Quite a lot of times, I have seen this issue with a lot of people, especially our students, when they struggle mixing drums in their music production project. It always amuses me, why is this the case, why it would be so difficult for people to know exactly what they want from their drums.
So, I will try my best to simplify most of the things about mixing drums processing and to go about it. This is going to help people even those who knows how mixing their drums sounds, because a lot of things are quite unknown to the people until they stumble upon the problems associated with it. So, let’s get started.
Things you need to know about mixing drums/percussions:
There are certain things which any drum or a percussion sound would have:
If you look in the above-mentioned picture, the light blue highlighted portion depicts the transient. It is the initial burst of energy which is responsible to create a click sort of sound when played. The above sample is an example of a kick drum, and the initial highlighted portion is the click sound of the kick which typically have a lot of high-frequency sound associated to it.
The latter part of the sample in the above-mentioned diagram is the body, the body could be comprised of either the low-frequencies or even upper-mid or mid frequencies. And this is completely dependent on the type of sample or percussion element you’re selecting for the track.
At this point, we all can agree that there are two things associated to any of the percussion sound. This gives us a better understanding of how to deal with our percussion sound.
The Mixing Chain:
When we talk about a typically mixing chain, I really want to break the stereotyping of having conventional mixing chain; and to go about in this process of doing so will not only limit you from getting the best out of the sound but will also make certain constraints or bounds with respect to the audio you’re treating.
Here’s an example, I have been talking to a lot of people, especially our students. Those people who are self-taught producers are not very well informed about the technicalities of the plugins they use in the mixing chain. Of course, if it sounds good, then nothing like it. Andrew Scheps himself conveys this message that, “Whatever you do inside your DAW doesn’t really matter, what matters is what comes out of your speakers!”. I am a huge proponent of this statement, because at times, we break the rules to make sure that it sounds a bit more interesting and different from the rest of the tracks out there just to be unique. I mean, if you know the rules very well, there is no harming in breaking one.
But, blindly breaking the rules without knowing the consequences is something which I call it as blind optimism. This is where the knowledge we seek comes in to foreplay. The mixing chain is there for you to guide you through, how to go about mixing one element. But it doesn’t necessarily mean that you need to blindly follow a typical mixing chain without knowing the consequences of it.
This typically happened with one of our students and he got it all wrong, he applied always the same settings on every track, for instance, the settings I had shown in the classroom, he noted down those parameter settings and applied the same settings on every kick drum he has ever used in every track. This is why understand a plugin will help you a lot in terms of where to use what, rather than just blindly copiously following some of the settings dictated by people who have an absolute valid reason for their doing.
So, a typical mixing chain would go like this;
Now, only, if you know things about each plugin in its entirety, only then you would be able to exploit the mixing chain to the fullest by placing these plugins according to your subjective preference.
A lot of people by looking at the picture would start catcalling me, by saying, how can you so sure about the mixing chain, that the compressor would fall before the EQ or After it. And I would love to answer that question as well. But here’s the catch, I just can’t explain every plugin which is mentioned above in the mixing chain, in its utmost details, because that would be a lot time-consuming and it would require for me to sit an entire week writing about compression, types and how to go about it. But will surely try and explain certain things which are essential to mixing drums or percussion.
The biggest question of all, “Compressor before or after EQ?”. The answer to be honest is not that simple, the here is why. Firstly, we need to know what a compressor does, I presume that you guys have the basic understanding of what a compressor is. So, when we add a compressor in the mixing chain, there are two reason why we adding it, one to make the audio track more consistent, in terms of the level of course, and second, for shaping the audio. When we add an EQ before a compressor, it would make a lot more sense, if we are subtly compressing the audio and not using compressor in an over exaggerated setting.
When used in an over exaggerated and aggressive settings, the compressor may or may not change the sonic characteristics or in short, the Timbre of the sound. Which will make you to add another EQ plugin afterwards to fix and undone the Timbre which was disrupted by a compressor. So, the answer here is this, if you want to make the compressor work in an exaggerated setting, it best that you add a compressor before and add the EQ later to reduce the redundancy and save the CPU cost deliberately. Or else, you can keep stacking piles of EQs after every plugin that you have added, just to make it sound better.
