Technique #1: Spectral Analysis

In techno the kick drum and bassline are a fundamental aspect of the track, but they are often hanging out in the same frequency range, and we all know that this is like tequila joining the party, there’s going to be a fight at some point. 

You know about rolling off the low end, and EQing the bass to “fit” around the kick, but despite having decent monitors, it’s hard to hear “down there” in an untreated bedroom/slash music space. 

So, how does one monitor low down in upside down acoustic spaces?

Spectral Analyser: It’s no magic pill, but a spectral analyser helps see what you might not be able to hear.

http://www.voxengo.com/product/span/ 

http://www.bluecataudio.com/Products/Product_FreqAnalyst/

https://www.meldaproduction.com/plugins/product.php?id=MAnalyzer are a good place to start, and they’re all free.

How to do it:

• Strap a spectral analyser across your main stereo out. 
• Solo your kick and the analyser will show you the frequency range it is using.
• Solo your bassline next and you’ll soon know exactly which frequency it is occupying. 
• Repeat for all elements in the track.

Voxengo SPAN allows routing two elements into the analyser so you can do side by side comparisons within the same plug-in.

Now you can EQ sounds with much more confidence and accuracy because you can see where they are conflicting with each other.

Technique #2: Check The Mud

Low end monitoring and mixing is a major issue for most modern producers working at home - poor acoustic environments, less than accurate monitoring, and neighbours who don’t like their walls rattling at 3am in the morning.

There are forums and discussions abound tackling this muddy, boomy issue, but answers involving acoustic treatment, speaker placement, speaker type, diamond coated cables dipped in gold flakes and dried by moonlight are of little practical use to beginner and advance tweakers alike.

So how can one do it with the tools at hand?

Parametric equaliser. Specifically, a low pass filter with a cut off at 100Hz exactly. It looks like this: 

How to do it:

• Strap a parametric EQ across your master out. 
• Set a LOW PASS filter centered at 100Hz and use the steepest cut off it can muster, you are cutting out all the frequencies above 100Hz. 
• Enable the EQ and play your full mix through it.
• Mute all instruments that SHOULD be hanging out down here - your kick and bass line, for example.
• Now that we have isolated the lowest frequency range, we can hear which elements are reaching below that range and shouldn’t be there. 

Even if you have terrible monitors/acoustics, this method still helps. Engaging the 100Hz lowpass makes it obvious which sounds shouldn’t be down in this range. You will be amazed at what you find taking up space down there. 

It even works with headphones.

Repeat through all elements in the mix (if necessary) until you are satisfied that the low end has been cleared up, then let kick and bass back into the room and they will have much more room to play nicely without getting all up in each others grille, which is when they start fighting. Again.

Technique #3: Mix Against A Pink Noise Test Signal

In techno the main reference level is straightforward, it’s pounding away under the song. Most dance music producers will “mix to the kick”, i.e., all other elements in the song are balanced and leveled relative to it, but in less beat driven music or more complex arrangements a kick is not necessarily prominent.

So what then?

A calibrated pink noise reference level. The idea in a nutshell: mix into pink noise and use it as a global reference level for all mix elements. Use pink noise (and not white noise) because pink noise more closely matches how we hear. (How and why that is, is not in the scope of this article.) 

With this method, sonic elements that are too loud immediately stand out, and elements that are too low are equally revealed. Of course some sonic elements will cut through (or be masked) more than others and you will always need old fashioned ears, tweaking and adjustments after the fact, but having a calibrated, dedicated “anchor” to mix around works wonders. 

No one sound tends to dominate, and incremental volume creep of mix elements over a session are held in check.

This technique takes a bit of practice and getting used to, but once you get the hang of it you’ll be amazed at the results. 

You will need a test tone generator that has a pink noise option and a metering plug-in like Voxengo SPAN. 

The free MDA Plugins bundle http://mda.smartelectronix.com/ has a tone generator that allows signal to pass through, but it does not work in Logic X unless you have a 32 bit host like 32Lives running.

I will explain how to set up this technique using Logic X native tools as well, but the MDA test tone/SPAN is the easiest to set up. 

How to do it:

First the set up with the MDA Test Tone running into the Voxengo Span metering plug-in. 

• Load the MDA noise generator across your main stereo out.
• Set mode to Pink Noise.
• Set Output to all channels.
• Enable “Audio Through” check box.
• Set level to around -23db.

