--- title: "IEEE 754 Converter - Float, Hex and Binary | CanDoYa" description: "IEEE 754 converter for 32-bit floats and 64-bit doubles. Convert decimal to hex or binary, decode raw bits, and inspect every field locally." url: https://candoya.com/tools/ieee-754-float-converter/ locale: en type: browser-tool --- # IEEE 754 Float Converter Category: Developer ## What is an IEEE 754 converter? An IEEE 754 converter shows how a decimal value is stored as a binary32 float or binary64 double. Convert decimals into binary and hexadecimal words, or decode raw bits back into a number. The field view separates the sign, biased exponent, and fraction while the exact value reveals any rounding. The conversion runs entirely in your browser with the same typed-array operations used to read and write binary data in JavaScript. Choose 32-bit for C floats, GPU buffers, and many binary protocols, or 64-bit for JavaScript numbers and C doubles. Both directions preserve signed zero and identify normal values, subnormals, infinity, and NaN. ## How to use it 1. **Choose a direction** - Select Decimal to bits to encode a number, or Bits to decimal to decode a complete hexadecimal or binary word. 2. **Choose the precision** - Use binary32 for a 32-bit float or binary64 for a 64-bit double. The required bit and hex lengths update with the selection. 3. **Enter the value** - Type a decimal, scientific notation, Infinity, or NaN. In decode mode, paste an 8 or 16 digit hex word or a 32 or 64 bit binary word. 4. **Inspect and copy** - Read the field breakdown, classification, decoded number, and exact stored value. Copy the raw hex, binary, or exact decimal result. ## Who it's for - **Software developers** checking float fields in network packets, file formats, memory dumps, or serialization code. - **Students** learning how a sign bit, biased exponent, and fraction combine into an IEEE 754 value. - **Embedded and graphics engineers** translating binary32 register values, sensor data, and GPU constants between decimal and hex. - **Data analysts** investigating why a decimal such as 0.1 changes when stored at a lower precision. ## Expert note The decimal 0.1 becomes 0x3DCCCCCD in binary32, whose exact value is 0.100000001490116119384765625. In binary64 it becomes 0x3FB999999999999A, much closer but still not exactly one tenth. The difference comes from rounding an infinite repeating base-2 fraction to the available 23 or 52 stored fraction bits. ## Frequently asked questions ### Is this IEEE 754 converter free? Yes. The converter is free, requires no account, and has no conversion limit. It supports IEEE 754 binary32 and binary64 in both directions, including decimal, hexadecimal, and raw binary representations. ### Are my numbers uploaded? No. Every conversion runs locally in your browser using ArrayBuffer and DataView. Decimal values, memory words, and bit patterns are not sent to a server, which makes the tool suitable for private debugging data. ### What formats and limits are supported? The tool supports 32-bit binary32 and 64-bit binary64 values. Decode mode requires exactly 8 or 16 hexadecimal digits, or exactly 32 or 64 binary bits. Half precision, bfloat16, decimal floating point, and extended 80-bit formats are outside its scope. ### How do I convert a float to hex? Choose Decimal to bits, select 32-bit float or 64-bit double, and enter the decimal value. The hexadecimal word updates immediately. For example, 1.0 is 0x3F800000 as binary32 and 0x3FF0000000000000 as binary64. ### Can it convert hex back to a float? Yes. Choose Bits to decimal, select the matching precision, and paste the complete hex word with or without 0x. The tool decodes the number and displays its sign, exponent, fraction, classification, binary word, and exact decimal value. ### Why is 0.1 not exact in floating point? One tenth has an infinite repeating expansion in base 2, just as one third repeats in base 10. IEEE 754 must round that expansion to a finite significand. Binary32 keeps 24 significant binary digits including the implicit leading bit, while binary64 keeps 53. ### What are subnormal numbers? Subnormal values use an all-zero exponent field and a nonzero fraction. They fill the gap between zero and the smallest normal value with reduced precision. The converter labels them explicitly and applies the fixed minimum exponent for the selected format. ### Does the converter handle negative zero, infinity, and NaN? Yes. It preserves the sign bit of negative zero, recognizes positive and negative infinity, and classifies any all-ones exponent with a nonzero fraction as NaN. When decoding NaN, the raw payload bits remain visible even though JavaScript displays the numeric result simply as NaN. ## Related tools - [Number Base Converter](https://candoya.com/tools/number-base-converter/) - [Data Storage Converter](https://candoya.com/tools/data-storage-converter/) - [HEX to RGB Converter](https://candoya.com/tools/hex-to-rgb/) - [IPv4 Subnet Calculator](https://candoya.com/tools/ipv4-subnet-calculator/) - [JSON Formatter](https://candoya.com/tools/json-formatter/) --- *Markdown edition of , published for AI agents and other automated readers. Index of key pages: .*