Dynamics

Sidechain Ducking Calculator

Sidechain pumping is the most abused effect in club production and the most misunderstood. A plugin says twelve decibels of reduction, but what the audience hears is not twelve decibels of reduction, it is the average amount of level that was removed across the whole bar, and that depends on how fast the compressor attacks and how long it takes to recover. Set the release too short and the duck collapses into a gate chop that stutters the arrangement. Set it too long and the level never comes back, so the effect turns into a slow level ride with no pulse at all. This calculator takes the source level, the target depth, the attack and release times, the tempo and the time signature, and turns them into the numbers that actually matter: the envelope drawn across a bar, the percentage of the bar under the duck, the percentage still audible, and whether the result will read as subtle, pumping or extreme.

Source signal level

dBFS

Level of the material that gets ducked

Target attenuation

How deep the duck goes at the bottom of the dip

Attack

Time from the kick to full depth

Release

Time taken to come back up to full level

Track BPM

BPM

The kick is assumed to land on every beat

Beats per bar

Result

How to use it

STEP 1

How to use this calculator: enter the level of the material you are about to duck, not the level of the sidechain key. The duck depth only makes sense relative to what is underneath it, and a twelve decibel duck on a source sitting at minus eighteen dBFS sounds different from the same duck on a source at minus six. Then set the target attenuation to the number you actually want to hear rather than the number the plugin is dialled to.

STEP 2

Next enter the attack and release times, and read the average attenuation row. This is the number that predicts how the effect will feel. A twelve decibel duck held for a fifth of a bar removes roughly one decibel from the track on average, which is the classic pumping sound. The same twelve decibels held for most of the bar removes far more and will bulldoze the arrangement.

STEP 3

Set the BPM and the time signature next. The kick is assumed to land on every beat, so a bar of four at 128 BPM gives you four dips and a bar of 1875 milliseconds to fit them into. The envelope drawing then shows exactly where each dip sits, how long it lasts and how much of the bar is left at full level, which is usually more informative than staring at the plugin.

STEP 4

Finally check the release window row. The shortest release listed keeps the dip clearly deeper than half the target for long enough to be heard as a pump, and the longest one is the point where the level stops recovering between kicks. Anything outside that window either chops the arrangement or turns the effect into a slow level ride instead of a pulse.

Three common mistakes

Common mistake: reading the depth setting on the plugin as the depth of the effect. A compressor labelled twelve decibels of reduction is not removing twelve decibels from the track; it is removing twelve decibels for the length of the dip and then giving the level back. The average attenuation row is the honest figure, and it is usually a fraction of what the plugin claims. Judge the pump by that number and by the envelope drawing, not by the value on the knob.
Watch out for a release that is shorter than the gap between the kick and the next element you care about. With a short release the duck chops the arrangement into stutters, and the bass or a vocal line that used to sit between the kicks gets cut in half. This is the most common reason a sidechain feels like it is fighting the music instead of supporting it.
Key point that gets missed: a very long release does not give a bigger pump, it gives a level ride. Once the release runs past the gap between hits, the signal never returns to full level and the compressor just holds the whole track down. The result has no pulse at all, which is why the long release threshold in the release window row matters as much as the short one.
Easy to get wrong: applying the same settings to a break with no kick. The calculation assumes a trigger on every beat. In a breakdown, a half time pattern or a track with a kick that only lands on the one and the three, the real number of dips is smaller, so the ducked share of the bar is smaller and the pump is lighter than the calculator suggests.

What the percentages actually mean

The two percentages in this calculator answer different questions, and mixing them up is why sidechain settings get argued about. The ducked share is the fraction of the bar spent under attenuation, measured from the attack time through the end of the release. The audible share is everything else, where the material sits at full level. Neither of them is a measure of how hard the effect pushes.

A deep duck over a short window can be almost invisible in the average and still clearly audible on every hit.
A shallow duck over a long window can remove more average level than the deep short one and feel far more intrusive.
The average attenuation row integrates the envelope across the bar, which is the closest thing to a single honest number for the effect.
Anything above about one and a half decibels of average attenuation starts to make decisions for the arrangement, so it needs a reason.