Also, we add Saturation before compression and EQ is because, Saturation adds something called as Fold-back Error, which you may want to avoid it at all cost. As long as you’re using good A2D converters, and using higher sample rates, typically more than 96,000Hz, it shouldn’t be a problem. But if you’re working in 44.1Khz or 48Khz of sample rate setting, you may want to add the EQ after the saturation because some of the harmonics may get folded back. Another alternative solution is to use a Saturation plugin which has oversampling mode, but this is again at the cost of CPU. Hence, the best way to deal with this problem is to add an EQ in the way shown in the mixing chain above.
Now, let’s start with our mixing drums chain which we are going to apply on percussion/drums:
Not all of the people know about this one, but majority of the samples we get from various sample banks are already heavily processed and saturated. So, here is the catch. You should really know when to apply Saturation.
One of my favorite plugins for adding saturation is from the company called, “IK Multimedia” and the plugin’s name is T-Racks Saturator X. There are various reasons why people like a saturation plugin, for me, why this is clearly a winner is because it has all sorts of saturation modes. For different emulations, I certainly don’t have to go to look for other saturation plugins, Saturator X gives me one stop solutions for all types and emulations of saturations. TAPE, MASTERING, TUBE (Push-pull, Class A), FET – Solid State (Push-pull, Class A), TRANSFORMER (Iron, Steel) are various saturation emulations which T-Racks offers. You can also choose to do mid-side saturation in this plugin. Also, there is a built-in brickwall limiter and the plugin has 4x times oversampling modes.
Now, how much of the saturation you may want to add on your drums on percussion? This may be a completely subjective process but to make it objective, you can start by increasing the input gain, while turning the link knob ‘on’, so that whenever you’re increasing the input gain, the output is also decreasing with same intensity.
A lot of people have propensity to add saturation, without managing the amount of it. The saturation adds harmonics, and the more the harmonics the more distorted it is going to sound. Unless, it is intentional, I wouldn’t recommend adding too much of color, mainly for two reasons; it would not sound great when we’d be mastering the track and pulling up limiter gain., and it would make everything sound way too harsh.
The idea of adding subtle saturation is so undermined and a lot of people underestimate this power. Even by simply not choosing to do anything but add the Saturator X plugin in its default configuration mode, helps a lot. Try this yourself, just add Saturator X plugin on all of your tracks in its default setting and map the device on/off buttons of all the Saturator X plugin to one key and try A/B. You’ll notice a minute difference, but that minute difference is also substantial when we master our track. In genres like Dubstep, where the kicks and snares are supposed to sound very distorted, the idea of making everything sounding harsh could work wonderfully here.
Compressor is a main game changer of percussion or mixing drums. Compressor is not here to make things uniform, since, in drums or percussions, there are barely any instances where we may want to make things sound equal. So, what could ideally be done at this point? Two things, either you can use a compressor to change its shape or, you can use the compressor bring up the whole body by using compressor in different configuration.
Like I said, the transient and the body is what we are trying to focus while using a compressor when dealing the percussion/drums. The compressor here with the fastest attack and release is going to compress the transient of the element, and when used in vice versa setting, it would tame the body.
In this case, if things are getting too vague by using one single compressor, since it changes the sound quite drastically, I divide the workload of a single compressor between two different compressors. The first compressor would focus on taming the transient and the second compressor would focus on taming the body of it. In this way, we have a lot better control over the transient of the percussion sounds. If you didn’t realize, we have just made a transient shaper. This is exactly how transient shaper plugins work.
Coming back to compression, for me, I usually like to add an FET compressor, if the transient is not up to the mark. If the transient is too loud and the body is too dull and would want to bring up the body a little up, this is where I use a digital compressor (could be a stock DAW compressor as well).
FET 1176 works like a charm if you want to bring up and give more prominence on the transients rather than on the body. In most of the American based FET compressors, you’ll find the attack time is to be fastest clockwise, and slowest counter-clockwise, which is exactly the opposite of most the compressors. So, 7 here would make the fastest time setting and 1 would make the slowest. Make sure to select the attack time typically somewhere in between of 5 -7. The attack time of FET compressors is very fast, typically around 4microseconds. Hence, trial and error method are your best bet.
