Contact

Tools  /  Security  /  Password Entropy Calculator

SECURITY

Password Entropy Calculator

The actual crack-time economics behind a password's bit count -- across online, offline-slow-hash, offline-fast-hash, and nation-state guessing rates -- plus a charset-length reference grid and a passphrase-vs-password entropy comparison.

More Security tools

JWT Decoder & Generator HMAC Generator Hash Identifier Random Token Generator UUID Validator OAuth URL Builder

What this tool does

Computes a password's theoretical entropy -- bits of randomness, from its character-set size and length -- and translates that bit count into actual crack-time estimates across four real attacker classes, from a rate-limited login form to a nation-state-scale cracking array. This is a focused, math-and-economics companion to this site's Password Strength Checker (which scores real-world predictability patterns like dictionary words and keyboard walks) and Entropy Calculator (general-purpose Shannon entropy for any text) -- here specifically for understanding what a given entropy figure actually buys you against a real attacker.

The entropy formula, and its one big assumption

Character-set entropy is bits = length × log2(pool size) -- each character contributes log2(pool) bits, assuming every character was chosen uniformly at random from that pool. That assumption is the whole story: a password that could come from a 94-character pool but was actually a dictionary word with a capital letter and "!" appended has nowhere near its charset-implied entropy, because a real attacker doesn't guess uniformly at random -- they try common passwords, dictionary words, and known patterns first. This tool computes the theoretical maximum; for a prediction of how an actual cracking tool would perform against a specific password's real patterns, use the Password Strength Checker instead.

What a bit count actually means in crack time

Entropy only becomes meaningful once it's translated into time against a specific attacker, because attackers differ by many orders of magnitude in guessing speed. A rate- limited login form might allow a realistic 100 guesses an hour. An offline attack against a properly slow hash like bcrypt or Argon2 might manage roughly 10,000 guesses a second on capable GPU hardware. An offline attack against a fast, unsalted hash like raw SHA-256 can exceed 10 billion guesses a second on the same hardware -- a roughly million-fold difference driven entirely by hashing algorithm choice, not password strength. And a nation-state-scale cracking array pushes that further still. The same password can be "uncrackable in practice" against one of these and trivial against another.

Passphrases: a different way to get the same bits

A Diceware-style passphrase built from truly random words out of a 7,776-word list contributes about 12.9 bits per word (log2(7776)) -- four random words is roughly the same entropy as a 10-character fully-random mixed-case-plus- symbol password, while usually being considerably easier to actually remember. The Passphrase vs Password tab compares the two directly. This site's Passphrase Generator produces real Diceware-style passphrases using genuine rejection-sampled randomness.

Privacy

Everything runs entirely in your browser. Nothing you type here is sent anywhere.

FAQ

Frequently asked questions

Does a high entropy score mean a password is actually strong?
Only if the password really was generated uniformly at random from its charset. A dictionary word with substitutions can have high charset-implied entropy while being one of the first few thousand guesses a real attacker tries. For a realistic, pattern-aware assessment, use the Password Strength Checker instead.
Why do the crack-time estimates vary so much for the same password?
Because attacker guessing speed varies by many orders of magnitude depending on what's being attacked -- a rate-limited login form, an offline attack against a slow hash like bcrypt, an offline attack against a fast hash like raw SHA-256, or a large-scale cracking array. The same entropy can mean "centuries" against one and "minutes" against another.
Why is an offline attack against a fast hash so much worse than against bcrypt?
Hashing speed, not password strength. bcrypt and Argon2 are deliberately slow and memory-hard by design, capping an attacker to roughly thousands of guesses per second even on capable hardware. A fast, unsalted general-purpose hash like raw SHA-256 can be computed billions of times a second on the same hardware -- a difference of roughly six orders of magnitude that has nothing to do with the password itself.
How much entropy does a Diceware passphrase word actually add?
About 12.9 bits per word, from log2(7776) -- the size of the standard Diceware word list. Four random words is roughly 51.6 bits, comparable to a 10-character fully-random password using the full 94-character printable ASCII set.
Why does adding one more character matter so much?
Entropy scales with the exponent, not linearly -- each additional character multiplies the total keyspace by the pool size, while the crack-time estimates double or grow faster with each added bit. The charset-vs-length grid shows concretely how much length buys versus charset size at a glance.
How is this different from the general Entropy Calculator?
The general Entropy Calculator measures the actual information content of whatever text you give it (Shannon entropy), useful for any kind of data. This tool is password-specific: it computes theoretical charset-based entropy and translates it directly into crack-time economics against real attacker classes.
Is anything I type here sent anywhere?
No. Every calculation runs entirely in your browser -- this tool has no server component.
Theme Accent