Please note the examples on this page will be easier to understand with headphones or two speakers placed a few feet or more apart.


Panning describes the strength of a signal between two channels, typically left and right.

Panning is useful for three main reasons. It can act as a way of emulating the position of sound sources within an environment, it can help seperate sounds to make them easier to discern (something outlined in the "Mastering" section), or it can simply be an artistic touch on a musical track.

An audio clip that pans from left, to center, and finally to right.

Panning does not always have to be for giving the listener the impression of instruments being in different positions. Many times just a slight offset in panning left or right can help the listener differentiate between sounds which might otherwise overlap or overpower each other when they are centered.

This is an especially crucial step in the art of foley work (adding sound effects to a video track) which has specific examples in a later section.

Mono VS. Stereo

A strict definition of stereo sound in a digital format could be two audio tracks, one for the left and right channels, being played back at the same time. The reality is slightly more complex.

Stereo sound takes advantage of the precedence effect, a kind of aural illusion, to make it seem like what is actually one sound doubled is still just one sound. It is dependent on a sound from a single source being delayed, or otherwise modulated, in an incredibly minute way. It's similar to how your eyes actually receive two different visual signals but your brain combines them to percieve one singular stereoscopic image. In the real world this is achieved through two factors.

  • The direction of the sound. Allowing you to place the "position" of the sound.
  • The delay between the sound hitting one ear and then the other.

The second factor is the important one in emulating the effect. The engineering of this is something better discussed later in the mastering section however. For now we just need to be aware that most engineered sound, especially music, make use of two seperate tracks to emulate the effect of sound coming from a left or right direction and at the same time increase or decrease the *percieved* delay between left and right ear.

A small audio clip WITHOUT stereo separation.
A small audio clip WITH stereo separation.

The sound should feel more as if it is coming from all around as opposed to right in front of you. This effect can be used to immmerse the listener in an artifical environment that does not try to recreate a physical setup.

The key thing to remember about stereo sound is that it is NOT the opposite of mono sound. You can't have stereo sound with a mono track but just having a different audio track for each channel doesn't automatically create the impression of stereo sound. Having your tracks split to stereo is the opposite of having them panned "center" or "mid".

Applying stereo effects is covered in the "Mastering" section.

Envelopes, basic ADSR design

Some programs may add extra parameters, such as cut or delay, but for the most part the compression and playback of samples will be dictated by the ADSR envelope. The acronym stands for Attack, Decay, Sustain, and Release.

  • Attack
    This will describe the "ramp" up into a sound. Increasing it generally will modify the amplitude from nothing to the initial crest of sample shape.
  • Decay
    The ramp from the peak of the attack to the sustain level.
  • Sustain
    This is the part of the signal that will be consistent while a note is live (or while a key remains pressed).
  • Release
    The release will dictate the ramp of the sample from the moment input is stopped to the end of the actual output. In other words, how much of the sound continues to play after a note has stopped. Though it might sound similar this is *not* the same as a reverb effect. Samples within the sound are not repeated.

As seen on the official FL studio site

The main benefit of being able to shape a sound envelope is to change the impression it leaves the listener. For instance if you wanted a more stocatto like effect for a sample, like a snare drum, then you may want to lower the Attack and Release values to next to zero. This would have the effect of the sound reaching its initial peak very quickly, and having minimal effect as the sound fades, making the sound spike in and out of existence quickly.

On the other hand you might want something slower like the stroke of a violin bow. In this case, even if you're using a synthesized sound, you may want to increase the Attack and Release values to give the sound more of a buildup and falloff like a real instrument would have.

What is gain?

While using software that modifies sound in any way you may encounter a dial or setting that is labled simply as "gain". When played with you'll notice the volume getting quieter or louder and no other noticeable change. So why is this dial simply not labeled "volume"?

The difference between "volume" and "gain" is fairly simple.

  • "Volume" describes an exact signal power.
  • "Gain" describes an addition to an existing signal power.
  • Gain can describe many measurements but in sound it is more often than not measured in decibels. Specifically using the following formula.

    Gain = 10 * log (Power out / Power in)

    For comrpessors specifically you'll often see a knob labeled "type" or "knee". This allows you to attenuate the way the amplitude change is applied after a given level has reached the threshold amount.

    Clipping happens when the input signal strength exceeds the amount that can be recieved / recorded and information is lost.

    With any gain plug in you want to make sure to use the effect in conjunction with some kind of waveform monitor to get a visual representation of your peak values. It is very easy to apply too much gain to a sound, creating clipping, and therefore the loss of information.

    Lastly, it should also be noticed that you shouldn't be surprised if someone differentiates between not only volume and gain, but "level" and "loudness" as well. Typically "level" might also be a magnitude coompared to a predefined reference but might be measured in pressure levels while "loudness" tries to include the "percieved" volume (which can in fact change between different frequencies even if they are at the same volume level).

    The compressor plug-in included with FL Studio has a "Gain" knob that lets you adjust the final signal strength. You'll see knobs like that on many plug-ins.

    Pitch stretching

    Many times when editing sound you'll find you'll want to adjust the pitch of a particular sound. This is not as easy as you might think.

    Remember that what we percieve as "pitch" in sound is dictated by the frequency of the sound waves detected by our ears. Higher frequencies are percieved as higher pitches. So if we want to increase the pitch of a pure tone then we increase the frequency. This is not as simple with any sound that is not a pure tone however.

    More complex sounds like voices or music are based on the idea of a constantly changing frequency. They are one and the same. Changing the frequency rate of the incoming signal therefore changes the rate of the percieved sound.

    This is why we have the problem of pitch shifting changing the percieved "speed" of a sound.

    So how do we go about changing pitches without changing the percieved speed? It's almost always something that will require a specialized plug-in.

    Most pitch shifting will be done through a kind of slicing. Imagine a pure tone with 10,000 cycles over a second. We want to lower the pitch of the clip one octave. Normally if we were to make the cycle rate 5,000 a second then the clip now lasts two seconts. But if we simply drop every other sample in the clip (or interpolate them in some way) then our new rate of 5,000 can fit into 1 second again.

    If we were doing the opposite and wanted to heighten the pitch then we might look at each sample, find the average of them, and use that new value to fill in the spaces in between the old samples to make new samples.