You are sure you are using the correct user_worker name? Most importantly, that it starts with your correct user name and an underscore?

It would be very useful if those who want the "estimated pay per block" number back could explain what they get out of it.

Here is how I see it:

The shifts page shows how much work you did at which time and how well it was paid. The payment for a block is many of these lumped into one bigger number, which is much less informative.

Imagine you do a little mining for a couple hours in the morning. Later you mine a couple hours in the evening. We find a block and both the morning and evening work is paid. Later we find another block and only the evening work is paid, as the morning work is no longer among the 10 latest shifts.

Looking at the two payments you got one big first and then one small payment later. You could look at the time the blocks were found and correlate that with your work in the 10 latest shifts at the time, and you'll understand what's going on. But most people won't. They will believe they are being punished for not mining 24/7. Or that "I need to wind up that PPLNS-thing, so I get paid fairly again."

Looking at the new shifts page you can instead look at how much work you did in the morning hours, and in the evening hours. And you can see that the morning work was paid an average rate, while the evening work was luckier and received double that.

I really believe this shows much better the relation between work you did and the pay you received.

What is perhaps missing is showing the 10 paid shifts on the block page (when you click a block in the block list). Would that be of help?

Try us2.bitminter.com - it has extranonce2 size of 4 bytes.

I'll be switching mint.bitminter.com (us1) from 6 to 4 bytes in the future.


For unconfirmed payments there is only the unconfirmed balance on the account details page. Hmm. Maybe I could add your share from the block in the block list. There's not much room though.


You're welcome

That's where it is. There is one payout per block found.

But I think pay per work done is more useful, which you find on the shifts page, is better than pay per block.

Will new versions of the relay client be announced somewhere? (here maybe?)

It may take some getting used to, but can you try looking at the shifts page for this?

It shows your average hash rate in the latest shifts, and this is what the "pay per block" estimate was based on. Sure, it's 10 numbers instead of 1. But it also tells you roughly when there was a problem if you had some downtime. This is your mining health in the pool. The shifts page also shows the pool hashrate, although I think that is less useful as it doesn't affect your average earnings, only the variance.

The live stats wasn't showing the total for all workers for all time. It was the total for the current round, which doesn't appear very useful. Maybe the total for the current shift does what you want? A total for all workers for all time is a good idea though - I'll add that to my TODO list.


The last 10 shifts are still being paid. Since their pay rate is not known yet, they show as question marks.

Some shifts have very high pay and some have zero. If you mine 24/7 this evens out and doesn't matter. If you are mining part time it will increase your variance, causing payouts to vary a lot depending on luck.

The old one is in Los Angeles. The new one is in Kansas City.

Not very far apart, but it works for failover. We'll have servers in other areas soon.

If you know any good hosting providers in Asia please send me a message.

Yeah, it is still there in the transaction history. There is one payout for each block.


If you have suggestions for more stats, send them to me. I don't know any useful measure of pool efficiency except the stale rate. Most users are at 0.1% or below. As long as the stales stay that low I don't know how interesting it is to show more stats on that..


Glad you like it Yeah, I'm not sure what to do with the 10 latest shifts. I suppose I could show how well they have paid so far, but I wonder if people will misinterpret that. I could show a text saying "not yet fully paid", or perhaps show the pay so far and some text or graphic showing those shifts still receive pay.

New server is up, and stats have been updated on the website!

We now have a new server up at us2.bitminter.com:3333 (Stratum only)

Use it as backup or primary, and keep mint.bitminter.com:3333 as your other server. It's helpful to have a failover server in case you lose connectivity with your primary one. For some users the latency to us2 will be lower as well, which may help a tiny bit with rejected (stale) work. More servers are coming soon.

Website updates:

A lot of people misinterpret the numbers on the website, so I'm trying to remove stats that are either not really useful or that people tend to misunderstand, and instead add data that is both easy to understand and generally useful.

• Shifts page: now displays pay/hour for mining with 1 TH/s. It's very simple. Work during a shift was paid at a certain rate (which varies with luck). You did a certain amount of work in that shift. Seeing it from this angle makes it much easier to understand the relation between work and pay. No more wondering how much you will get paid per block, and how that is affected by your work in the shifts.
• Workers page: data per "current round" removed. Instead you can choose between "all time" and "since reset". There is now a "0.0" reset button. This works like the trip meter of an odometer in a car. Hit the reset button any time you want. Afterwards you can see how much work the worker has done since that time. And the total amount of work the worker has ever done is always available.
• Account details: "expected pay per block" has been removed. This was the source of most misunderstandings and caused trouble to no end. It invited you to try and understand the relation between work and pay in a very complicated way. You can of course still see how much you got for each block in your transaction history.
• Live stats: user-data per BTC and NMC rounds removed. These numbers didn't matter. You can still see your data for the current shift, which does matter. The total (pool wide) amount of work per round is still kept for luck/cdf stats, which is useful.

