Mix Rescue: Patrizio Cavaliere

Patrizio Cavaliere, a highly regarded, well–travelled DJ, has been producing underground house and electronica for a number of years. He’s inspired by a blend of eclectic rhythms, rare disco, Balearic beats, house and techno. He is the founder of Also Ran Music and resident DJ at Cosmic Soup and this production ‘Dionysus’ will be his third release on German label Movida Records.

Our engineer sorts out the bottom end of a deep–house track.

DJ Patrizio Cavaliere is an accomplished producer of his own material, and he usually takes care of his own mixing in the comfort of his home studio. As with many home studio users, though, he often finds it tricky to judge the lower regions of the frequency spectrum, and thus to get the ‘bottom end’ of his mixes right. With that in mind, he got in touch to ask if I could lend some fresh ears and perspective to his deep-house track ‘Dionysus’.

Why Mixing Bass Is Tricky

The low end can be a tricky part of the mix to get right, both technically and subjectively, but home studios often make this challenge even tougher. The limitations of budget monitor speakers or headphones don’t help, of course, but the biggest issue is usually the room acoustics: low frequencies are often difficult to bring under control using acoustic treatment in typical small domestic rooms, and without an even low–end frequency response, free from unwanted resonances and nulls, it’s pretty much impossible to judge things reliably — you don’t even know whether you have any problematic issues, never mind what action you should take to address them!

Decisions about the bass end of a mix are complicated further by the variety of playback systems used by listeners. Some will be hearing your track via a laptop or mobile phone, with their bass-light responses, while others may listen to systems with over–hyped bass, whether in car stereo systems or popular headphone designs. It’s not surprising that the novice engineer finds it difficult to strike the right balance. In the case of a producer/DJ like Patrizio, a major concern is also how the mix will translate to a club environment, with a PA often containing large subs, which might have been brought along for a specific event and not have been tuned to the room.

This month, then, I’ll focus on how I worked on the bass elements of Patrizio’s track, and how sorting out the low–end instruments left me free to help shape other aspects of the mix as well. I’ll also give a few suggestions about how you might improve your listening environment and how your bass will fare when it’s played back out there in the real world.

Prep!

Before starting on the bass elements themselves, I felt it important that I got whole the mix organised and roughly in shape. There were about 30 parts to play with, including individual drum hits, shakers, bongo samples, bass, pads, stabs and other assorted synth parts, which came in and out at different parts in the track. Patrizio had also provided me with his mix and a reference track, which was both relevant to the style and a good example of a mix he felt translated well when DJ’ing.

I performed my usual grouping and organising of parts, and got a basic rough balance of the track using nothing but the faders and a little panning. I also used the time spent doing that to develop a broad idea of what the mix would require of me. My first step in terms of processing was to apply some high–pass filtering on elements that didn’t seem to require any frequencies around 50–100Hz. The effect of such filtering can be quite subtle — clearing out this unnecessary clutter doesn’t seem to change the sound of the individual parts you’re filtering, but it does have a cumulative effect, eventually creating much more space for the lower frequency elements in a mix.High–pass filtering out any unnecessary low information on various mix elements helped make space that the bass tracks could occupy.

Splitting Up The Bass

The bass elements themselves were quite straightforward, with just one kick drum and one synth bass part that Patrizio had created with his Mini Moog. Each needed its own place in the mix, and obviously had to be full and generous in the bass department, but I also needed to ensure they’d make a worthwhile contribution when the track was played on smaller speakers.

I decided to use an approach called ‘multing’, which simply means creating a duplicate copy of a track, in order that the two copies can be processed differently. Sometimes you’d do this to apply different processes in different sections of the song, but here my aim was to create a sort of crossover filter, so that I could process a ‘low’ bass track and a ‘high’ bass track independently.

