Shares are immediately building on top of these from the new round, however, so the longer the next round the less shelved shares will be paid when a block is found, if any.

Long story short, you can't estimate your earnings with that level of granularity.  Your best bet is to use a longer time frame, something like at a minimum, "I should earn about X BTC in 14 days."  30+ days makes more sense.  CPPSRB works out so that you get paid as close to 100% PPS as possible.

I'll just clarify that not all shares from the 25% round would be paid (yet) in this scenerio.

A 25% luck round, as per luck in wizstats, means that the round took 4x as many shares as expected, earned the normal 25 BTC but 75 BTC in shares were shelved for a total of 100 BTC in shares submitted.  A 400% luck round would be 1/4 of shares expected paying 6.25 BTC to pay off all shares submitted this round, and 18.75 BTC towards shelved shares. (I changed it to 400% from 440% for simplicity).  No more than 25 BTC (+txn fees) can be paid per round.  So, in this example of a 25% round then a 400% round, 50 BTC are paid out.  25 BTC to those working on the 25% round when the block was found, 6.25 BTC paid to those working when the 400% block was found, and 18.75 BTC paid towards the shelved shares for the 25% round, leaving 56.25 BTC in shares shelved still from that round.

Hope that makes some sense.

I guess I should have been more specific to what I was asking here. I'm trying to figure out what a payout would be for an individual miner, not the whole pool. There's no way for the user to see what his shares on the shelf are from unlucky round, but going by what their weighted averages are, you can guess as a payout.

So is that formula how it's figured out correct? What I'm asking is, are the payouts calculated 25/173366316979(100000+50000)=0.0000216305 , where 100,000 is the current round shares and the 50,000 is from a previous unlucky round that made it? So 150,000 is just something for an example.

No, the payout for the block is always 25 BTC.  They are paid out to the topmost 173366316979 shares in the stack ("shelf").  If this is an unlucky round (greater than 17.3 Billion shares accepted) then these will all be from the current round, and some shares from the current round will be shelved to hopefully be paid out later.  If it is a lucky round, all shares from that round will be paid, and whatever shares are on the top of the stack will be paid out until a total of 17.3 Billions shares are paid out.  These shares may be from the last unlucky round, or if there has been a run of good luck, you may have to go back several unlucky blocks.


Yes.


Yes.


No, not really.  Longevity (or just sticking around for more than a couple of days) will reduce your variance, but Eligius is pool-hopping-neutral and will pay out your shares regardless of how long you stick around.  If you are pool-hopping, and happen to mine for 6 hours with Eligius during a lucky streak, you could very well be paid out for 100% of shares.  Or if you happen to mine for 6 unlucky hours, only 25% of your shares might be paid out immediately, while 50% more are paid out in the next 36 hours during a good run streak, while 25% might remain unpaid forever.  Leaving you with only 75% of total shares paid.

While if you mine continuously with Eligius (i.e., weeks/months) then you will see these numbers converge towards the long-term payout rate which tends to around 97%, and is almost always between 95% and 99% unless the pool happened to have a severe bad luck streak which would drag you down a little more.  But as always, the bad luck tends to be balanced out by good luck in the long run, but there's no way to predict it, as each block mined is an independent event and there is no way to say that the pool "deserves" or "is due for" a block after going 12 hours without one (see Gambler's Fallacy).  So I would say, the longer you mine, the more likely you are to converge around 97% of shares paid.  If you mine for a short time, you could see 100% paid or 50% paid.  Just simple variance.

Note that in this whole discussion I am using "paid out" to mean "BTC credited to your account", not "entering the payout queue" or "actually seeing the BTC in your wallet".  The payout queue is an entirely separate mechanism from CPPSRB, see the post in my sig.

So what you're saying is 25/173366316979 *(round time in weighted shares + shelved shares from previous round)= payout for that block. As long as those shares in the entire pool don't out weigh the total amount of shares for the current round.

Now in the case of shares from a previous round, those would apply in the case of a over 100% luck round, or is that a different scenario?

So if we had a 440% luck round, you would be paid the shares you submitted that round plus round shares that are still on the shelf from say a 25% luck round, or what have you?

So the benefits of having more hashes are obvious, but the longevity is where you can potentially be paid your fullest possible earnings?

I'm gonna explain with a simple example, in which there are only two miners: a slow one and a fast one. Both mine continuously at a constant hashrate.

Every time someone submits a share, that share is added to the shelf. The 'value' of such share is 25 BTC/bitcoin_difficulty. That is totally independent of the speed of the miners.

The share shelf contain mostly shares of the 'fast' miner, with the slow miners shares more or less uniformly dispersed trough it. The relative proportion of the fast vs slow shares in the shelf is the same as the relative proportion of their hashrate, and thus the same as the relative proportion of their fair reward.

Every time the pool finds a block, it will pay the topmost 25 BTC worth of shares of the shelf. As stated earlier, since the shares are uniformly distributed and in a fair proportion between the miner, the payed amount to each miner is fair too.

