Mining alone, called solo mining, can mean a long wait between rewards. The electricity bills keep arriving the whole time. The reward is uncertain, but the costs keep adding up.
A mining pool lets many miners work together and split the payouts under one set of rules. The interesting part is how a pool measures how much work each miner contributed.
Use a more achievable checkpoint
The pool sets its own target, which is much easier to meet than the network’s block target. A result that meets the pool’s easier target is called a share. At the same pool target, a machine that makes more hashes finds more shares, on average. So shares give the pool evidence of the work each miner contributed, even though most shares are not blocks.
Occasionally a result meets the network target too. That is the block everyone was hoping for. Most shares never become blocks, and that is normal.
Agree how to divide the proceeds
Pools pay miners in different ways. Pay per share (PPS) usually pays for each accepted share, based on what that work is worth on average. Full pay per share (FPPS) usually adds an estimate of transaction fees too. Pay per last N shares (PPLNS) splits the pool’s actual block rewards across its most recent shares. N is the number of recent shares counted, which the pool chooses.
Those labels are only a starting point. Each pool sets its own definitions, fees and payout conditions. The payout method also decides how the ups and downs of block-finding luck are split between the pool and its miners.
The dashboard is an estimate, not a heartbeat monitor
A pool cannot measure a machine’s hashrate directly. It estimates hashrate from the shares it receives over time. Over a short time window, the estimate can jump around even while the machine runs steadily. A longer window smooths out the jumps, but it can still include a period when the machine was offline.
Share difficulty can differ too. A higher-difficulty share stands for more work than an easier one. So just counting shares, without checking their difficulty, can mislead you. Before you call it a problem, make sure you are comparing the same worker, the same time window and the same kind of measurement.
Follow the work past the network port
Imagine the miner looks healthy on its own web page, but the pool shows no accepted work for the worker it should be using. Your job is not done yet. Check the worker name, the pool address it connects to (the endpoint), its current connection status and any details about rejected work (results the pool refused).
A machine can use power without delivering useful work to the right pool. This is why “the miner’s web page loads” and “the miner is contributing” are different conclusions.
Optional reading for this lesson
Continue your study
Reading for this lesson
Further reading on pool participation.
Technical background on pooled mining.