To do this you just need to add a high–pass filter to one part and a low–pass filter to the other. After a little experimentation I settled on a low–pass filter set to 335Hz, but with a very shallow curve that meant it reached down to something like –20dB around the 1kHz mark. For the high–frequency track, a slightly steeper high–pass filter centred on 85Hz reached 0dB at 1kHz, rolling off to about –12dB at 100Hz. I spent a while playing around with the steepness of the filter curves because I find they can otherwise sometimes create a significant ‘null’ at the point they cross over. Being careful not to overdo it — which would create a big and unwanted hole in your track! — you can use this effect to scoop out some potentially ‘muddy’ frequencies around the 200–350Hz range, without resorting to further EQ on the part as a whole.The bass part was multed, with each version filtered so that the low- and higher–frequency aspects of the sound could be processed independently.

I wanted to make sure that the low–bass part was controlled enough that I could use it quite generously in the mix, and that meant controlling it both in the frequency domain and in terms of dynamic range. Thinking about the variety of playback systems the track might encounter, I decided to remove the really subsonic frequencies (below 25Hz), as they weren’t really adding anything useful and removing them would leave me with more mix headroom. I then applied a serious amount of compression and limiting, to pin this low bass firmly in its place, removing most of the dynamics and movement in level between the different notes of the bass line. Using a compressor with a fairly fast attack and release, I ended up with about 6–10dB of gain reduction, and then paired with a brick–wall limiter, which tamed things by a further 5–8dB. When you start focusing on this process, it’s quite interesting to observe just how much difference in terms of gain reduction there can be between the different notes of a bass line — this can give you a good feel for just how much energy and space is often taken up by the low bass in a mix.The low bass was heavily compressed and limited to hold it firmly in place.

A Test Of Character

With the low bass now providing the desired power and chest–moving qualities at high volumes, I could use the high bass region to give the part a bit of character, whilst also ensuring it would make a positive impression on smaller speaker systems.

In contrast to the approach I used for the low–bass track, I didn’t set out to apply any compression but, instead, set out solely in search of character. I experimented with a number of different plug–ins before settling on what, when it comes to bass processing, is an old favourite of mine — Sound Toys’ Devil–Loc Deluxe, which is an emulation of the old Standard Audio Level–Or.

Technically, the Devil–Loc was compressing the signal, but that was more of a side-effect; I was using it more for distortion and tonal sculpture than anything else. Used fairly sparingly, this helped bring a nice bit of ‘growl’ to the bass line, which seemed to work really well. I found myself automating the blend control, to avoid the effect becoming too stale, with the amount increasing towards the end of the track. Introducing this touch of distortion, along with a whisker of gentle EQ boost at 1kHz, meant that the bass now had a slightly pronounced mid-range — perfect for helping it to be audible on smaller playback systems.High–bass processing: a touch of distortion, courtesy of SoundToys Devil–Loc, and a small EQ boost to provide some extra mid–range presence.

Kick Drum

With the kick drum, I took a similar approach to the bass synth by ‘Multing’ or duplicating the track. This time, rather than copying the part, I simply sent some of the signal to another track (some DAWs let you route to other audio tracks, others will require you to use a group bus or dedicated effects channel for this). Although I quite liked the sample Patrizio had used for this part, I felt it was a bit lacking in low bass; the sound was nice but it didn’t quite feel powerful enough. So, after setting a low–pass filter turning over at around 160Hz on one of the two channels, I exaggerated the low end considerably with the Waves Puigtec plug–in, a Pultec EQP1A emulation, at 100Hz. On its own this channel sounded quite over the top, but the idea was that it could be blended in to taste with the higher, more ‘normal’ kick drum track. As with the low synth–bass track, I also compressed and limited this part heavily, so those low frequencies were firmly held in place — and that meant that it could be pushed in level where required, without risking any nasty surprises.The low frequencies of the kick were treated to parallel processing, with heavy compression and EQ greatly exaggerating the low end.

