Hamming Distance Calculator: characters, binary and integer XOR

Real-world use cases

• Check error-correcting code coverage

  You are designing a fixed-width code and need to know how many bit errors it can survive. Enter two valid codewords in binary mode and read their distance; the minimum distance across all pairs tells you the guarantee. A minimum distance of 3 detects two bit flips and corrects one, so if any pair you test comes back as 2, that pair is too close and the code will not correct a single-bit error between them.

• Compare two perceptual hashes for duplicate images

  Two photos produced 64-bit pHashes and you want to know if they are near-duplicates. Paste both hashes as integers, switch to integer mode, and read the bit distance from the XOR. A distance under about 10 out of 64 usually means the same image at a different size or compression, while a large distance means genuinely different pictures. The BigInt math keeps the full 64 bits exact.

• Spot a single-bit flip in transmitted data

  A register or packet came back different from what you sent and you suspect line noise rather than a logic bug. Put the expected and received values side by side in binary mode; a distance of exactly 1 points to a single bit flip at a specific position, which the highlighted aligned view pinpoints for you. A larger distance suggests a framing or addressing problem instead.

• Teach or check a coding-theory homework problem

  A student is learning Hamming codes and needs to verify distances by hand. Enter the two binary strings, confirm the count, and use the highlighted positions to see exactly which bits the textbook answer is counting. Share the URL with a classmate and the same two inputs and mode open for them, so a study group can check each other without re-typing.

Common pitfalls

• Feeding two inputs of different lengths. Hamming distance is only defined when both sides have the same number of positions; there is no insert or delete to absorb the gap. If you need to compare strings that can differ in length, use edit distance (Levenshtein) instead, which this tool links to.

• Confusing Hamming distance with edit distance. They agree only when no insertions or deletions are involved. For abc versus acb, Hamming counts 2 substitutions while edit distance is also 2, but for off-by-one shifts they diverge sharply, so do not quote one number when you mean the other.

• Padding integers to different widths and expecting a different answer. In integer mode the result is popcount of the XOR, which already ignores leading zeros, so 0001 versus 0010 and 1 versus 2 give the same distance of 2. Do not hand-pad before entering; just type the plain numbers.

Privacy