MDA Test Tone Pink Settings

• Load Span.
• Select K-14 metering. (K-14 metering places the zero -14db below Digital Full Scale and will leave you with a ton of headroom after leveling your mix.)
• Adjust the level of the pink noise so that your peaks hover around 0 and RMS level sits around -8 or -9. For this technique we are more concerned with RMS than Peaks, but I will not go into details on why here.

Span k14

Set up in Logic Pro X. 

Since the Logic Pro X Test Tone Utility does not allow for audio to pass through, it needs to be set up on a stereo bus that feeds to the main stereo outs.

• Load Utility Test Oscillator onto a dedicated Stereo Aux channel.
• Select Pink Noise.
• Set level to -16 dB.

Pink calibration in Logic Pro X

• Load Span.
• Select K-14 metering. - 16dB input level of the pink noise should have your peaks hover around -0.5 and RMS level sits around -9. 

dBFS equivalent levels to K14

Span dbfs

The walk away here is that this technique works on two levels; first, it gives you a calibrated, objective mix element to use as a reference for all elements, and second, it aids finding a balanced mix with tons of headroom for post processing and mastering.

Technique #4: Process Mid Side Information

Mid-side processing is the ability to attenuate or accentuate elements of a sound, bus or final mix based on its location in the stereo image.

The Mid Channel is any sonic information in the center of the stereo image, when this is accentuated we perceive a more centered, narrow, or mono sound.

The Side Channel is made up of sonic information at the edges of the stereo image. When this is accentuated it appears to widen the stereo image and add space around the centered image.

Conversely if we attenuate the mid channel a few dBs but leave the side channel at 0dB we can create “space” in the middle of the stereo image without raising levels, or we can lower the sides and create a more mono,  centred sound, effective for bass lines and low end elements that should be mono anyway.

How to do it:

Load up an EQ/plug-in with M/S capabilities like the Logix X stock EQ, Ableton’s EQ 8, Fab Filter Pro Q 2, and Izotope Ozone to name a few.

Processing Single Sounds, a plastic example:

Let’s say you have a bassline with a chorus effect and would like the lower end to be mono but want to keep the “width” the chorus provides to the upper harmonics.

The EFX chain goes like this bass > chorus > M/S EQ:

In the first image all EFX and EQ are off and the resulting spectra look like this:

Let’s switch on the chorus (using stock preset Mega Wide Chorus) and see what happens:

MS Bass EFX chain add chorus no MS processing

MS Bass EFX chain add chorus no MS processing

Now if we look at the phase analyser on the right we obviously have stereo information in the audio, but looking top to bottom there is some stereo information reaching all the way down to past 30 Hz. The FFT looks the same.

Let’s see what happens when we add a side only high pass filter at 48db per octave and center it around 220 hz.

MS Bass EFX chain add chorus with MS processing - side only 220Hz high pass

MS Bass EFX chain add chorus with MS processing - side only 220Hz high pass spectra

Now when we look at the spectra we do see some changes on the FFT in the fundamental,  but now we have a hybrid of the mono/wide signal. What this has done is effectively make everything below 200Hz mono but kept the stereo chorus effect in the upper frequency range, the best of both worlds.

Bus processing or on the final mix:

Applying the same technique as in the plastic example above, but across a bus of instruments or across the whole mix (and way more subtly) can work wonders on bringing clarity and stability to a mix.

Carefully attenuating some of the side information from the whole mix can quite often “anchor” all the elements in a surprising way because it clears out the bottom end, pushes the sides “up” and can give the whole arrangement space to breathe - when done well.

Creative MS processing.

Voxengo have a free MS processor plug-in called the MSED. Its actual purpose is to decode audio recorded using the mid-side stereo micing technique, but it can be used for so much more. Rather than being an equaliser, it effectively works as a mid side GAIN controller, meaning you can raise or lower the gain of both independently. This plug is great for taking out the “middle” of a sound while keeping its side presence without overtly affecting the tone of the sound like an EQ will. It also widens the stereo image without overt phase issues common with stereo techniques like the Haas effect or most stereo enhancers.

The only limit is your imagination, and processor speed.

Conclusion:

We covered a lot of ground with this article, the idea is to go and tinker with these techniques and find your own way of making them work for your music production and mixes.