How To Optimise Your Reverb Treatments

There’s more to great‑sounding reverb than loading and tweaking your favourite plug‑in.

No matter how good the reverb plug‑ins that you own are, there are times when the best results can only be obtained using additional processing, either before or after the reverb. In this article, I’ll look at a number of techniques. Some can prevent the reverb clashing with other sounds and help it to sit better in the mix, while other options can add an impression of power, or inject interest and movement. For most of what follows, I’ll assume that the reverb plug‑in is hosted in a bus fed by aux sends from the various source tracks in your DAW session.

Filtering & EQ

Since reverbs add sound to a mix, they have the potential to obscure other sounds, and often cause a build up of ‘mud’ in the low and low‑mid frequencies. To counter this, try rolling some low end off the reverb signal with a high‑pass (low‑cut) filter. Assuming you’re using your reverb as a send effect, try putting a 200Hz 12 or 18 dB/octave filter before or after the reverb. This should work pretty well for most sounds, including vocals. Similarly, rolling off the top end can help if the reverb sounds too sharp or ‘glassy’. In a reverb, there’s rarely much of use above 10kHz, so set a 12 or 18 dB/octave low‑pass (high‑cut) filter there and roll it down to taste. Engineers have combined low‑ and high‑pass filters on the reverb for decades — using both at once in this way is sometimes known as the ‘Abbey Road reverb’ trick.

More detailed sculpting is possible with a parametric EQ, but unless you particularly want to change the tonality of the reverb compared with the source there’s usually no need. If you do go down this road, broad‑brush moves (wide‑Q or shelving boosts and cuts) are usually better, though if the reverb unhelpfully accentuates any resonances, esses or fricative sounds you can try countering that with narrow cuts. That said, there are often better tactics than EQ for that, as I’ll explain later.

Some reverb plug‑ins include high‑ and low‑pass filters, so you won’t always need to use dedicated EQ plug‑ins. Also note that other common reverb controls allow you to damp the highs or change the relative decay times of high and low frequencies; using these can sometimes sound better than filtering everything.

High‑ and low‑pass filtering a reverb can help it sit better in the mix. Neatly, Cockos Reaper can EQ the send differently for every source: each plug‑in’s connector pin matrix allows you to route its output to separate channels on the same track, while letting the main signal through unaltered.

A separate EQ does come in handy, though, where several tracks share the same reverb but you see a benefit in EQ’ing the reverb differently for each source. To do that, create a separate bus for each source you’re sending to the reverb and put an EQ on each bus. Then route these busses to your reverb track. This way, you can have a master reverb EQ on the reverb track, and an independent one for each source on its own bus. Reaper users can also EQ a send without this intermediate bus stage, as described in Martin Walker’s SOS March 2020 workshop: https://sosm.ag/reaper-0320.

Send & Blend

The signal routing for ‘send and blend’ reverb, whereby you use a channel’s send level to blend between the dry signal on the source channel and a fully wet signal on the reverb’s aux bus.

Normally, for a reverb set up as a send effect, you’d feed it with post‑fade sends from the dry sources. That way, if you change the source levels the reverb level will change accordingly. But if you want a sound to fade and leave only its reverb audible, to create the illusion of a sound moving into the distance, you can use a pre‑fade send instead: bring down a source channel’s fader and you’ll kill the dry sound but still hear that source in the reverb.

Taking this idea a step further, how about turning your send level controls into wet/dry blend controls, and making the source level fall increasingly as you send more of it to a reverb? Setting this up the first time can be a bit fiddly, but when you’ve done it once you can rinse and repeat fairly easily. For full details on how to implement this effect in your DAW, check out Matt’s article in SOS January 2017: https://sosm.ag/send-and-blend.

De‑essed Reverb

Some musical styles demand a bright, ‘steamy’ reverb sound. Such effects can be problematic where vocals are already on the verge of becoming sibilant; the bright reverb exaggerates the problem. If the vocal sounds OK, though, de‑essing it risks introducing lispy artefacts.

The reverb channel can usually take a lot more assertive de‑essing than the dry source before the artefacts become distracting.

