Hello everyone and welcome to this week's article!
Today we're going to talk about a particular way to make a solo cut through, which is used by many musicians: to emphasize only a certain mid range area which is obtained by using a wah in fixed position.
Let's start by reminding what is a wah (click here for a dedicated article with free vst plugins): it's an envelope filter, which means basically an equalization filter that can be fixed or controllable in real time via a pedal control and that according to its settings will emphasize a certain area of the midrange of the guitar tone adding gain and harmonics.
A wah boosts and cuts a certain area of the midrange of a guitar to make it cut more through the mix, and when it's used with the foot control in a tasteful way it can really make the guitar almost talk, or cry like a baby (the famous Dunlop CryBaby is called like that for this reason). This mid range boost, which is usually a rather narrow range between the 300 and the 3000hz makes the sound from dark to very bright and harmonic rich, but the most usual application when doing distorted solos is usually when it's kept around the central position, which means a boost around 1000-1500hz, which makes the tone very honky and recognizable.
This nasal tone is the core of the solos of many famous guitarists, which sometimes just set the wah in a certain way without even moving it, just to have that distincitve mid range, boxy boost when doing solos, and the tone obtained can be heard for example in many Queen, In Flames or Blind Guardian songs, bands very different among themselves which make extensive use of the wah both moving the filter in real time or leaving it in a fixed position according to the song, and this use of the wah also just in a fixed position as mid boost has led to the creation of several stompboxes which replicate just that: a wah with a fixed setting (for example the Dunlop CryBaby Q-Zone or the Magnetic Effects Midphoria).
When trying to recreate this type of effect in the studio when we don't have a wah at hand we can start with a not too steep high pass and low pass filter, between 300 and 3000hz, then we can draw a bell quite wide, around 1200hz and add some db in this area according to taste. Obviously these settings are just a starting point, feel free to play with the filters amplitude and the position of the bell according to the tone you are looking for, and if you feel like you lose a bit too much sparkle try to be less aggressive with the high end roll-off or maybe to add a harmonic exciter.
Hello everyone and welcome to this week's article!
Today we're going to deep dive in a subject already touched in our main equalization article, but explaining more the differences between the various types of equalization and which one to use according to the needs.
- Shelving eq and Filter eq: those are the simplest of all, and can be either with an analog interface (with knobs) or with some graphic representation, but they all recreate the fact that you can choose an eq shape (in this case a filter or a shelf), choose the frequency range where to put it and dial in how many dB to add or subtract to that area (if you are filtering, obviously you're bringing to zero db the sound from a certain point up or down). By concept, for shelf we mean that for example from a certain area (for example 1khz) up (or down) we will start adding or removing gain, and this usually happens with a not too steep curve, to not make the change too unnatural.
- Graphic eq: this is a way to intervene in the frequency areas with more precision, and also this type of eq comes from the analog world. Initially the spectrum was split in just few areas, like 4, then they started adding more and more faders to the units up to 30 bands or more. You can intervene with the individual bands (the more the number of faders, the smaller the frequency range for each band), and the models with more bands obviously allow much more surgical correction, to the point that the most precise ones are used mostly for room correction, meaning to clean up specific resonance areas of a sound. When you have plugin recreation of those old units, usually they try also to recreate the way they used to colour the sound, for example adding some harmonic content or some saturation.
- Parametric eq: the way those equalizer works is an evolution of the shelving ones, basically they work the same way but they let you choose among more shapes (for example a bell shape), decide the Q (which is the width of the shape) and add usually many of those shapes, also making them interact-overlap among each other (this is a specific more of the digital ones, like the Fabfilter pro Q). This type of eq allows much more control and by consequence also much more possibility to make mistakes, but it's one of the few that for example allows you to make a wide boost in a frequency area and then, in that same area create several very narrow cuts to tame the resonances. The most feature-rich digital ones allows also to make corrections only in the part of the sound panned in the middle or only to the one on the sides (MID-SIDE eq) or to affect only one of the 2 sides of the stereo field.
