Binary Translator

How text becomes binary

Computers do not store letters. They store numbers, and every number is a row of switches that are either off (0) or on (1). Translating text to binary is therefore a two-step job: first look up the number assigned to each character, then write that number in base 2 and pad it to a fixed width of eight digits. Eight is not arbitrary — eight bits make one byte, the smallest chunk of memory a processor addresses directly, so every group here lines up with one real byte in a real file.

Worked example: "Hello" byte by byte

ASCII assigns H = 72, e = 101, l = 108, l = 108, o = 111. Convert 72 to base 2 and you get 1001000, which is only seven digits, so a leading zero pads it to 01001000. Repeat for each letter and you get the full translation:

So "Hello" is 01001000 01100101 01101100 01101100 01101111 — five characters, five bytes, 40 binary digits. Check one by hand if you like: 01001000 reads as 64 + 8, which is 72, which is capital H. Paste that string back into the tool with the direction set to binary to text and the word comes straight back out.

Reading a byte without a calculator

The eight positions carry the values 128, 64, 32, 16, 8, 4, 2 and 1, left to right. Add up the values sitting under a 1 and you have the byte. For 01100101 that is 64 + 32 + 4 + 1 = 101, the letter e. This is the whole trick behind every "decode the binary" puzzle in a homework sheet or an escape room, and it is worth doing once by hand before you trust a converter.

You may also notice that every ASCII letter starts with 0. Original ASCII only defined 128 characters, which fit in seven bits, so the eighth bit was spare. That spare bit turned out to be the hook that made Unicode possible.

Beyond ASCII: what UTF-8 changes

English text is the easy case. The moment you type café, Москва, 東京 or an emoji, one character stops equalling one byte. UTF-8 solves this by making the length variable and using that spare high bit as a flag: a byte starting with 0 is a standalone ASCII character, a byte starting with 110 opens a two-byte sequence, 1110 opens three bytes, and 11110 opens four. Continuation bytes always start with 10, which is what lets a decoder resynchronise in the middle of a stream.

Concretely, é is U+00E9 and encodes as two bytes, 11000011 10101001. A grinning face emoji is U+1F600 and encodes as four bytes, 11110000 10011111 10011000 10000000 — 32 binary digits for a single character. That is why this translator shows a character count and a byte count separately: for plain English they match, and for anything else the gap is the point. If you decode a stray pair of bytes that does not follow those rules, the tool refuses rather than printing a replacement diamond, because a silent U+FFFD hides the actual problem.

What people actually use this for

Three uses dominate. Students working through a computer science unit convert their name to binary and back to prove they understand place value and character encoding. Developers debugging a protocol want to see the exact bit pattern of a header byte, especially when a flag lives in a single bit. And hobbyists building novelty projects — binary clocks, cross-stitch patterns, engraved jewellery, geocaching puzzles — need a clean string of 0s and 1s that they can trust.

For all three, the reverse direction matters as much as the forward one. Paste in the groups you were handed, drop the spaces if you like — the parser strips all whitespace before decoding — and read the message. If a digit is wrong you get told which position broke the rule instead of a blank screen.

Limitations worth knowing

Binary is not encryption. Anyone can decode it in seconds, so it hides nothing; treat it as a display format, not a secret. It is also bulky: eight characters of output for every byte, which is why hexadecimal (two characters per byte) is standard in hex dumps and colour codes. Finally, this tool assumes UTF-8, today's default on the web. Text saved as UTF-16 or a legacy code page such as Windows-1252 will produce different bytes, and pasting those bytes here may fail validation — correctly, because they are not valid UTF-8.