Also abbreviated as subtractive EQing, is used to eliminate those frequencies which are either resonating/harsh or are not the part of the sound. The frequencies which are not the part of the sound typically falls under the lower category, below 30Hz are called low-end rumble or rumbling frequencies. These frequencies are usually not a part of our sound and should be eliminated because a lot of consumer audio systems doesn’t full frequency response ranging from 20Hz to 20Khz.
One of the best plugins which I would recommend for surgical EQing on everything would be FabFilter Pro Q3. One of the alternatives to this would be F6 from Waves Audio. Both of them are equally competent and have equal potential to do what an EQ is supposed to. Only in one aspect, Waves Audio’s F6 takes the lead in giving the attack and release time controls in the dynamic EQ section. But, nonetheless, both of them are great plugins. We already know how to get rid of the rumbling frequencies, is by applying the high-pass filter with higher slope at 30Hz.
How to get rid of the resonating frequencies? This could be really difficult at the very beginning as your ears may or may not be trained to notice these subtle frequencies.
These frequencies have a character of a whistling sound, so the moment you would solo/audition a bell curve with a high Q factor, you want tumble upon these frequencies which would sound like someone is intentionally whistling. You can dip these frequencies as these are very harsh in nature and may cause irritability and pain to the end-user or the listener. Dipping around 6dbs – 8dbs is what I advise. Lastly, taming a bit of low-mids, upper-mids, presence or brilliance is completely a subjective choice but keep a reference track so you can match the frequencies of the reference with yours.
Never use conventional EQs, or analogue emulation EQs to boost the low-end of the percussions. There might be a chance or a possibility that you may be changing the phase relationship of that frequency which may alter the entire sound, or may cause phase interferences with the bass, in spite of sidechaining it properly.
This is where Linear Phase EQs comes in handy, I would suggest to use these EQs only when you want to increase the low-end of the percussion/drums. It is not advisable to add this EQ on every track as it would take a lot of CPU processing power making it difficult to workout without freezing all the tracks.
You can either use, Fabfilter Pro Q3 in linear phase mode, or you can use another third-party plugin from Waves Audio called Lin Broadband EQ. Increasing mid or high frequencies from this EQ would make no difference as even if the phase relationship changes, it doesn’t make a huge difference with rest of the phases of different elements.
One of the main things people tend to usually forget is to check the mid information as well as the side information of a particular element. Some of the percussions are majorly supposed to stay in the mid information range, like Kick, Snares, Claps, Toms etc. Whereas, some of the percussions are supposed to be present in the side information range, like cymbals. But remember to note one thing, the correlation meter is always going to be your ally while deciding the levels of mid information and side information. Make sure that the mid information of the kick, snare, claps is more as compared to the side information.
The correlation meter should range somewhere in between 0.5 to 1. For cymbals, the side information could be a lot more here but make sure that the correlation meter is not going below 0, towards -1. For MSED from Voxengo is a plugin which I would highly recommend.
It’s a very user friendly and intuitive plugin. One can also mute the mid information or the side information to check exactly what is being played in the mids and the sides.
It is not a great idea to add time-based effects processors on percussion, unless the track is required to or its sole intention is to deliberately add it. You can use reverb, delay, chorus, flanger and phaser here only if it is intentional. This will come at the last part of processing. Adding time-based effects on a return track is highly recommended as it may avoid majority of the phase issues. You can also choose to add a compressor in the return track to apply parallel compression on the drums by choosing aggressive compression settings. This will definitely bring out some of the drum sounds, but not all.
The ratio setting of this parallel compressor is going to be very high, the attack time is going to be as fast as possible and the release time is going to be around 120ms to 150ms. I would specifically choose, FET 1176 compressor here, because of the All-Ratio Mode, which allows the compressor to work in the feedback mode, giving out more color and offer aggressive compression. Works best on drums parallel processing.
And that’s pretty much everything you need to know about how to process your drums/percussion sounds. I hope this has helped in many ways, and I will see you in another blog like this one. Thanks for reading this.