An FPU (Floating-Point Unit) is a specialized part of a [microprocessor](https://www.ampheo.com/c/microprocessors) that performs floating-point arithmetic operations—like addition, subtraction, multiplication, and division—on real numbers (with decimals).

**What Is Floating-Point Arithmetic?**
Floating-point numbers represent real numbers (e.g., 3.14, -0.001, 2.718) in a format that allows a wide range of values.
They're commonly used when:

* High precision is required
* Very small or very large numbers are involved
* Decimal calculations are necessary

**What Is an FPU?**
FPU = Hardware component that accelerates math operations involving:
* float (32-bit floating-point)
* double (64-bit floating-point)
* Scientific or engineering computations

**Why and When Do You Need an FPU?**
**You need an FPU when your application includes:**

**Without an FPU?**
If your MCU has no FPU, floating-point operations are done in software, which is:

* Much slower
* More power-hungry
* Larger in code size

Example: An [STM32F103](https://www.ampheo.com/search/STM32F103) (Cortex-M3, no FPU) is much slower at float math than an [STM32F407](https://www.ampheo.com/search/STM32F407) (Cortex-M4, FPU).

**STM32 Example**

**Summary**