How Bitcoin mining works: Follow the hash, get the coin

Bitcoin mining still comes down to the same simple loop it has since 2009: run a hash, check the result, repeat until the network says you win. The prize is newly minted coins plus fees, but the mechanics have tightened since the April 2024 halving cut the block subsidy to 3.125 BTC. For readers watching ETFs and exchange prices, the question is what actually happens between pressing start on a rig and seeing new coins in a wallet.

Hashing the block header

Miners first pull pending transactions from the mempool and assemble them into a candidate block. They then run the block header through SHA-256 twice. The header includes a version field, the previous block hash, the Merkle root, a timestamp, the current difficulty bits, and a 32-bit nonce that miners change on every try.

Each attempt produces a 64-character hexadecimal string. If that string is lower than the current target, the block is valid. The target resets every 2016 blocks so the average time between blocks stays near ten minutes. No miner can predict the outcome; they simply generate new nonces until one hits.

The process is closer to a lottery than to traditional computation. The hardware does not solve an equation; it guesses until the network accepts one guess as correct. Success depends on hash rate, electricity cost, and luck.

Proof-of-work competition

Every miner on the network is racing against the same target at the same time. The first to produce a qualifying hash broadcasts the block. Other nodes verify the transactions and the hash, then append the block to their copy of the chain. The winner receives the subsidy and any fees attached to the included transactions.

Because blocks arrive roughly every ten minutes, the probability of finding the next block is proportional to a miner’s share of total hash rate. A single modern ASIC can produce trillions of hashes per second, yet even the largest facilities hold only a small slice of the global total.

Difficulty adjustments keep the race fair. When hash rate drops, the target loosens; when hash rate rises, the target tightens. The adjustment happens automatically and prevents the network from speeding up or slowing down for long.

Current subsidy and fees

After the 2024 halving the subsidy sits at 3.125 BTC per block. At 144 blocks per day the network now issues roughly 450 new coins daily. The next reduction is scheduled around 2028, when the subsidy will fall to 1.5625 BTC.

Transaction fees have become a larger slice of miner revenue. In quiet periods fees still represent only a few percent of income, but during congestion spikes they can exceed the subsidy for individual blocks. Post-halving economics therefore reward miners who can operate at lower cost per hash.

The 21-million-coin cap remains fixed. Roughly 19.95 million coins have already been mined, so the remaining issuance continues on the same geometric schedule regardless of price or hash rate.

Hashrate and difficulty in 2026

Global hash rate has spent more than 300 days below its 2025 peak. Recent readings hover between 900 EH/s and 1 EH/s. The drop reflects miners turning off machines when revenue falls below power cost.

Difficulty has followed hash rate downward in a series of adjustments. Lower difficulty reduces the average number of hashes needed to find a block, which helps remaining operators stay profitable. The network’s automatic response keeps block times stable even as participants leave.

Three mining pools currently control more than half of all blocks. Foundry USA accounts for roughly 27 percent, followed by AntPool and ViaBTC. Pool payouts smooth individual miners’ earnings, but concentration raises questions about long-term decentralization.

ASIC hardware and efficiency

Application-specific integrated circuits dominate Bitcoin mining. The newest models reach efficiencies below 10 joules per terahash. Hydro-cooled units from Bitmain and MicroBT lead the current generation; air-cooled machines sit a few joules higher.

Older S19-series rigs drew roughly 25–30 J/TH. Replacing those machines with sub-10 J/TH units cuts electricity cost per hash by more than half. Operators still running older fleets face steeper losses when hash price declines.

Power draw remains the largest operating expense. A single high-end ASIC can consume several kilowatts continuously, so electricity contracts and cooling infrastructure determine which miners survive each cycle.

Energy consumption and cost

Annual electricity use for the Bitcoin network is estimated between 138 and 204 terawatt-hours. That range places mining at roughly half a percent of global electricity demand. The exact figure fluctuates with hash rate and regional power prices.

Many large facilities now locate near stranded or renewable energy sources. Hydroelectric sites in Texas, Wyoming, and parts of Canada host clusters of machines that can power down quickly when grid demand rises. This flexibility sometimes earns miners payments for curtailing load.

Publicly traded mining companies have begun shifting capacity toward AI and high-performance computing workloads. Contracts worth tens of billions have been announced, offering steadier revenue than hash price alone. The shift reduces network hash rate but does not change the underlying protocol.

Profitability after the halving

Hash price, the daily revenue per unit of hash rate, sits near multi-year lows. At current levels only the most efficient machines remain consistently profitable. Operators with higher power costs or older hardware have powered down or pivoted.

Difficulty adjustments provide a partial cushion. When hash rate falls, remaining miners need fewer hashes on average to find a block, which raises their effective revenue per machine. The network therefore self-corrects without external intervention.

Transaction-fee revenue offers a second buffer. During periods of high on-chain activity, fees can double or triple a miner’s income for individual blocks. Fee pressure tends to rise when price volatility increases or when large transfers move on-chain.

Pool structure and payout models

Most individual miners join pools to reduce variance. Full Pay-Per-Share and similar models pay a steady daily rate based on contributed hash rate, regardless of whether the pool finds the next block. The pool operator takes a small fee and manages block construction.

Pool dominance has stayed relatively stable since 2024. The top three pools continue to mine more than half of all blocks, while smaller pools and solo miners account for the remainder. No single pool has yet approached the 51 percent threshold that would allow unilateral chain reorganizations.

Pool operators also decide which transactions to include and at what fee rate. During high-demand periods they can prioritize higher-fee transactions, influencing both miner revenue and network congestion.

Outlook for miners and the network

The next halving will cut the subsidy again, further tilting economics toward efficiency and fee capture. Operators who cannot reach sub-10 J/TH performance will face continued pressure to exit or diversify. Difficulty will keep adjusting, maintaining the ten-minute block interval as long as at least some hash rate remains online.

For users holding or trading Bitcoin, the mining process itself stays unchanged. Coins are still created by finding a hash below the target, and the network still self-regulates through difficulty adjustments. The only variables that shift are who participates and how much each participant earns.