Home / Wiki / H / Hashrate

Hashrate

Cryptocurrency and Digital AssetsUpdated 05 Nov 2026

Hashrate is the number of hash calculations performed per second by a mining device, a mining pool or an entire network that uses Proof of Work (PoW). Each calculation is one guess at the puzzle that wins the right to add the next block. The figure is used to compare machines, to divide rewards inside pools and to gauge how much computing power protects a blockchain.

Hashrate is closely tied to mining difficulty: when total hashrate rises, the network raises difficulty so blocks keep arriving on schedule. The two numbers together describe the state of crypto mining on a network. Hashrate is not a feature of networks that use Proof of Stake (PoS), which do not run a guessing contest.

01

Hashrate in plain terms

Imagine a raffle where each ticket is a random guess and you can print as many tickets as your printer allows. Hashrate is how many tickets your printer turns out each second. A faster printer does not make any single ticket more likely to win, but it gives you more tickets in the drum.

The network's hashrate is the combined output of every printer taking part. That total matters to everyone, not just the people printing, because anyone wanting to rig the raffle would have to out-print all of them together.

02

Units of measurement

Hashrate is written in hashes per second (H/s) with standard metric prefixes. Because modern mining hardware is so fast, the larger units dominate:

UnitHashes per secondTypical use
MH/s (megahash)10610^{6}Processors and GPUs on memory-hard algorithms
GH/s (gigahash)10910^{9}Early Bitcoin ASICs, around 2013
TH/s (terahash)101210^{12}A single modern Bitcoin ASIC
PH/s (petahash)101510^{15}A mining farm or small pool
EH/s (exahash)101810^{18}Large pools and the Bitcoin network

Hashrates are only comparable within the same algorithm. A hash of Bitcoin's double SHA256 is far cheaper to compute than one of Scrypt (used by Litecoin (LTC)) or RandomX (used by Monero), so a Monero miner's few kilohashes are not "slower" in any useful sense than a Bitcoin miner's terahashes.

03

Estimating network hashrate

Nobody can count every guess made around the world, so network hashrate is inferred from how quickly blocks are found at the current difficulty. On Bitcoin (BTC) a block at difficulty DD takes, on average, D×232D \times 2^{32} hashes. Dividing by the target block time gives:

H=D×232600H = \frac{D \times 2^{32}}{600}
  • HH is the estimated network hashrate in hashes per second.
  • DD is the current difficulty.
  • 2322^{32} (4,294,967,296) is the expected number of hashes per block at difficulty 1.
  • 600 is the target block interval in seconds.

Worked example: at a difficulty of 100 trillion (101410^{14}), H=1014×4,294,967,296÷600≈7.16×1020H = 10^{14} \times 4{,}294{,}967{,}296 \div 600 \approx 7.16 \times 10^{20} H/s, or about 716 EH/s. Real block times vary by chance, so published charts usually average over several days. Short-term swings in such charts often reflect luck rather than machines switching on or off. Mining pools publish their own hashrate, and adding up pool figures gives a second, independent estimate that analysts compare with the difficulty-based number.

04

Hashrate in mining economics

For an individual miner, hashrate is the main sales figure of a machine, quoted alongside power use. The ratio of the two, energy per terahash, decides whether an Application-Specific Integrated Circuit (ASIC) or a mining rig earns more than it costs to run. A miner's expected share of new coins is roughly its own hashrate divided by the network's, multiplied by the blocks found and the block reward.

Pools use hashrate to divide income. Members submit "shares", easier solutions that prove work was done, and payout methods such as Pay-Per-Share (PPS) and Pay-Per-Last N Shares (PPLNS) credit them accordingly. In cloud mining, customers buy hashrate by the TH/s, and inflated claims about how much hashpower actually exists have been at the centre of several fraud cases.

After each halving, revenue per terahash drops sharply, and the network's hashrate often dips as older machines are retired.

05

Hashrate and security

High network hashrate makes a 51% attack expensive, because an attacker would need more hashing power than all honest miners combined to rewrite recent blocks and achieve double spending. Concentration matters as much as the total: if a handful of pools control most of the hashrate, the network depends on their behaviour. Smaller chains that share an algorithm with a much larger one are especially exposed, since hashpower can be redirected or rented to attack them, as happened several times to Ethereum Classic (ETC).

Last updated 05 Nov 2026