The difference between a short and a fast round is that in a long round, all miners will receive for that round an amount less than what would be expected just by the number of shares they submitted (and the shelf grows), whereas for a short round, miners will receive more that what would be expected just by the number of shares they submitted (and the shelf shrinks). However each round pays exactly the same amount, and that payout is fair, because it is in the same proportion of their respective hashrate.

Not on the website. However, I think the server honors difficulty requests sent by stratum as a command-line option of your miner.

I've been searching a bit, there is no way to set a minimum worker difficulty?

Incorrect

Seems like to me, the shorter rounds benefit the higher hash rate miners/guys, and longer rounds will slightly favor slower folks.

baddw does a great job explaining it with numbers and examples.  I'll try to simplify it for you.  The expected shares is equal to the current difficulty, so changes every 2016 blocks.  Every block that Eligius finds, it can ONLY pay that many shares.  So, if it took fewer shares than expected to find a block, there are shares left over, which payout the "shelved" shares.  If it took more than the expected shares to find a block, Eligius "shelves" the extra shares it took to find the block.

The payout queue pretty much works based on how long it has been since a miner was last in the queue, which is why it is recommended that you set your payout threshold to be at least 3 days worth of mining.  If you set your threshold too low, you get placed lower in the queue because you were just there, and other miners have been waiting longer.  Your position in the queue is dynamic, with miners constantly being added before and after your position.  Eligius uses this to show you how many blocks are ahead of you.

Hope this helps.

It can get very confusing. I'm not great at math, so I've hung back and watched the pool for a long time. I don't currently mine BTC, as my rig is more profitable on alts (it's very small), but when I did, and in the future if I can get better equipment, I have and will have Eligius as my primary pool.

Not because I fully understand the payout system (I don't), but because for the time that I was mining here, the operator has been dilligent, honest, and responsive. I never made a lot mining BTC, as I was late to the game. But after a couple of months on Eligius, I was never concerned about whether or not I would get my payout, either. It came, always. Not always on time, but always.

Whenever I had a question, even dumb ones, either Whizkid or some of the other rather smart members of this thread helped me out. I learned a great deal from my time here, and I still keep track in antiicipation of both gaining more knowledge and (hopefully) getting more powerful equipment.

Thank you so much for explaining it so that I could understand it.  It all makes since completely.

The shares accepted should not "flat line", they should continue to go up for as long as the round continues.  Your earnings for that round will, however, "flat line" as the luck turns from >100% to <100%.

Looking at the round that ended when Eligius found Block Height #311302, this round took about 30 minutes and there were about 3.15 billion shares  accepted by the pool.  This comes to approximately 6.3 billion shares per hour, or 105 million shares per minute.  So this is the baseline hash rate of the pool.

On average, the pool expects to find a block once every 17.34 billion shares.  (This = Bitcoin difficulty.) So this comes to the pool expecting to find a block every 165.1 minutes, or around 2 hours and 45 minutes.

In the round ending in Block Height 311302, the pool accepted 3155503232 shares, which 3155503232/17336316978 =  18.201% of number of shares expected to find a block.  Take the inverse of .18201 (i.e., 1/.18201) and you get 5.4934 = 549.34% which is the "Luck" shown.

In the block ending in Block Height 311364, the pool accepted 55,282,473,728 shares, which 55282473728/17336316978 = 318.88% of the number of shares expected to find a block.  Take the inverse of 3.1888 = .31359 = 31.36%, the "Luck" for that round.

Now to the shelved shares. 

In the second-to-last round, ending in Block Height 311302, the pool accepted 3.15 billion shares, while it paid out 17.34 billion shares.  This means that the 3.15 billion shares accepted during those 30 minutes of that round were paid out, as well as an additional 14.18 billion shelved shares got paid out when the block was found.  Looking through the round history, and given the run of good luck up until that point, it seems likely that these shelved shares came from the round ending in Block Height 311,175, which was a round that took 12.33 hours, and the last significantly unlucky round before the round ending in Block Height 311302.  Although the shelved shares paid out could have come from older blocks, if the shelved shares from the round ending in Block Height 311175 had already been paid out (which would not be difficult to figure out given a spreadsheet or just some time with a calculator).

In the most recent round, ending in Block Height 311364, the pool accepted 55.28 billion shares, while it paid out 17.34 billion shares.  The 37.94 billion excess shares were shelved, and they will be paid out when good luck allows them to be paid out.  The pool will need at least 3 "good luck rounds" in succession, in order to pay out these 37.94 billion shelved shares.  Over the last 24 hours, the pool had 9 "good luck rounds" in a row, which was outstanding and which paid off a lot of shelved shares.

Ok the mind just kicked in thats where shelved shares comes in.  I understand now.  The shares are shelved when the round percentage is low, and when we have great averages finding blocks the shelved shares are then paid.

Ok got it now.

Please correct me if i am wrong.