Principles of Delay

From Audacity Wiki
Jump to: navigation, search
Peter 3Sep12:
  • Material here is transferred directly from the old pages in the Manual (now deleted).
  • The page title is purely a working-title. I welcome suggestions for a revised title prior to publication.
    • Steve 12Sept12: References to "RAM buffers" etc. imply hardware or real-time effects. Would it be more useful to confine the description to "what" the effect is rather than specifics of how it may be implemented? For example, if the user is aware that the audio signal is delayed for a period of time, I doubt that they need to know anything about "RAM buffers" or other methods by which that delay might be achieved.
      • Peter 17Sep12: following Steve's comment I edited out the stuff abot RAM buffers.

A delay effect is similar to an echo, in that the sound is repeated (after a brief time delay)one or more times after the original sound. If the time between the original signal and the repeated (delayed) signal is sufficiently small, it will not sound like a distinct echo, but rather as a simple reverberation effect.


This effect stores the audio for a certain amount of time, after which it outputs this audio. Many delay effects allow for their delay parameter to beautomated.

The Delay Signal-flow
Diagram of the Delay Effect

Multi-Tap Delays

This effect incorporates several discrete delay lines, each with its own delay time, level(amplitude) and panning. The feedback parameter is usually shared by all delays in a multi-tap delay. Some multi-tap delays feature a modulation section that modulates the delay times, which can be used to create Chorus effects.

Common parameters of a digital Delay

Parameter Range : 0.000 to seconds

The upper boundary of this parameter( depends on the amount of RAM in the hardware unit, or the amount of RAM reserved by the software plug-in. For software plug-ins, the programmers or designers of the plug-in set this upper boundary.

The delay software plug-in Supertap by Waves for example comes in two flavours on Protools TDM systems. One allows for delays of up to two seconds, but another version accommodates delay times of up to six seconds. This is due to the availability of RAM on the DSP hardware of a TDM system. The native(->CPU) version of this plug-in for Windows for example only allows for delays of up to four seconds in length, but there are delay plug-ins that allow for delays of several minutes.

Today's delay units and plug-ins will feature one of several different types of feedback parameters :

a) Percentage Range : 0 to 99(or 100)%
This is used by most hardware units and some software plug-ins. It determines the amount of a delayed signal to be sent back to the RAM buffer. The same delay time is applied. This is the most common parameter type for feedback.

b) Decay Time : 0.1 to xx seconds
A slightly different approach. The plug-in or hardware unit determines the amount of delayed signal, that it sends back in to its RAM buffer, itself. Some may find this more intuitive. It isn't widely used however.

c) Feedback Delay Time : 0.000 to seconds
This is like an additional delay line that takes the already delayed signal as input. This kind of parameter also requires a level setting, which is usually a percentage parameter, like a).


Many delay plug-ins feature highshelf and/or lowshelf EQs, or they feature high-pass and/or low-pass filters. These EQs or filters usually process each delay. The idea behind this is usually to simulate distance, because with increasing distance, the lower and higher frequencies are dampened.

Hi- & Lowshelf EQs

  • Gain : -xx to xx dB
  • Frequency: 20 Hz - 20 kHz

Low- & High-pass Filters

  • Frequency: 20 Hz - 20 kHz
  • Filter Strength : 6, 12, 18 or 24 dB/Octave

These are the most common types of stereo panning parameters you'll find.

-63 +64 0% 100% -100%100%

Simply a panning control for the delay. The feedback parameter may also have its own panning control.