Why louder sounds better
Play two pedals back to back and the louder one wins. Almost always, almost regardless of what the two pedals actually do. This bias is the oldest confound in gear evaluation — and being fooled by it one time too many is, honestly, the reason this app exists.
The bias
Human hearing is not flat, and it is least flat at low levels: as sound gets louder, bass and treble come up disproportionately, detail seems to emerge, and everything reads as richer, fuller, more exciting. A change of one decibel — far too small to register as “louder” in a quick A/B — still shifts the impression in favor of the louder source. Showrooms and shootout videos have traded on this forever, usually innocently: a pedal with a bit of output boost sounds “better” than a unity-gain rival even if their distortion is identical.
The cruel part is that knowing about the bias doesn’t turn it off. You cannot listen around it; you can only remove it.
What the app does about it
Every comparison PedalScope plays is loudness-matched first: the Hear-it panels’ variants, the Compare view’s audition lanes, and each simulation A/B are all adjusted to the same measured loudness before playback — not the same peak level, the same perceived loudness, using the broadcast-industry measure built for exactly this problem. When one lane still sounds bigger after matching, that difference is real character, not level.
The same rule shapes the words: a comparison summary’s “After loudness matching” line lists only the differences that should survive equal loudness, because everything else on the list might be the bias talking.
What this means for your own tests
Match levels before judging. Rolling a pedal’s volume knob until an A/B sits at equal loudness is the single highest-value habit in tone evaluation.
Distrust “it just sounds bigger”. Check the Compression chart: if the pedal’s output simply sits a few dB above unity, “bigger” may be free — and available from any clean boost.
Beware demos. Any comparison you didn’t level-match yourself tells you as much about gain staging as about tone.
For the curious: the matching standard
Loudness is measured per ITU-R BS.1770 / EBU R128: the signal is K-weighted (a filter approximating the ear’s sensitivity), measured in 400 ms gated blocks, and integrated with the standard absolute and relative gates into LUFS. Every comparison render is iterated to the same LUFS target (matching is repeated because the gating makes a single gain step land slightly off for wide-dynamic material). Near-silent lanes are deliberately left quiet rather than amplified to target — silence is an honest answer.