What a measurement can't tell you
This app exists to replace folklore with measurement — which makes it doubly important to say where measurement stops. Every chart in PedalScope is honest; none of them is complete. Here is what the numbers do not, and sometimes cannot, capture.
The measurements are snapshots
A Fingerprint is one set of knob positions, one moment — and most of its sub-measurements probe one drive level at a time. The app widens the frame where it can — the Gain Map makes level an axis, family runs make a knob an axis — but a pedal is a surface with more dimensions than any finite set of sweeps. The charts interpolate; your playing extrapolates.
Time-variance mostly escapes the net
The measurement model assumes the pedal behaves the same during second five of a sweep as during second one. Many interesting circuits don’t, quite:
Bias drift and sag — germanium fuzzes shifting operating point with temperature and battery voltage, sometimes over minutes.
Envelope-dependent behavior — circuits whose character depends on how long you’ve been sustaining, not just how hard you hit. The transfer curve’s hysteresis loop area catches the fast end of this; slow drift between measurements shows up, if at all, as row-to-row wobble in a gain map.
Temperature — a pedal measured cold is not the pedal an hour into a set. The metadata field is there to note it; the chart can’t see it.
When a summary hedges (“appears to”, “at this measurement’s level”), this is the family of reasons why.
The rig is part of the instrument
PedalScope measures the pedal against the interface’s line input — a clean, high-headroom, mostly-resistive load. Your rig is messier in ways that matter:
Your pickups see the pedal’s input impedance, and some circuits (fuzz faces, notoriously) change character entirely with what drives them. The series-resistor trick in rig setup approximates a guitar’s source impedance; it is an approximation.
Your amp and speaker are their own nonlinear, band-limiting system. A pedal’s measured 8 kHz fizz may never survive a 12-inch speaker.
The room, the volume you play at, and the loudness biases of your own ears complete the chain. None of them are in the loop here.
The band edge is faded, not measured
Reading harmonic k at note f₀ means reading output frequency k·f₀ — and for high harmonics of high notes that runs off the edge of what the measurement chain can honestly report. Where it does, the app fades the trace instead of pretending: faded, dashed regions on the Harmonic Distortion and Compare charts (and faded columns on a single-harmonic Gain Map map) are analysis artifacts past the reliable edge, shown so you can see where the limit is, excluded from every fit, summary, and quoted number. The bound is computed from each record’s own sweep and calibration parameters, not assumed.
Separately from that limit, every quoted THD/IMD percentage integrates the audible band, 20 Hz–20 kHz — a pedal’s real ultrasonic content is measured (and feeds the Hear-it models), but content no rig reproduces doesn’t move the numbers. That keeps the same pedal quoting the same figures whatever the capture rate.
THD is a number, not a verdict
Total harmonic distortion says how much was added — nothing about how it reads. Two pedals at an identical 10 % THD can sound worlds apart: one concentrated in a warm H2, the other spread across a fizzy high series. That’s why the app leads with the full Harmonic Distortion measurement and treats THD as one summary row of it. Distrust any comparison anywhere that reduces distortion character to a single percentage.
Black box means black box
The app measures behavior at the jacks. It cannot see components, topology, or clipping-element chemistry — and its summaries are explicitly forbidden from claiming them. If a description here ever seems to know what’s on the circuit board, that’s a bug in our honesty, and we’d like to hear about it.
“Just use your ears”?
Always. Ears are the final arbiter of tone, and nothing in this app says otherwise — a measurement is never a verdict about what you should like. The reason to measure is different: to reconcile what you hear with what actually happened.
Because the folklore vocabulary, it turns out, keeps landing on measurable quantities. A “tight” drive pedal usually turns out to be starving its clipper of bass — the lows are filtered down before the distortion happens, so the low notes stay cleaner than the mids (watch that gap in the Harmonic Distortion chart). “Spitty” or “sour” chords are out-of-key intermodulation products — sum and difference tones landing between pitches, which the Chord Roughness chart draws stem by stem. When a word you’d use lines up with a number the app measured, you’ve learned what to change to get more or less of it.
And in one respect ears are reliably, measurably wrong: loudness. Nobody judges character fairly across even a one-decibel level difference — the louder side wins — which is why every comparison here is loudness-matched before you hear it. That matching exists precisely because ears can’t be trusted about level; it’s the one part of listening the app quietly does for you.
The honesty cuts the other way too. The instrument resolves differences far smaller than anyone can hear — a plain cable measures around 0.01 % THD and the charts resolve it — so a measured difference is not automatically an audible one (and a single number never ranks sound; see THD is a number, not a verdict above). That’s why the app reports differences as quantities (“H2 differs by 8 dB at this note”), never as judgments. Use your ears to decide what’s better; use the measurement to find out what “better” was.
And feel is still yours
Latency of your own muscles against the attack, the way a note blooms at stage volume, whether the thing makes you play differently — the parts of “feel” that live in the loop between your hands and your ears aren’t in any chart. The measurements are a map of the pedal. You are still the territory’s judge.