You don't need to keep all the shifts at full hashrate to get proper pay (common misconception). You should expect in any pool that if you don't mine for 4-6 hours then your income drops a bit.

For 24/7 miners, if you have x% of the hashrate then you are getting x% of the coins. You are earning the same coins as before. The payouts also happen at the same time as before. For you nothing has changed. When we find a block the payout to your account is the same as before.

For part time miners, if you mine 1 hour and leave the chance is now higher that you get extremely high pay or extremely low (even zero) pay. If we get more pool hashpower in the future then the shifts will be extended again to reduce the luck factor for part time miners.

If you mine 24/7 and you think zero pay shifts are hurting you, why do you think that? Isn't the important thing that you get the same pay as always? Do you really care that the calculations have changed, if they always produce the same numbers for you as before?

Did you make any firewall changes?

The connection refused is not coming from the Bitminter server.

Everything looks fine on this end. Are you unable to connect to stratum on port 3333?

Green is CDF from just above 0% to 33%. Black is from 33% to 66%. And red is CDF from 66% to just below 100%.

CDF at X% means that in X% of cases we would have found 1 or more blocks with this amount of work at this difficulty.

CDF never reaches 100%. As time approaches infinity, CDF approaches 100%. CDF can also never be 0%. That would mean we found a block with an amount of work that can never find a block.

You are never certain to find a block. Ever. Just like you are never guaranteed to win the lottery, no matter how many times you play.

It could happen that bitcoin just stops right now and noone at any pool ever finds another block. It just isn't very likely. The chance that all the hashpower in the world finds one or more blocks in a week's time is something like 99.999999999999999%. Exercise for the reader: find the actual chance, instead of just putting in lots of nines like I did.

But it's never 100% certain. Not even with a hashpower of 100 yottahashes per second mining for 100 years.

9 lucky blocks in a row. 2-3 of them very lucky. That's pretty good.


Hmm, maybe I should make it possible to switch the shift payout display so you can choose.

I'll look into adding some bitcoin price info.

You're welcome.

Profitability does look scary right now, with bitcoin dropping like a rock and the only thing going "to the moon" being the difficulty.

It's not going to look so good until the wind turns.

Everyone's work per shift is less because the shift is shorter. Your percentage of work in the shift is the same.


The shorter shifts do not change what you get per block. If you have x% of the pool hashpower then you get x% of the coins that are paid out. Same as always.


If you got 0.01 BTC per block then you will keep getting that until your hashrate or the pool hashrate changes (changing your percentage of pool hashpower).

Pay per block is very confusing and pretty much irrelevant though. Better shift stats coming soon, which will be much more useful.

You can still see how much you got paid for a block in your transaction history.  Do you still need the prediction of what you might get if we found a block right now?

The reason I want to remove it is that it has caused so much confusion and invalid interpretations of how the reward system works. I want to remove stats that people don't understand and that leads to wrong assumptions, and replace it with stats that are understandable.

When people see they get low pay per block until they have mined 10 shifts, they assume they have to mine for free to ramp up the shifts before they get proper pay. That's incorrect and the logic is wrong. But the stats shown today easily lead people down this train of thought.

I'd rather have people look at how they mined from 06:00 to 09:00 and got paid X coins for that work. Then they worked from 09:00 to 12:00 and got paid Y coins for that. Here you see clearly what is happening. There is no mining for free to ramp up the shifts, and the stats shown do not lead you to that wrong assumption either.

I think stats that lead people to broken logic and misinterpretation are best removed.


Sorry, I had forgotten to update the score/completion display in the livestats. If you refresh the page you should see the correct shift completion percentage. You may need to clear your browser cache.

To show how your mining has been paying, which is better?

• Pay per proof-of-work (share) in bitcoins e.g. 0.0000000002056180 BTC
• Pay per billion proofs-of-work (shares) e.g.0.20561798 BTC
• Pay per proof-of-work (share) in satoshis e.g. 0.02056180 satoshis
• Pay per TH/s e.g. 0.0000000478741685 - what 1 TH/s earned per second
• Pay per TH/s/shift e.g. 0.0005170410195979 (3 hour shift) or 0.0010340820391959 (6 hour shift) - what 1 TH/s earned for a variable time period
• Pay per TH/s/hour e.g. 0.0001723470065326 - what 1 TH/s earned per hour
• Pay per TH/s/day e.g. 0.0041363281567834 - what 1 TH/s earned per day
• Something else?