- Dynamic eq: this is much closer to a multiband compressor than to an equalization, meaning that it's dependent on the level of the input. With a dynamic eq you can for example choose an eq bell and decide to reduce the gain on a certain area: the gain reduction in that area will be higher when the gain is higher in that specific area and vice versa. This is used to tame certain frequencies that spike only every now and then like the sibilant consonants in a vocal track.
Note that all these types of eq mentioned are "archetypes", but in the reality of today, especially in the digital domain, you will find equalizers which offers a combination of those functions (for example there can be an eq which can also offer dynamic eq functions, or a graphic eq which offers also a frequency analyzer and mid-side features, and so on).
With the term "sidechain" we define an interdependence between a trigger event (for example a kick hit) and an effect (usually a compressor that lowers the volume of another track for few instants), for instance in a song in which there is a synth pad, every time a kick hits, a sidechain compressor could lower the pad volume creating an artistic "vacuum" effect on the synth, like in the min. 1.43 of the song "Follow me" by Muse.
The "ducking" effect (that's how it's called), it's a creative choice that's functional to the arrangement of the song, but there are other instances in which a mix engineer would simply need to carve a little bit of room and avoid frequency masking in a very dense mix (for example one that is very fast or with many layers) without producing an effect that would take the attention of the listener away from the arrangement.
In this case, more than a compressor that would affect the whole tone, it's better to use an equalizer or a multiband compressor, because in this case one could clear a little bit of space just in the exact area in which we want for example our kick drum to cut the mix more clearly, without touching anything else.
How do we do it? In the Studio One interface (but surely it's very similar also in all the other DAWs) we need to load the eq or the multiband compressor in the insert of the track we want to affect (usually synths or bass guitar, but it could be really anything) and click on the sidechain button on top (as in the picture in this article), then in the track that should trigger the effect (for example the kick track) we click on the "+" button next to "sends" and there we'll see a menu with all the effects with the sidechain function active. From there is sufficient to choose the eq or multiband compressor we've loaded on the other track and every time there will be a sound in this "trigger track", the eq or multiband compressor will activate simultaneously in the other, and in this case it will lower the eq in the other track, in the area of our choice, which should be the same that we want to emphasize in our "trigger" track; for example if the kick track we are using is covered by the bass in the low end, for example in the 100hz area, we can sidechain an eq that every time the kick hits, it will lower 2 or 3 db exactly in that area of the bass track.
This way the kick will be more prominent, but without us boosting it and without ruining the balance of the song.
Besides the eq-based effects, it's definitely worth mentioning the reverb, one of the most important effects of all: it's an effect that mimics the bouncing of a sound in various ambients, which can be from small rooms to huge caves, and it's essential to add smoothness and realism to any tone and to make it less harsh; click here for an in-depth article about reverbs.
Moving towards the modulation effects, let's first start with the definition: they are filters that take a given signal and create a copy, with a given delay and pitch modification, to be summed with the original one, creating a wide range of different results.
Said this way it's a very wide definition, but modulations are the core (together with the reverb) of sound effects, and they are many and capable of obtaining very different result.
Let's start with the chorus: the chorus is a type of modulation thatdoubles the sound creating a slighly delayed (usually around 20ms) copy whose delay will not be stable but will keep on variating, oscillating 5ms more and less, plus the copy's pitch is slighly detuned, giving the impression that the copy is (in case of a Vocal track) another person singing along with the first one: not identical, but very similiar, and this effect is used to make the original sound wider and fatter.
Moving to other types of modulation, other two types which are quite important are flanger and phaser.
These 2 effects starts from the same idea: to split the signal in 2 copies and putting one of the 2 rhytmically out of phase with the other. The fact that this phase changes all the time following a certain tempo creates a very "sci-fi" effect, and the flanger takes this effect one step further by delaying the second sound copy and moving it back and forth in time, creating an even stronger phasing effect.
