Vibe Engines is free, interactive AI engineering & system design — runs in your browser. Pick a learning path
Password Entropy Calculator
A password strength calculator based on entropy. Type a password (analyzed only in your browser) to see the character pool it draws from, its entropy in bits, a strength rating, and an estimated time to crack at realistic offline attack speeds. Explains why length beats complexity and where the entropy model’s limits are.
Time to crack (offline, fast hash @ 1010/s): about 8,561 centurys
This is the theoretical maximum for the character set, assuming every character is random. Real, memorable passwords use words and patterns attackers guess first, so practical strength is lower — the fix is genuine randomness (a password manager or random passphrase).
Password strength is really one question: how many guesses would an attacker need? Entropy answers it in bits, on a scale where every extra bit doubles the work. This calculator measures the theoretical entropy of whatever you type, rates it and turns it into an average time to crack, all inside your browser.
What it computes
pool = 26 (any a–z) + 26 (any A–Z) + 10 (any 0–9) + 33 (anything else)
bits = length × log₂(pool)
guesses = 2^(bits − 1) half the space, on average
time = guesses / 10,000,000,000 per second- Pool is the size of the character sets your password draws from. One character from a class adds the whole class.
- Bits is the entropy if every character were picked at random from that pool.
- Guesses is the expected number of tries: on average an attacker finds the password halfway through the space.
- Rating: under 28 bits is Very weak, under 40 Weak, under 60 Fair, under 80 Strong, and 80 or more Very strong. The bar fills at 100 bits.
Times are printed in the largest fitting unit, as powers of ten beyond a million years, and as “longer than the age of the universe” past 10¹⁷ years.
A worked example
The default is Tr0ub4dour&3, 12 characters.
- Classes: lowercase (r, o, u, b, d), uppercase (T), digits (0, 4, 3) and a symbol (&), so the pool is
26 + 26 + 10 + 33 = 95. - Bits:
12 × log₂(95) = 12 × 6.570 = 78.8, shown as 79 bits, which is just under the 80-bit line, so the rating is Strong. - Guesses:
2^77.8 ≈ 2.7 × 10²³. - Time:
2.7 × 10²³ / 10¹⁰ = 2.7 × 10¹³ seconds, about 8,561 centuries, or roughly 856,000 years.
| Password | Length | Pool | Bits | Rating | Average crack time |
|---|---|---|---|---|---|
password | 8 | 26 | 37.6 | Weak | about 10.4 seconds |
P@ssw0rd | 8 | 95 | 52.6 | Fair | about 3.8 days |
Tr0ub4dour&3 | 12 | 95 | 78.8 | Strong | about 856,000 years |
correct horse battery staple | 28 | 59 | 164.7 | Very strong | longer than the age of the universe |
The table shows the formula’s blind spot. Real attackers try password and P@ssw0rd within their first few guesses, whatever the bits say, and the famous xkcd comic on password strength estimates about 28 bits for this kind of pattern, an uncommon word with substitutions plus a symbol and a digit, against 44 bits for four random common words.
Assumptions and limits
- Random characters assumed. Words, names, dates and substitutions are not detected, so human-chosen passwords score far higher than their real strength.
- Class-based pool. One symbol adds 33 to the pool; an accented letter or emoji also counts as a symbol, and the length counts code units, so an emoji counts as two.
- One attack speed. 10 billion guesses a second models a fast unsalted hash. Slow password hashes cut that by thousands of times; leaked plaintext needs no guessing at all.
- Averages. An attacker can get lucky early; the figure is the expected time, not a minimum.
How to use the result
- Generate, do not invent. Let a password manager create random passwords; then the bits shown are the bits you actually get.
- Aim high for what matters. About 80 bits of true randomness, such as 13 random printable characters or six to seven random words from a large list, is strong against offline attacks.
- Prefer length over symbol rules; each added random character is worth 5 to 7 bits.
- Store with a slow hash if you run a service. Argon2, scrypt or bcrypt multiply the attacker’s cost for every user.
How it works
- Detects the character pool your password draws from (26 + 26 + 10 + 33).
- Entropy = length × log2(pool size), shown in bits.
- Average crack time at 10¹⁰ guesses/second against a fast hash.
- Analyzed entirely in your browser; never transmitted.
Frequently asked questions
What is password entropy?
Entropy measures how unpredictable a password is, in bits. If a password is drawn at random from a pool of P possible symbols and is L characters long, its entropy is L times log2 of P. Each extra bit doubles the number of guesses an attacker must try, so a 60-bit password is about a billion times harder to brute-force than a 30-bit one. It is the cleanest single number for comparing strength, with the big caveat that it assumes the characters really are random.
How is the character pool counted?
The tool adds 26 if the password contains any lowercase letter, 26 for any uppercase letter, 10 for any digit and 33 for anything else, which covers the printable symbols and the space. A password using all four classes therefore gets a pool of 95, the printable ASCII set. One symbol is enough to add the full 33, so the pool describes the character sets an attacker would have to search, not how varied your particular characters are.
Why does length matter more than adding symbols?
Because length multiplies entropy while the pool only changes the logarithm. Growing the pool from 26 lowercase letters to all 95 printable characters raises each character’s contribution from about 4.7 to 6.6 bits, a modest gain. Each extra character, by contrast, adds another full 4.7 to 6.6 bits. That is why a long random passphrase usually beats a short password full of symbols, and why current guidance from standards bodies favors length over composition rules.
How realistic is the "time to crack"?
It is an order-of-magnitude estimate, not a guarantee. It assumes an offline attacker who has stolen the password hashes, needs on average half the possible guesses, and makes 10 billion guesses a second, a plausible rate for a GPU rig against a fast, unsalted hash. Against a deliberately slow hash such as bcrypt, scrypt or Argon2 the rate is thousands of times lower, and an online login with rate limiting is slower still. Use it to compare passwords, not as a countdown.
Why is my real password weaker than the bits suggest?
Because the formula assumes every character was chosen uniformly at random, and human-chosen passwords are not. People use dictionary words, names, dates, keyboard patterns and predictable substitutions such as P@ssw0rd, and attackers exploit exactly that with wordlist and rule-based attacks. The default example, Tr0ub4dour&3, a variant of the password in the well-known xkcd comic on password strength, scores 79 bits here, yet the comic estimates about 28 bits for that pattern. The fix is genuine randomness from a password manager or a random passphrase.