Attack and release, and the shape of the dip

The envelope drawn here is a linear attack into the target depth followed by a linear release back to zero. Real compressors do not behave that way. Most release on an exponential curve, so the recovery is fast at first and then slows down, and the drawn dip is a little deeper through the middle of the recovery than a real plugin will be. A soft knee also means the approach into the attenuation is rounded rather than a corner.

Very short attacks, under five milliseconds, let the compressor grab the initial transient of the source as well as the kick, which adds a clicky edge to the dip.
Attack times around ten to thirty milliseconds are the usual range for a smooth pump on dance material.
Release times between one fifth and one half of the beat are the safe zone, which is where the release window row points.
If the source material is bass heavy, a longer attack keeps the low end of the kick from being chopped and gives a heavier, slower feel.

Pump feel, and when to pick a different effect

The pump rating is not a quality score. Subtle pumping is what most house and techno records do, and it works because the ducked element is a supporting layer while the kick and the bass carry the weight. Extreme pumping works in a specific register only: it needs a kick that stays clear, a bass that is not fighting for the same space, and a mix where the ducked element is disposable enough that the listener reads the movement as energy.

If the ducked element is the lead vocal, extreme pumping will destroy the lyric and the mix will sound amateurish.
If the ducked element is a pad or a reverb return, you are usually better served by a slower attack and a shallower depth, or by an LFO on the filter instead.
Try a different key source for a different feel. Ducking from a snare gives a half bar pattern that reads as a different groove entirely.
Check the result in mono and on a small system, because a pump that is exciting in stereo can disappear entirely on a phone speaker.

FAQ

Why does my duck sound deeper than the calculator suggests?

Because the drawing shows the level of the ducked material in isolation, while the ear hears the change in the balance between elements. A bass that drops twelve decibels on every kick does not just get quieter, it stops competing, and that shift in the mix is much more obvious than the raw number. The calculator reports the level change honestly; the perceived effect is larger because of how the elements sit against each other. If the depth feels excessive in the mix, the answer is usually to duck something else rather than to reduce the number blindly.

Should the attack be faster than the kick transient?

Not usually. The point of the attack is to let the compressor respond to the kick without clipping the very first transient, and on most material a fast attack grabs that transient and produces a click on the front of the dip. Between ten and thirty milliseconds is the range where the duck feels smooth but still reads as being triggered by the kick. A faster attack is worth trying on sparse, sharp material where the kick transient is short and you want the dip to land right on it.

How do I stop the bass from pumping with everything else?

Move the bass out of the sidechain key. Most arrangements have the kick and the bass playing the same rhythm, so ducking the mix ducks the bass too, and the low end disappears on every hit. The usual fixes are to high pass or split the bass so the sub below the duck region is not touched, to use a slower and shallower duck on the bass bus only, or to use a separate key such as a snare for the other elements so the bass keeps moving. Higher pumping mixes often keep the sub completely out of the key.

What happens if the track is not at a steady tempo?

The calculation is anchored to a fixed BPM, so a live recorded or badly timed kick will not sit on the grid the drawing assumes. The percentages still describe the average case, but the gaps between dips will vary, and a fixed release time behaves very differently when the gaps change. When the timing is loose, set the release from the shortest gap you actually have rather than from the average, and check the envelope in the arrangement rather than trusting the bar drawing.

Is a deeper duck always more energetic?

No. Depth is only one part of it, and the average attenuation row is the better guide. A shallow duck over a long window removes more level from the track than a deep duck over a short one, and it also removes the dynamic contrast that makes the kick feel like an impact. The records that read as most energetic are usually the ones where the ducked material is a layer rather than an element the listener is following, so depth can stay moderate and the arrangement keeps its clarity.

Limits of this tool

Limits: the envelope is a linear attack and a linear release, not the exponential curve most compressors use, so the drawn dip is slightly deeper through the recovery than reality.
Not for: multi source keys, external sidechains or tempo synced plugins. The model assumes one kick on every beat with a constant tempo.
When this stops: with a release longer than the gap between kicks the model no longer describes the behaviour, because the level never returns and the envelope merges into a single curve.
Scope: the sidechain key is the kick on every beat. Half time patterns, ghost kicks and breakbeat arrangements change the real trigger count and the percentages with it.