June 9, 2026
FFT? More like FIGHT fight
Show HN: Resonate – Low-latency, high-resolution spectral analysis
A new audio breakthrough lands on HN—and the comments instantly yell “been there”
TLDR: Resonate promises a quicker, lighter way to analyze sound in real time, which could matter for music and audio apps. But the comments stole the show: one developer claimed they’d already shipped this years ago, while others demanded messy real-world tests before calling it a win.
A shiny new Show HN post about Resonate tried to sell a simple dream: a faster, lighter way to “see” what’s happening in sound in real time, without the usual waiting around that comes with older methods. In plain English, it’s a tool for spotting pitches and sound details quickly, potentially useful for things like tuners, chord detectors, and other music apps. But on Hacker News, the real performance started in the comments, where the crowd immediately turned this into a classic internet showdown: breakthrough or reheated leftovers?
The loudest reaction came from bialamusic, who barged in with serious “sir, this is not new” energy, saying they’d already built apps using the same idea years ago and dropping store links like receipts. That instantly set the mood: less polite applause, more tech paternity test. Then the thread split into factions. Skeptics wanted proof, not promises, with one commenter begging for side-by-side demos on ugly real-world audio—fast pitch changes, messy chords, noisy recordings—the kind of stuff that exposes whether a tool is actually useful or just pretty on paper. Meanwhile, others defended the post, saying there is something interesting here: a clever twist on familiar sound-analysis ideas that could cut delay and make real-time tracking smoother.
And yes, there was comedy. One user dragged piano tuners into the chat, joking about ultra-devoted experts who think they can out-tune any digital gadget. Suddenly this wasn’t just about audio math—it was man vs machine, app store receipts vs research papers, and vibes vs benchmarks.
Key Points
- •Resonate is described as a spectral analysis algorithm designed for low latency, low memory use, and low computational cost on audio and other signals.
- •The method updates each resonator state at every input sample using recursive complex-valued equations, without buffering.
- •The algorithm uses EWMA both to accumulate signal energy around each resonant frequency and to smooth the resonator state.
- •For audio applications, the article gives a heuristic formula for the update parameter alpha_k over the 20 Hz to 20,000 Hz frequency range.
- •Banks of independent resonators can estimate perceptually relevant spectral content in real time with memory and per-sample cost linear in the number of resonators, and the 2026 extension allows resonant frequencies to vary over time.