Before we look into some tips, let's cover what mid-side is - in short, the mid image is everything that’s completely centered. A signal that’s centered comes out of the left and right speaker identically giving the listener the impression of it being right in front.
This is also known as phantom center, and it’s why a stereo signal, when both channels are identical, sounds mono.
The side image is all signal that is completely left and right, or signals that are not identical but are played at the same time. When the mid and side combine, phase cancellations cause signals to be perceived in a 180-degree field.
Adjusting the mid and side changes this perception, which we’ll definitely cover in future chapters.
Let’s listen to a stereo mix, and isolate the mid and side image.
Any stereo signal can be separated into mid and side by inverting the phase of either the left or right channel. I like to use the free MSED plugin by Voxengo to create mid and side channels quickly and reliably (meaning without artifacts or unwanted signal changes).
I’ll first duplicate the stereo track I want to make mid and side, insert MSED on both and then mute the Mid image on one and Side on the other. Let’s do this and then manipulate the mid and side images to hear how we can change the balance between the 2.
Also, again, this plugin is free so please try this out for yourself.
Let’s start looking at using Mid side tips, starting with using an M/S EQ on the drum bus - I’ll use the Kirchoff EQ, insert it on the drum bus, and create some bands. First, I’ll use a high pass filter on the side image, up to about 80Hz.
This is going to make my lows, like the 808 or kick more focused and driving. I may even want to make a mid bell band and amplify 80Hz to further emphasize the kick.
Then I’ll create a high shelf on the sides and amplify the highest frequencies, causing pleasant-sounding stereo expansion.
I could also make a mid band around 200Hz or on the snare’s fundamental and amplify that.
Let’s listen to how this EQ targets and augments important aspects of the drums.
On our instrument bus, we can do the same this we did in the last chapter, but of course, tailor the bands to the signal we are affecting. One thing I’ve found that makes synths and guitars sound great is a boost to roughly 300Hz to 600Hz on the side image.
Although we typically don’t amplify this range in a mix due to how it can quickly sound cluttered, on rhythm and support instruments, it creates a full and wide sound. Let’s take a listen.
Let’s go back to separating a signal into mid and side using the MSED plugin, but this time insert the NewFangled Audio Punctuate transient shaper onto the side image signal. The side image is typically full of softer sounds like reverb reflections and signals with a long attackS to their ADSR.
By putting a transient shaper on the side image, we artificially introduce transients to this signal, resulting in a much more detailed and interesting signal.
Let’s take a listen to this on our drums to better understand it.
If we’ve separated our signal into mid and side, we can compress the mid but leave the side alone to cause dynamic stereo expansion - or we could compress the side only to cause dynamic stereo narrowing. Stereo expansion will be more common so let’s use the Impressor by PSP.
I’ll insert it on the mid image, and compress it. This way, whenever the mid is attenuated, the side will be louder relative the to mid image, creating a wider image.
Let’s take a listen and notice how the signal becomes wider.
On our vocal bus, we may have multiple performances that have been panned or reverberated, causing info to be moved to the side image. Using a mid-side EQ, we can affect the balance between this side and mid image, as we’ve done in previous chapters.
Let’s boost 3-5kHz on the mid image and dip it on the side to make the vocal cut through. Then let’s boost the side image’s highs to introduce pleasant stereo expansion and air, and do it to a lesser extent on the mids.
Lastly, let’s subtly boost the vocals fundamental on the mids, and use a high pass filter on the side, as we did to the drum bus.
Let’s listen and consider how this improves the vocal's presence and power where needed and augments some details on the side image.
Last up in terms of processing busses, let’s use a mid-side EQ on our full mix bus. Honestly, I’ll do a lot of the same, cut the lows on the sides and maybe boost some of 3-5kHz on the mid image while also boosting the side’s highs.
This time though I’ll make the changes subtle, find important target frequencies like the root note, and center the boosts on various harmonics or high orders of the root note. This will make the overall mix sound both more impressive in terms of the stereo image, and also more musical.
Let’s take a listen.
As we discussed in chapter 5, the side image has softer signals when compared to the mid image. This can cause the more powerful mid image to mask or cover up the side - to avoid this I like to use upward compression on the side image.
In short, I'll insert an upward compressor on the side channel and increase low-level detail so that the side image, and it’s easier to miss details, has a better chance of being heard. Keep in mind this will amplify the side, so adjust the output as needed to maintain the balance between the 2 images.
Let’s take a listen.
Since the stereo image is usually static or kept the same, we can surprise the listener by introducing dynamic stereo expansion or narrowing. For example, I could attenuate the side image, but then drastically increase its value to rapidly expand the image - causing unnatural but entertaining stereo expansion.
A more subtle method would be to enable the side high-frequency band on an EQ when the chorus hits to help differentiate it from the verses. Let’s take a listen to the first method we described.