Hello and welcome to this week's article! Today we're starting a new serie called "How to improve", in which we will add some element to our basic mixing articles. This time we're talking about 5 ways to improve our amp sim guitar tone, which even if we are using some high end plugin, still needs some polish before it can perfectly sit in the mix. These tips are to be applied after recording, editing, finding the right guitar amp sim and tweaking it.
1) Impulse responses: impulse responses are the single most revolutionary thing happened to the world of guitar tones in decades, since they can change radically a sound and make it realistic in a way that few years ago was unthinkable without microphoning a real amp. Get yourself a nice collection of impulses, at least a couple of hundreds, it doesn't matter whether free or paid, just make sure to have enough variety to cycle through them until you find the one (or the combination of more than one) that fits perfectly the song, and in order to do it, use a referencing track with a sound similar to what you have in mind. A good starting point would be our free IR Pack Trident, which has 14 impulses ready to use.
2) Create a guitar bus: route all the guitar tracks that should have similar tones (e.g. the rhythm guitar tracks) into a stereo guitar bus, in order to control with one fader the general presence of guitars in the song.
3) Compress the minimum necessary: the general rule here is "the more the gain = the more the natural compression", so if we have clean guitars we will need to compress more, while if we have high gain ones, we're going just to use a multiband compressor in the low end area, shaving off a couple of db when needed in the area from 80 to 350hz (usually during palm muting).
4) Equalize with wisdom: here the general rule is "boost wide, cut narrow". We could spend hours talking about how to clean a guitar tone, but in this article let's just add the fact that we need to find balance: the sound must be natural, similar as if it's coming out of a real amp, so: - high pass up to 50 to 80hz - if the guitar still sounds weak and lacks of presence, add some body in the area between 1000 and 2000hz, finding where it suits, with a wide bell boost of 1 or 2 db. - cut a couple of db with a narrow bell in the 4k area, sweeping the eq until you find the "nails on a fingerboard area" - add some air, with a wide bell around 8-10khz
5) Don't be afraid of saturation: this is another tool to be used if our tone (especially if clean or overdriven) is too weak, thin, and we can't solve by using the eq; a touch of tape or tube saturation can add the weight and the midrange our tone needs without altering too much the sound. This tool can be used on other instruments too!
This article will only talk about the distorted, hi-gain guitar and the hi-gain part of a bass track (if you are using a dual or more tracks technique), and starts from the assumption that a distorted track has by nature a good part of noise and resonances, and we need, from that noise, to minimize the bad part and emphasize the useful one.
How do we tell the good part from the bad part? By nature a distorted track adds a big amount of gain and saturation in all frequencies, therefore along with an important increase of level for what is the original sound of our instrument we end up with a lot of "dirt", which eventually might end up by covering the other instruments, or producing unwanted peaks that prevents us from rising the overall volume.
By finding and notching out those peaks and resonances we will free a considerable amount of headroom (very hard to point out at first listen, unless you have a trained ear) that will let us raise the volume of the track, making it more present and at the same time it will sit better in the mix. Sometimes, in facts, we might feel our guitars weak, low in volume, not very present, but if we rise the volume we will have it reaching the maximum level and clipping while it still sounds weak and low; that's because some frequency that we might even not notice at first is so high in volume that hits the ceiling before all the rest, thus keeping all the rest of the curve low.
How do we find these resonances to notch out? The nature and the modern technology comes in our aid, with our ear (here's an ear training tutorial) and with frequency analyzers (here's a dedicated article). Obviously frequency analyzers are importants because we can see right away if there are resonances and take them down some db until they are aligned with the rest of the curve, but the thing that makes the difference between a regular mix engineer and a good one is by using the ear to clean up the track.
It's very simple to do once you know how to do it: you create with an eq a narrow notch and boost it considerably, around 6/10db, then you start sweeping left and right through your distorted guitar and bass sound. You will hear every kind of weirdness but what we are looking here are the frequency areas that does not resonate with the note that the instrument is playing in that moment (for example in a guitar riff with a sequence of notes), leave the parts in which you can hear strongly the note played and focus on the parts in which there is a noise, a weird sound, that remains the same regardless from the note played. Those are the frequencies we are looking for, and once we isolate them, we must take them down 2 to 4db, until they are not so prominent anymore.