In this situation, you can try using a de‑esser on the reverb bus instead, with the de‑esser placed before the reverb. You can afford to be quite heavy‑handed with the de‑essing where required, as the blend of dry vocal and processed reverb will still sound natural. If the sibilance occurs in the same frequency range throughout the song, you may also be able to put a notch filter after the de‑esser to further dip the offending sounds. Alternatively, use a dynamic EQ, or try automating the source’s send level to the reverb channel.

Pesky Transients

When treating percussive sounds, you may find that the attack of the reverb is too fast, and that this creates too obvious an effect. Some reverbs include an attack time control but many don’t. If not, try placing a Transient Designer‑type plug‑in before the reverb. This feeds slower note attacks to the reverb without affecting the dry source, making the effect bloom after the original hit rather than sounding like a hard repeat.

The attack of percussive sounds can be unhelpfully smeared by reverb. To address that, try using a Transient Designer‑style plug‑in before the reverb.

You don’t have to save this for percussive sounds; it can be a cool effect on various sources. And if you like to experiment, also try a transient designer after the reverb: take the attack right down and bring the sustain up to taste.

Ducked Reverb

Sometimes, a reverb seems to muddy the mix yet still doesn’t feel loud enough during pauses in the action. Before changing the reverb sound or getting into detailed automation, try using a ducker on the reverb signal. Put a compressor (or gate with negative attenuation capability) after the reverb, and set the compressor so that it reacts to its external side‑chain, rather than to the reverb channel’s signal. Send the dry source (eg. lead vocal) to the compressor’s side‑chain input.

Ducking the reverb by sending the source both to the reverb and to the side‑chain input of a compressor on the reverb track.

Now, the reverb sound should drop when the vocal is playing back, and swell up during the gaps in between. Use the compressor’s attack and release controls to adjust the speed at which the reverb swells up between phrases, and how quickly it’s brought under control again when the vocal gets louder again. Set these timings by ear.

Smooth Delays

Feeding repeating, filtered delays into a reverb (or vice versa) can sound oh‑so smooth.

Putting a delay without repeats before the reverb is no different from using the pre‑delay parameter of most reverbs. But setting a delay (before or after the reverb) to produce a repeating echo, and using its onboard filters to roll off the lows (around 300Hz) and the highs (above 1.2kHz or so) achieves a very different effect. It makes the delay sound less obtrusive, and helps the repeats recede into the distance as they decay. Whether using the plug‑ins as inserts on a track or on a bus fed from an aux send, the wet/dry mix controls on each effect need adjusting so that you hear the desired contribution from each effect. When used as sends, one of the plug‑ins, usually the reverb, should be set fully wet.

Squish & Crunch

Placing a compressor or saturator before the reverb can create a really dense‑sounding effect.

Sometimes, you want a really dense reverb sound, with the tail remaining high in level during the initial part of the reverb decay. One way to achieve this is to put a compressor before the reverb. A saturator or even an amp modeller can work too, while also warming/darkening the sound somewhat. Go too far, and you may hear obvious distortion in the reverb, but because a reverb tends to diffuse what passes through it, you can add more dirt than you might think before things get rough.

Panning For Gold

Using a tremolo to alternately autopan a piano sound and its reverb.Most of the time, people tend to use the stereo output from a reverb, with each channel panned hard left and right. That maximises the illusion of spaciousness and emulates the way reverb occurs in nature. But working this way can also dilute the sense of location for the original sound. A different approach, used by a number of classic producers and engineers including George Massenburg, is to use multiple mono reverbs and pan each one to the same left/right position as the source sound to which they relate. This preserves the position of the sound source and still sounds suitably wide once a few other tracks have been treated in the same way.

Another approach is to use a mono or narrow‑stereo reverb, such that the source is panned to one side and its reverb to the other. This can work well for creating a sense of distance and width. If you really want to mess with your listeners, though, put an autopanner on the dry sound and another, with identical settings but 180 degrees out of phase with the first, on the reverb; the source will always move in the opposite direction to the reverb!

Pitch Processing

If placed on an aux channel before a reverb, a real‑time pitch‑corrrector can be used fairly assertively to create a lovely thickening effect.

