{"id":7061,"date":"2026-07-17T12:38:26","date_gmt":"2026-07-17T10:38:26","guid":{"rendered":"https:\/\/www.pschatzmann.ch\/home\/?p=7061"},"modified":"2026-07-17T13:57:18","modified_gmt":"2026-07-17T11:57:18","slug":"microcontroller-fft-ifft-performance-benchmark-n64","status":"publish","type":"post","link":"https:\/\/www.pschatzmann.ch\/home\/2026\/07\/17\/microcontroller-fft-ifft-performance-benchmark-n64\/","title":{"rendered":"Microcontroller FFT & IFFT Performance Benchmark (N=64)"},"content":{"rendered":"

The Fast Fourier Transform (FFT)<\/strong> is an efficient algorithm for computing the Discrete Fourier Transform (DFT)<\/strong>: it converts a signal from the time domain<\/strong> into the frequency domain<\/strong>, revealing which frequencies are present in a signal and how strong they are. It is the workhorse behind spectrum analyzers, audio effects, filtering, and countless other DSP applications.<\/p>\n

I recently published a new FFT library<\/a> that supports both integer and floating-point<\/strong> data types for its calculations. Using this library, I benchmarked DSP performance across a wide range of microcontroller architectures \u2014 from a vintage 8-bit AVR all the way up to a 480 MHz ARM Cortex-M7. The results reveal some surprising winners and a few important lessons about choosing the right chip (and the right data type) for your signal-processing project.<\/p>\n

\n

Methodology<\/h2>\n

Each benchmark measures the time to compute a single N=64 complex FFT and IFFT<\/strong> for five data types: int8_t<\/code>, int16_t<\/code>, int32_t<\/code>, float<\/code> (32-bit), and double<\/code> (64-bit). All timings are averaged over repeated runs and reported in microseconds (\u00b5s); throughput is derived as 1,000,000 \/ latency.<\/p>\n

Executive Summary & Key Takeaways<\/h2>\n