Password Entropy Explained — The Math Behind Password Security
Password entropy is the only objective measure of how long a password would take to crack by brute force. Unlike 'strength meters' that reward character variety with a colored bar, entropy is a precise number in bits derived from the character set size and password length. This calculator gives you the exact bit count and crack time estimate at realistic GPU speeds — no guesswork.
N = character set size, L = password length. Example: a 12-character random password using all 95 printable ASCII characters has entropy = 12 × log₂(95) = 12 × 6.57 = 78.8 bits.
Character Set Sizes
The base of the entropy formula is the character pool your password draws from. Digits only (0–9): N = 10. Lowercase letters: N = 26. Mixed case letters: N = 52. Alphanumeric (letters + digits): N = 62. Full printable ASCII (adds !@#$%^&* and other symbols): N = 95. Adding symbols increases N from 62 to 95 — a factor of 1.53. Adding one more character to the length multiplies possibilities by N. For a 95-char set, length adds more entropy per unit of effort than expanding the character set.
Entropy vs Crack Time
Each additional bit of entropy doubles the number of guesses required. At a realistic offline GPU attack speed of 10 billion guesses/second: 40 bits takes about 54 seconds on average. 50 bits takes about 16 hours. 60 bits takes about 33 years. 70 bits takes about 36,000 years. 80 bits takes about 37 billion years. The exponential relationship is why a 16-character password is not merely 'twice as secure' as an 8-character one — it is astronomically more secure.
Why Entropy ≠ Actual Password Strength
The entropy formula assumes the password is chosen uniformly at random from the full character space. Human-chosen passwords are not random — people pick words, names, dates, and predictable substitutions. 'Password1!' technically draws from a 95-char alphabet at 10 characters, giving 65 theoretical bits of entropy. In practice, attackers find it in seconds because it matches common patterns. True entropy only applies to randomly generated passwords — which is exactly why password managers and generators like this one matter.
Practical Entropy Targets
For most online accounts: 60+ bits of entropy from a randomly generated password provides strong protection even against offline attacks. For password manager master passwords and encryption passphrases: 77+ bits (e.g., 6 Diceware words or a 12-character random full-ASCII password). For cryptographic keys and API secrets: 128 bits minimum. NIST SP 800-63B recommends allowing and encouraging long passwords — up to at least 64 characters — and avoiding mandatory complexity rules that do not meaningfully increase true entropy.
Frequently Asked Questions
What is password entropy?
Entropy is a measure of unpredictability, expressed in bits. For a password, it is the base-2 logarithm of the total number of possible passwords of that length drawn from that character set. A password with H bits of entropy requires on average 2^(H-1) guesses to crack by brute force. Entropy is what security researchers actually measure — not password 'strength scores' based on character rules.
How many bits of entropy is considered secure?
For randomly generated passwords: 56 bits is a minimum for low-risk temporary use, 72 bits is the practical security threshold against offline GPU attacks at 10 billion guesses/second, and 128 bits is used for cryptographic keys and API secrets. NIST SP 800-63B does not set a minimum entropy requirement for user-chosen passwords because human-selected passwords are never truly random regardless of complexity rules.
What does entropy tell me that password length alone does not?
Entropy incorporates both length and character set size simultaneously. A 12-character lowercase-only password has 56 bits of entropy (log₂(26^12)). A 12-character password from full ASCII (95 chars) has 78 bits. Entropy lets you compare passwords constructed differently on equal footing — for example, a 6-word Diceware passphrase (77 bits) versus a 12-character complex password (78 bits) are roughly equivalent in resistance to brute force.
Does adding one character really matter?
Significantly. Each character added multiplies the search space by the character set size. Adding one character to a 95-character-set password multiplies possibilities by 95 — adding about 6.5 bits of entropy, or roughly doubling crack time 6–7 times over. Going from 12 to 16 characters in a random 95-char password takes entropy from 78 to 105 bits — the difference between decades and longer than the universe's age.
What is the recommended minimum entropy for passwords?
For online accounts with rate limiting and lockout policies, 40 bits is often adequate against online guessing. For offline attacks where crackers operate at GPU speed against stolen hashes, 72 bits is the practical threshold. For cryptographic applications — API keys, session tokens, encryption keys — a minimum of 128 bits. NIST's current guidance favors password length over complexity requirements.
How does entropy relate to crack time?
Given H bits of entropy, an average brute-force attack requires 2^(H-1) guesses. At 10 billion guesses per second (GPU MD5 benchmark), each additional bit doubles crack time. At 60 bits: about 33 years. At 70 bits: 36,000 years. At 80 bits: 37 billion years. At 90 bits, you are well beyond any realistic attack timeline regardless of near-future hardware improvements.
What are the character set sizes for common password types?
Digits only: 10. Lowercase letters: 26. Uppercase + lowercase: 52. Alphanumeric (letters + digits): 62. Full printable ASCII (adds symbols like !@#$%): 95. Extended ASCII: 256. Adding symbols to an already-long password has diminishing returns compared to simply adding more characters — length scales the exponent, character set scales the base.