reverb is a trade: space for position

a producer sends me a vocal that “sounds small” after he added reverb to make it bigger. that sentence is the whole issue in one breath. he reached for space, and space cost him presence. nothing he did was wrong. the trade is built into the physics, and once you can see it you stop fighting it and start spending it on purpose.

the frame

reverb gives a sound a place to live, and the price of that place is foreground. distance and space are the same control. you don’t get one without paying a little of the other.

that’s how your ear reads a room, whatever anyone’s mixing opinion is.

why distance and reverb are the same dial

your brain judges how far away a source is partly from the ratio between the direct sound and the reverberant energy around it. it’s called the direct-to-reverberant ratio, and Pavel Zahorik measured its role directly in 2002: as that ratio drops, the source is heard as farther away. Bronkhorst and Houtgast, writing in Nature in 1999, named the same ratio as one of the two primary cues we have for distance, alongside loudness.

the reason it works is simple geometry. move a source back and the direct sound falls off fast, roughly six decibels every time the distance doubles. the reverberant field, bouncing off the whole room, stays about the same. so the ratio shifts, and your ear reads the shift as distance. when you add reverb to a track, you are doing the exact same thing to the ratio that physically moving the source away would do. the plugin can’t tell your ear “more space but stay close.” the cue is one cue.

worth knowing where the edges are: it’s a coarse cue, not a ruler. Zahorik’s follow-up the same year put the just-noticeable change at around five to six decibels of ratio, which is more than a doubling of apparent distance. so reverb-as-distance is a feel you paint with, not a number you dial.

the cost side

the spend is presence. late reverberant energy arrives after the direct sound and smears across it, and it masks the very detail that made the source sit forward. Leclère and colleagues showed in 2015 that this temporal smearing measurably reduces intelligibility, the reverberant tail acting like noise laid over the target.

but here is the distinction that saves the technique: early and late aren’t the same thing. the first reflections, arriving within a few tens of milliseconds, reinforce the direct sound and add body. it’s the long late tail that does the damage. so the goal is never “less reverb.” it is “more early, less late,” and a way to keep the front edge of the sound clear of the wash.

the escape hatch: pre-delay

that way exists, and it’s older than any plugin. the precedence effect, mapped by Wallach in 1949 and refined by Litovsky’s group fifty years later, says the first-arriving wavefront wins: it owns the perceived position, and reflections that land within a short window fuse into it instead of registering as separate, smeared events.

pre-delay is how you use that. hold the reverb back ten, twenty, forty milliseconds, and the dry transient lands first and claims the front of the image. only then does the tail bloom in behind it. the word stays crisp, the snare keeps its crack, and the space opens up underneath. as a bonus, longer pre-delay reads as a bigger room, so you’re buying size and protecting presence with the same control.

the upside you don’t want to lose

it would be easy to read all this as “reverb is a tax.” it isn’t. the same energy that costs you presence is what pulls a sound out of your head and into a place. Brimijoin and colleagues showed in 2013 how strong that effect is: sources that behaved like real things in a real room were heard as outside the head most of the time, while ones stripped of those cues collapsed to inside-the-head listening far more often. reverb is part of how a sound becomes a thing in a space rather than a signal in your skull. the trade buys embodiment. that is worth paying for, as long as you know you’re paying.

the same shape, outside the studio

i can’t unsee this one either. you add context to make a thing land, a longer intro, more backstory, more setup, and past a point the extra space pushes the actual point into the background. presence and place trade off in a sentence the same way they do in a mix. the move that makes something feel situated is the same move that can make it feel distant. the skill is spending just enough room to place it and not one decibel more.

the tools

in the mix it comes down to three dials used on purpose. pre-delay to keep the transient in front of the tail. an early-to-late balance to choose how much presence you spend on how much distance. and decorrelation for width that still survives a mono sum, because a stereo tail that cancels when summed is the one way reverb can cost you something with no upside at all. i keep CHECK, the free analyser, open to watch for that: how much of the space i just added is still there when the signal folds to mono.

this early-and-late, pre-delay-first way of thinking is the whole philosophy behind OPEN, the algorithmic reverb i have been building. it’s in beta now, and it’s built around spending the trade deliberately rather than hiding it.

from the studio

reply with the worst case of this you have caused: the vocal you reverbed into the next room, the snare you sent to a stadium and never got back. i collect them. and if you want first crack at OPEN when the beta opens, just say “OPEN” and i’ll put you on the list.

jonas