the guitars that vanished on a phone speaker
a CHECK user asked a question i could not answer with a meter. why a mix can look fine on your meters and still lose something in mono, and how to tell the normal kind of loss from the kind worth fixing.
this one’s a day late, sorry. i’m deep in the big update for all five plugins, and tuesday slipped past me.
a CHECK user wrote to me last week. he records acoustic guitars, pans them hard left and right, and when he checks the mix in mono (like a phone speaker or a club system playing one channel) some of the guitar just goes missing. CHECK showed him the risky areas. but what he really wanted was to hear only the part that disappears, and nothing in CHECK does that yet.
i thought it would be a small feature. it turned into a week of reading, because the thing he hears missing is actually three different things wearing the same coat.
why the meter looks calm
most stereo meters give you one number between +1 and -1. +1 means left and right are the same sound. 0 means they have nothing in common. -1 means they’re mirror images, and in mono they cancel each other out completely.
but that number is an average. a part that reads 0.5 can spend half its time at +1 and half at -1, and the meter shows you something calm the whole way through. the same happens across the frequency range: the whole mix reads fine while one narrow band is quietly cancelling.
three kinds of disappearing
the guitar panned hard left gets quieter. in mono, left and right are added together and halved, so anything that only lives on one side comes out about 6 dB lower. that’s a level change, and you fix it with a fader, if at all.
wide, airy sounds get a bit quieter too. pads and rooms where left and right are different from each other drop around 3 dB in mono. also normal.
the third kind is the one to hunt. left and right are almost the same sound, slightly out of step, so they push against each other and subtract. a doubled part with a tiny offset, a widener pushed too far, a reverb that folds badly. that’s where a vocal loses its edge or a bass note goes hollow on a phone.
i had the first two filed under “problem” at first, which is how you end up fixing the wrong thing.
what to take into your next session
look at your stereo meter band by band. one number for the whole mix hides the band that’s actually in trouble.
if a hard-panned part drops in mono, decide on its level and move on. turn it up a touch or leave it. there’s no phase problem to chase.
check mono every time you add something wide. the user who wrote to me builds his mix dry first (levels and panning only), checks it, then checks again after every reverb or widener. when something breaks, he knows which move did it.
i’m working on a way to let you hear only the part that cancels. no date yet. if you’ve got a mix where mono hurts and you can’t work out why, reply and tell me about it.
more field notes
-
sep 29, 2026 · sound science
OPEN a week in, and what comes next
OPEN is a week old, the other four plugins are headed for one big update through the same blind test, and this week's film is about a music study nobody has run twice. a result counts once someone runs it again.
-
sep 15, 2026 · sound science
your loudest master never arrives loud
streaming turns your master down to a target before anyone hears it, so mastering louder buys nothing but lost dynamics. and the loudness war mostly did not do what people think: measured across five decades the spread between loud and quiet did not narrow. what came down was the transient crack, by about a decibel and a half. so the thing to watch is heavy limiting and clipping, more than compression as an idea, and the move is to master until it is finished and let the platform set the number.
-
sep 8, 2026 · sound science
the dialogue you mixed is not the dialogue they hear
a chain of reasonable decisions ends at the sofa with the voice buried. the film is about the chain. this issue is about your end of it: how to mix dialogue so it survives a stereo fold-down, a flat panel and an ordinary pair of ears, with the numbers i use, including one my own films failed on.