I find there to be many a common misconception in understanding the stereo field so I thought I would clarify some things. Many thanks to Aliki Rodgers for his contributions.
What is Mono?
Mono (short for monophonic) means one channel of audio. Stereo means more than one. When you record a vocal into your DAW using one microphone and one channel on your interface, into a mono track, that is mono. When we have a mono signal in our DAW, eventually it gets sent to a stereo bus, which effectively converts the signal to stereo, then it’s played out of both output channels (L and R) at the same time and at the same volume. This is what creates a centre image. Mono information in a stereo environment equals centre information. As our ears hear the same thing from either side of us at the same volume and at the same time, our brains localise that as centre. This makes sense as that’s how it works in the real world, when a sound comes from our front, it reaches both ears at the same time.
What is Stereo?
From this Wikipedia article
Stereo or stereophonic sound is the reproduction of sound using two or more independent audio channels
Couldn’t have said it better myself. In music, we use the left and right configuration of stereo as to create a three dimensional sound stage which would be impossible with just a single channel (mono). How would you make something sound like it’s coming from the left if you only had one speaker directly in front of you? You couldn’t.
How we localise sound in the stereo field
Interaural Time Difference
This is fundamentally the reason why headphones don’t produce a true stereo image. In the real world, when a sound comes from the left, the right ear hears the same thing, yet slightly later than the left ear does. Our brain uses this information as well as interaural intensity difference to localise the sound.
Interaural Intensity Difference
This is the other piece of information our brain uses to localise sound. Using the same example used above, the right ear hears a quieter version of what the left did due to the fact that the head shadows the sound resulting in a lower level compared to that the left ear.
How does one achieve stereo width?
For me, the term ‘stereo width’ means the sensation you get when you hear a wide sound. When it seems to be coming from multiple angles, like a wall of sound. Something much bigger than what would be possible if we only had one speaker.
Stereo width is achieved when what comes out of the left is different to that of the right. For example, if I’m tracking guitars, I will record the same part twice and pan them left and right respectively. This is effective because they are two different pieces of audio even if I am playing the same riff.
However, if I made only one recording, panned it hard left, duplicated the audio and panned that hard right, I’m simply achieving the illusion of centre. This is a pointless exercise as I might have well just kept the one copy panned to the middle. The effect is the same.
Stereo Balancing vs True Panning
Bit of a rabbit hole this one. Below is the gist of it, but I implore you to check out this article where Edgar Rothermich explains the difference in very good detail. It’s presented as a Logic Pro tutorial but it’s stuff we all need to know.
When you ‘pan’ a stereo track (unless with true panning), you’re actually turning one channel up and turn the other down. This is actually called stereo balancing as it’s likely the default behaviour of your DAW when you are presented with one pan pot for a stereo track. However, if you DAW supports True Panning, it’s a whole other kettle of fish. With this method, both channels (left and right) actually pan (move) in the direction you specify, as opposed to one being turned up and the other down. I’ll give you an example.
Imagine a situation where you had a stereo audio track in your DAW, and on that track was an audio clip, and in that audio clip you somehow had a piano on the left channel, but a vocalist on the right channel. When you use the pan pot in Stereo Balancing and ‘pan’ to the left, you’ll simply be turning the vocalist down and the piano up. However, with True Panning, you’ll be panning (in the true sense of the word) the vocalist over to the left to join the piano. Nothing on the right, both piano and vox to the left. Different DAWs support this in different ways so it’s important to know whether that pan pot you are using is in fact a true panner or a stereo balancer. In the context of a mono track going out to a stereo bus, it has to be a panner as there is no L and R channels to balance there.
Where does Pan Law come into all of this?
Without pan law, anything panned to the middle doubles in power. Pan law compensates for this by attenuating by 3dB on anything that is panned to the centre. This way, when you pan something from left to right, the apparent volume is consistent throughout the movement. There is actually more than one standard for pan law, yet another rabbit hole i’ve chosen to pass on for the sake of this article. However, I do recommend you read further.
What do Stereo Spreaders or Wideners do?
They simply take your stereo information and further manipulate the L and R channels so they are more different, resulting in a wider stereo image. They achieve this in many different ways often playing with phase but that is fundamentally the strategy, to make them more different. You can even use these on mono tracks, converting the signal to stereo. Often Stereo Spreaders include frequency bands to be treated individually. While it can feel like it opens up an individual track the resulting movement based on frequency can dramatically distort a stereo image. Which is why I personally feel that when you have the option to go back to the mix and further explore panning instruments, that’s usually better than using a widener on the mix bus.
Some spreaders use sum/difference processing AKA Mid/Side to increase the difference between Left and Right, resulting in a wider image. That nicely brings us onto one of the most misunderstood conceptions in audio today…
What is Mid/Side?
Mid/Side is actually Sum and Difference. Let’s break that one down!
What Mid/Sum is
L + R. Exactly the same as summing to mono.
What Side/Difference is
L – R. The difference between left and right. The same as taking a stereo file and flipping the polarity of one channel. That’s actually how it works.
What’s Mid/Side for?
Mid/Side allows you to treat the sum and difference channels independently. In my opinion it can have a place in mastering if used in a well informed manner.
When mastering for example, you may want to increase gain on the side channel to widen the stereo image. As you are increasing the level of the side channel, you are increasing the difference between left and right, thus producing a wider image. Conversely, if you increase the mid/sum channel, you are reducing the difference between left and right, thus narrowing the image. You never actually lose or gain any new content, you simply convert it.
To give you a more real world example, you may want to apply a low cut filter on the side channel with a view of converting everything below the cutoff frequency into mono/sum/centre information. This is how mono makers work. Again, you haven’t lost any energy below the cutoff here, you’ve simply converted it.
All being said, I wouldn’t practice any of this just because it’s there and because you can. Mid/Side processing is not any better than regular stereo processing. It’s just different. In stereo you are treating either or both channels. In Mid/Side, you are treating both or the difference between the two. That is fundamentally the difference.
Bonus: How is Mid/Side decoded back into L and R?
You won’t ever need to do this when using Mid/Side plugins yourself, however it is very simple and worth knowing.
The method of decoding a Mid/Side signal back to L and R is as follows:
L = (M+S)/2 (Mid summed with side)
R = (M-S)/2 (Mid minus side)
The division by 2 is there simply because the summing process nets a doubling of power of the original stereo signal. We simply negate 3dB to return to unity gain.
Hopefully this really basic arithmetic can help you understand what mid/side truly is.
That’s all, Folks!
That’s everything I wanted to set straight with regard to what all these things actually mean and I do hope it has brought some clarity where needed or just reassurance on what you already expected to be true. Please let me know if you have any questions or concerns about any of this using the comments below. I look forward to discussing further with you!