You can make a vocal reverb really pop out of the mix by putting an automatic pitch‑correction plug‑in directly before the reverb. If you set the pitch‑corrector to act fairly assertively, the reverb tail will essentially be pitch‑quantised while the dry vocal remains natural. The slight pitch variations between the dry vocal and reverb tail make for a very rich sound, but without anything sounding too obviously processed.

I covered shimmer reverbs recently (https://sosm.ag/shimmer-reverb-effects) so I won’t go into detail here. But, essentially, a pitch‑shifter before the reverb adds some ‘octave‑up’ sound. Happily, while many pitch‑shifters can sound grainy and unnatural on dry sources, a fully wet reverb will knock most of those rough edges off.

You can make a vocal reverb really pop out of the mix by putting an automatic pitch‑correction plug‑in directly before the reverb.

You’re probably familiar with the fake double‑tracking effect that uses a stereo pitch‑shifter to add a couple of extra parts to a vocal or instrument, usually one part between three and six cents sharp on one side, and the other opposite and flat by the same amount. Well, instead of applying this effect to the main vocal, try putting the pitch‑shifter before a vocal reverb. Again, this will make the reverb sound more expansive.

Rotary Club

A rotary‑speaker effect used after a reverb can add an interesting sense of movement — it’s much more subtle than using it on the source you’re sending to the reverb.

When using reverb to create ambient guitar sounds, combining compression, delay and reverb is often a good way to go. Sometimes, though, putting a slow rotary speaker emulation after the reverb will add a nice, slow, cyclic wash to the sound. Using an emulation of a Roland Dimension‑D after a reverb can also add a useful texture without becoming too obvious; it can work extremely well on minimalist acoustic piano parts.

Chop Shop

Reverb can be turned into a rhythmic element, simply by passing the reverb output through a tempo sync’ed chopper or a stepped filter plug‑in. When doing this, don’t be afraid to use really radical EQ to squeeze the sound into the audio band where it works best.

Feeding a long reverb into a tempo‑sync’ed chopper (ie. a square‑wave tremolo, as used here) can turn the reverb into an interesting rhythmic element.

Gated Reverb

One of the best‑known reverb effects is gated reverb. Originally this was created either using natural room reverb or a conventional synthetic reverb, then using a noise gate triggered via its side‑chain input to interrupt the reverb. Folklore attributes the effect to Hugh Padgham, who first used it in 1979 and later, more famously, on the classic Phil Collins single ‘In the Air Tonight’. The story is that it was discovered by accident when a drum overhead mic was fed into a heavily compressed and gated talkback mic channel by mistake. Earlier still, Tony Visconti had used gates on room mics for David Bowie’s voice: he set up several mics at different distances from Bowie, and each was processed by a gate with a different threshold. More gates opened up as the voice got louder, so the louder the voice the more ambience was added to the sound.

Using reverbs and gates to emulate the Tony Visconti ‘Hansa reverb’ technique famously used on David Bowie’s ‘Heroes’.Assuming you don’t have a huge stone drum room in your house, you can set up a gated snare effect by sending a drum overhead mic (or just the snare mic) through a reverb and then through a gate. The close snare mic signal also feeds the gate’s side‑chain, and you adjust the gate’s threshold so it opens on each snare hit. By setting a fast attack and fast release, with a hold time of maybe a quarter of a second, you should hear a short burst of reverb following each hit. If the signal is heavily compressed, by putting a low‑threshold, high‑ratio compressor before the reverb, the gated sound will sound big, dense and punchy. Many reverbs offer a synthesized gated reverb, but they don’t usually sound like the real thing in my experience; they’re actually a fairly coarse burst of early reflections. Using something like a plate or large room reverb followed by a gate that’s triggered via its side‑chain will get you much closer to the sound of the real thing.

You can also mimic the Visconti/Bowie effect in your DAW by setting up a few reverbs, each simulating a position further back in the room than the last, and sending the vocal to them all. Insert a gate before each reverb and set the gates’ thresholds so that the more ‘distant’ the reverb, the more signal is needed to push the gate open.