Skip to content

Attack time changes how a part feels

Attack time is one of the main cues the ear uses to tell sounds apart. A fast onset speaks on the beat and sits forward; a slow one swells and sits back.

Updated · 6 min read Lesson 4 of 10 in Sound Design

In short

  1. Attack time is one of the main cues the ear uses to tell sounds apart, so changing it changes what kind of instrument a part seems to be.
  2. A fast attack lands on the beat and sits forward. A slow one becomes audible later and sits further back.
  3. Shape the envelope before you reach for EQ, and move slow-attack parts a little earlier to keep them in time.
In this lesson (8)

Bright chords that still drag

You programmed a chord progression you like, but it feels lazy. It sits at the back of the mix and does not push the song forward. So you open an , boost the highs for bite and the mids for punch. The chords get brighter and louder. They still drag.

The drag comes from the envelope, which EQ leaves alone. How a note starts is set in its first moments, by the attack. A slow attack keeps a part relaxed and behind the beat however bright you make it. A fast attack makes it speak at once.

Why it matters

The onset is part of the timbre

is not a small detail of a sound. When listeners rate how different instrument sounds are, rise time comes out as one of the main dimensions they use, alongside where the energy sits in the spectrum (McAdams and colleagues, 1995). Change the attack and you change what kind of instrument the part seems to be: plucked, struck, bowed or swelling.

It changes timing too. A fast attack reaches full level almost the moment the note starts, so the part lands where you placed it. A slow attack takes time to rise above the rest of the mix, so the same note seems to arrive late and the part leans behind the beat.

It also changes depth. Listeners judge distance from level, from the balance of direct sound against room reflections, and over long distances from the loss of high frequencies (Zahorik, Brungart and Bronkhorst, 2005). A sharp onset belongs to the direct sound, and the reflections of a room blur onsets. A soft attack can therefore resemble a sound heard from further away, and parts with slow attacks tend to sit behind parts with fast ones.

Figure 1. Two ADSR envelopes for the same note, with the same sustain and release. The fast attack is at full level almost as soon as the note starts. The slow one is still rising for nearly half the note, which is why it sounds later, softer and further back.

Listen

Same notes, different attack

One phrase with an adjustable attack and release. A fast attack speaks and pushes. A slow one swells and sits back.

Loading demo…
100%

Start with your speakers or headphones low.

The science

The amplitude envelope

A synth shapes each note with an envelope, usually : attack, decay, sustain and release. In many analog envelope generators, and in digital models of them, the attack is a capacitor charging, which gives an exponential rise toward full level:

A(t)=Amax⁡(1−e−t/τ)A(t) = A_{\max} \left( 1 - e^{-t/\tau} \right)

τ\tau is the time constant. After one time constant the level has covered 63 percent of the way, after three 95 percent and after five more than 99 percent. With a short τ\tau the note is at full level almost at once. With a long one it is still rising when a faster part would already be on its next note.

Figure 2. An analog-style attack, computed from the formula. The level covers 63 percent of the way in one time constant and 95 percent in three, then keeps creeping toward full level.

Synths label the attack knob in different ways, often as the time to reach full level rather than τ\tau, and some use straight-line ramps. The same number can feel different from one synth to another, so trust your ears and the meter more than the label.

A fast rise also puts more high-frequency energy into the onset. A sudden step contains energy across the whole spectrum, while a slow swell adds almost nothing beyond the note's own . That is why a fast attack sounds like a click or a pluck and a slow one sounds soft, even at the same peak level.

Try it

The attack duplication test

Hear one part change its job when only the attack changes. If you use an audio part instead of a synth, step 2 needs a , which some DAWs call an envelope shaper.

0 of 7 steps done

The fast version speaks on the beat and sits forward. The slow version swells, leans behind the beat until you move it earlier, and sinks further into the reverb.

Common mistake

Fixing an envelope problem with EQ

An EQ can make a slow attack brighter, but it cannot make it faster. If a part feels lazy, change its attack before you touch its tone.

The opposite mistake happens on drums: a at its fastest attack, often well under 1 ms. It clamps the of every hit, so the kick and snare lose their crack and sink back. A slower attack, tens of milliseconds as a starting point, lets the transient through before the gain reduction starts. The lesson on compression and motion shows that in detail.

Takeaway

Shape the envelope before the EQ

Decide how each part should speak before you decide how it should sound. Leads and rhythm parts that drive the song need fast attacks. Pads that should sit behind the vocal can swell in slowly, and moving them a little earlier keeps them in time. If a part lacks punch, shorten its attack or use a transient shaper. If a pad crowds the vocal, lengthen its attack and let it sit back.

3 questions

Check yourself

  1. Your synth string patch has the right brightness, but it still sounds like a pad instead of a plucked string. According to the timbre study by McAdams and colleagues, which control does most to change what instrument it seems to be?
  2. An exponential attack has a time constant of 20 ms. Roughly how long until it reaches 95 percent of full level?
  3. You lengthen the attack on a chord part and now it feels behind the beat. What fixes the timing and keeps the swell?

Sources

Show all 2
  • McAdams, S., Winsberg, S., Donnadieu, S., De Soete, G., & Krimphoff, J. (1995). Perceptual scaling of synthesized musical timbres: Common dimensions, specificities, and latent subject classes. Psychological Research, 58(3), 177-192.
  • Zahorik, P., Brungart, D. S., & Bronkhorst, A. W. (2005). Auditory distance perception in humans: A summary of past and present research. Acta Acustica united with Acustica, 91(3), 409-420.

Next lesson · 5 of 10

Noise can make synths breatheAcoustic instruments carry noise that moves with each note. Give a synth's noise the same envelope and it fuses into the sound instead of sitting beside it.

Previous: Brightness is not clarity

Portrait of Virzy Guns

Written by Virzy Guns

Songwriter and producer, founder of Virzy Guns Production. Now building HealingWave.