How does Bitcoin mining work? Blocks, hashes and rewards

Bitcoin mining is the process that adds each new block of payments to Bitcoin's ledger and creates new coins. Miners bundle waiting transactions into a candidate block, then guess numbers until the block's hash, a kind of fingerprint, falls below a target. The first to succeed earns new bitcoin plus fees. The guessing does real work, measured in electricity, and that work is what makes the ledger hard to rewrite.

In short

• Mining is a lottery where each ticket is one hash: more computing power buys more tickets, and nobody can cheat the draw.
• The network retunes the difficulty every 2,016 blocks so that a block arrives about every ten minutes, however many machines join.
• Miners are paid newly issued bitcoin plus fees, which is how all bitcoin enters circulation and why the network spends so much energy.

What does a miner actually do?

A miner is a computer, today almost always a specialized chip called an ASIC (application-specific integrated circuit), running software that follows Bitcoin's rules. Each round of its work goes through the same steps.

1. Collect payments. The miner picks valid, unconfirmed transactions from its waiting area, the mempool, usually favoring those that pay the highest fee for the block space they take up.
2. Pay itself. It writes a first transaction, the coinbase, that sends the block reward to its own address.
3. Build a header. It packs a summary of the block into an 80-byte header: a version number, the hash of the previous block, a single fingerprint of all the transactions (the Merkle root), the time, the current target and a number called the nonce.
4. Guess. It hashes the header with SHA-256, twice. If the result is below the target, the block is valid. If not, it changes the nonce and tries again, billions of times a second.
5. Broadcast. The winner sends the block to the network. Every node checks it in a fraction of a second, and miners start the race for the next block on top of it.

That chain of previous-block hashes is what makes a blockchain. Rewriting an old block would mean redoing its puzzle and every puzzle after it, faster than the rest of the world is solving new ones.

What is the puzzle, and why can't anyone shortcut it?

A hash function turns any input into a fixed-length number that looks random. Change one character and the output changes completely, and there is no known way to work backward from a desired output to an input. So the only way to find a header whose hash is below the target is to try one after another, like rolling dice until you roll under a very low number.

Finding a valid hash is hard, but checking one takes a single calculation. That asymmetry is the core of proof of work. A block is accepted because anyone can verify, instantly, that its maker spent the work.

The scale is hard to picture. The nonce field holds only about 4.3 billion values, which a modern machine exhausts in a fraction of a second, so miners also change a spare field in the coinbase transaction, the extra nonce, to get a fresh header to hash, as the Bitcoin developer guide describes. In early November 2026, mining difficulty stood at about 148 trillion, according to mempool.space. By the formula on the Bitcoin Wiki, that means the whole network needs around 640 billion trillion guesses, on average, to find one block.

How does difficulty keep blocks ten minutes apart?

Every 2,016 blocks, about two weeks, every node runs the same calculation. It compares how long those 2,016 blocks actually took with the 14 days they should have taken, and moves the target in proportion. The idea goes back to Satoshi Nakamoto's 2008 paper, which set difficulty by a moving average aimed at a steady number of blocks per hour; the 2,016-block rule came with the software.

A worked example: suppose new machines switch on and blocks start arriving every 8 minutes instead of 10. The 2,016 blocks take 16,128 minutes, about 11.2 days. The network divides 14 by 11.2, gets 1.25, and raises the difficulty by 25%. Blocks slow back to about ten minutes. If miners switch off, the same arithmetic lowers the difficulty. Nobody votes on it; it is a rule every node applies.

This is why more miners do not create more bitcoin. They only raise the difficulty, so each miner's share of a fixed reward shrinks.

What do miners earn?

The winning miner earns two things through the coinbase transaction. The first is the block subsidy: new bitcoin created from nothing, the only way new coins ever enter circulation. It started at 50 bitcoin per block in 2009 and halves every 210,000 blocks. Since the April 2024 halving, it has been 3.125 bitcoin. The second is the fees from every transaction in the block, which rise when many people want block space at once. Coins from a coinbase can't be spent until 100 more blocks have been added, so a miner can't profit from a block the network later abandons.

Because one machine would almost never win a block, nearly all miners join pools. A pool gives its members an easier target, counts the near-misses they submit, called shares, as proof of how much work each did, and splits each reward it wins in proportion, minus a fee. Mining in a pool turns a lottery into a steadier, smaller income.

Why does Bitcoin mining use so much electricity?

Every guess costs a little electricity, and the network makes an enormous number of them. In early November 2026, mining ran at about 1,000 EH/s (one exahash is a billion billion hashes per second), according to mempool.space. A University of Cambridge study published in April 2025 estimated the network's annual use at about 138 TWh, roughly 0.5% of the world's electricity, and found that 75.4% of the mining it could trace was in the United States. The US Energy Information Administration put crypto mining at 0.6% to 2.3% of US electricity in a preliminary estimate published in February 2024.

The electricity is not a side effect; it is the security. To rewrite history, an attacker would have to outspend all honest miners on hardware and power, which on a network this size would cost more than the attack could likely earn.

Still argued

Whether that energy is well spent is argued. Supporters say miners can use power that would otherwise go to waste, such as gas flared at oil wells or surplus hydro and wind, and can switch off quickly when a grid is strained. Critics point to the emissions, the noise near mining sites, and the electronic waste from machines that are obsolete within a few years. The same Cambridge study, which surveyed miners covering about 48% of the network, estimated that 52.4% of mining energy came from renewables and nuclear and 38.2% from natural gas, with emissions of about 39.8 million tonnes of CO2-equivalent a year. Both sides use these figures; they disagree about what they imply.

Can you mine bitcoin at home?

Technically yes, but the numbers explain why almost nobody does. Take a machine of the kind sold for industrial mining in 2024 and 2025, rated at about 200 TH/s (200 trillion hashes per second) and drawing about 3,500 watts, and set it against a network of 1,000 EH/s.

Step	Calculation	Result
Your share of all hashing	200 TH/s ÷ 1,000,000,000 TH/s	1 in 5 million
Blocks you'd expect per day, alone	144 blocks ÷ 5,000,000	0.0000288
Expected wait to win one block, alone	1 ÷ 0.0000288 days	about 95 years
Expected subsidy per day, in a pool	0.0000288 × 3.125 bitcoin	about 0.00009 bitcoin
Electricity per day	3.5 kW × 24 hours	84 kWh
Electricity cost at 15 cents per kWh	84 × $0.15	about $12.60

At that power price, the machine earns more than its electricity costs from the subsidy alone only if one bitcoin is worth more than about $140,000, before paying for the machine, the cooling and the pool's fee, and before the next halving cuts the subsidy in half. Industrial miners win by paying far less for power, buying the newest chips in bulk and running them around the clock. This explains the arithmetic; it isn't financial advice.

Mining is also what makes Bitcoin work without anyone in charge: no one chooses who writes the next page of the ledger, and no one can quietly change an old one.

Check yourself

• Without looking back: why is finding a valid block hard, while checking one is easy?
• What happens to the difficulty if half of all miners switch off, and how long does it take?
• Why does a miner with one machine join a pool instead of mining alone?

Sources

• Bitcoin: A Peer-to-Peer Electronic Cash System — Satoshi Nakamoto, 2008
• Mining, in the Bitcoin developer guide — Bitcoin Project
• Difficulty — Bitcoin Wiki
• Bitcoin hashrate and difficulty — mempool.space, 2026
• Cambridge study: sustainable energy rising in bitcoin mining — Cambridge Judge Business School, 2025
• Tracking electricity consumption from U.S. cryptocurrency mining operations — U.S. Energy Information Administration, 2024