The remaining higher-frequency copy of the kick drum — the ‘normal’ kick drum sound — required very little processing. I set a high–pass filter at around 150Hz, to leave those frequencies free for the low kick part, and spent a while adjusting the shape of the curve to create a suitable crossover. The effect of the kick drum and bass synth having nulls at different frequencies, as created by my DIY crossovers, was that the two parts occupied their own space — if they occupy the same space, they tend to interfere with each other, robbing the track and its groove of impact.

Often, when working on electronic or programmed material, you can find yourself doing very little mix processing and worrying that you’re not doing enough, but remember that drum samples or loops have probably already been compressed and EQ’d two or three times before you get your hands on them, so always trust your ears! The only other treatment this particular kick track required here was a whisker of a boost around 3kHz and 4.5kHz, which could be fine–tuned later if the kick needed a little help to cut through in the mid–range.

Multiband Options

Whilst I settled in this mix on splitting my bass and kick drum into just two parts, you could easily develop the technique by adding a third band. This would enable you to go further with your split processing, perhaps by adding a stereo element to your higher bass part, courtesy of a phaser, chorus or some other spatial effect: whatever you find works. When doing this, you have to be careful to make sure those crossover points aren’t too wide, though. (Sometimes I’ll even draw a little sketch, just to get an idea of how the low and high–pass filters are crossing over.) Also, you need to make sure that your DAW’s plug–in delay compensation is up to scratch, because some CPU-heavy, or older, plug–ins can introduce a little latency even when automatic delay compensation is switched on in some hosts, and if you’re processing different versions of the same track with different plug–ins, that will mean they become slightly out of phase. The bottom line there once again is to use your ears.

The technique described in this article was effective here, but it doesn’t always work so well, and I certainly find it less effective when working with live drums and bass — where each hit and note is slightly different, and the result seems to ‘move around’ that much more. With dance or other programmed material, though, the nature of the music is more consistent, so you get a more pronounced sense of the low, medium and high–frequency parts of the sound — which means it’s a lot easier to home in on specific areas of those sounds.

The bass was checked on a few different speakers, including a ‘grot box’, which provided me with a good indication of whether the bass was becoming over–inflated.

It’s amazing how quickly things come together in a dance mix once the bass is working as it should. Patrizio had worked quite hard on his arrangement so the rest of my job was very much about balancing levels and setting the panning to create an interesting stereo field. Once I’d had some fun playing with delays and effects on some of the synth stab parts, my focus was largely about making sure the mix had the right broad balance between the bass elements I’ve already described, the synth and drum elements sitting in the mid–range, and the higher shakers and hand claps, which provided most of the top–end information.

Bass–ic Sanity Checks

I’ve developed a few very non–scientific techniques for checking if a bass part is too prominent in a mix, and I use these in conjunction with reference mixes to keep me on track. Although I don’t employ a dedicated subwoofer in my setup, I have a set of Mackie HR428 speakers to complement the Adam A7’s I use as my main monitors, and I can, if required, pump a mix out to a larger PA in a nearby rehearsal room. But checking bass isn’t all about large speakers — I find my ‘grot box’ particularly helpful in checking the bass. I certainly wouldn’t do any serious EQ work listening on it, but if I can get the volume up quite high without the speakers waving the white flag, that seems to be a good indicator that the bass is nice and tightly controlled! If I need further convincing, I’ll compare my mix to a couple of reference mixes on my headphones — removing the room completely from the equation.

People often seem happy these days to bypass the professional mastering stage, but if you never use a good mastering engineer you’re missing out on one of the most important contributions they can make. Proper mastering engineers have serious listening environments and ears that are tuned by experience to judge appropriate levels of bass for real-world playback systems — so even if you only get a track or two professionally mastered, that could provide you with a good sonic reference for your other projects. Despite spending unhealthy amounts of time on their own, most mastering engineers will also be happy to give you some often quite specific feedback on your mixes — and that can be enormously helpful in improving your own listening and mixing skills.