Think about those frequencies as junk to remove: what we are aiming here is to notch out 2 or 3 of these areas, quite surgically, in order to make room to the part of the sound that will be actually audible in the mix: the ones that comes with the note; the others are useless and eat up only headroom. I'm not writing specific frequencies here because they vary according to an infinity of factors: guitar, pickups, string gauge, amplifier, and most importantly, the tuning.
Hello everyone and welcome to this week's article!
We are starting the new year with a challenge that will teach us how to separate the various parts in the mix, carving the right place for each instrument.
The challenge consists in this: to mix a project, even better if a project with all microphoned parts (for example a microphoned drumset, microphoned guitar amps and so on).
Obviously we're not talking about a challenge with a prize, but an educational one: to dig into the root of the sound, and draw out a mix as natural and unmodified as possible, just avoiding the various parts to clash together (or to cover each other) without all the layers of effects, editing, autotune, quantization. The ideal would be to mix a live song, and luckily it is sufficient to google "live multitrack project" to find some free one, available for educational purposes.
For this challenge we just need to rise in each track the high pass filter until the sound starts to become thin and to lose its body, and then stop and take it a little back, then to use a low pass filter until the track starts sounding dark.
This way we will achieve separation, and we'll be forced to choose what element to leave in which area.
We could start from the picture on top.
I've drawn on this picture the starting points: we can start by filtering more or less the tracks bringing them on these areas, then from there it's all a matter of fine tuning, which changes from song to song, and from genre to genre.
Here is the website of Christian Budde, which offers a selection of free equalizers and high pass and low pass filters, among which Rubberfilter, but any eq/filter which lets us dial the frequency is good!
I'd be really happy to hear your mix just using a filter, if you want, share it with us!
Hello everyone and welcome to this week's article!
Today we're going to talk about a topic related to psychoacoustics: if we eq or compress a sound making it louder, somehow, regardless if it's actually better or not, it will probably sound better to us.
That's why people usually starts mixing at -15db, and once the sounds of all instruments are done and balanced, the peak in the mix buss is -5db: because we tweak one track to make it sound better and it ends up being louder, and then to compensate we rise the volume on the others, then we pass on the next one and make it louder, and then rise all the others... and so on... and then eventually before mastering we turn the master volume down when exporting like real cheapos.
The experienced producers can start with a chosen project level (for example -12db) and arrive at the end of the mixing phase without touching the master fader with the maximum peak still at -12db, because they did level matching when mixing, and they have kept the volumes of the single tracks at the same level of the balancing phase, thus preserving all the headroom of the project for the mastering phase.
Why not to just turn down the master fader at the end, e.g. 10db down to create headroom for the mastering phase? Because this way we will just raise the noise level 10db in our tracks, since the noise point is always there, in the same place, therefore if we sink the whole project 10db lower, it will eat up 10db of signal, the final result will sound dirtier and data will be lost for no reason.
What to do to keep the level of the tracks stable?
Match the in-out level when mixing or mastering, both when using Equalization and Compression.
While we have seen already how to do with compression, we haven't yet talked about the equalization; on an equalizer like the Pro Tools one (as you can see from the picture on top) you can monitor the level of input and output signal, which means that for every db of boost that we apply on the sound we can lower the overall output knob of the plugin in order to match the same level of input: this way we will avoid the process of raising and raising the volume of all the tracks that I have already mentioned.
If we don't have the input and output level in the plugin ui, we must use the track metering tool: we must see how loud is the signal with the plugin bypassed and compare it to how loud it is with the eq on: if with the equalizer on the signal is louder, we should lower the output of the plugin until it matches the loudness of the bypassed one.
Once the tone balancing phase and the tone shaping phase with compressors and equalizers are done, in theory the additional processors that we can add to our mix (like modulation effects) shouldn't impact much on the level of the single tracks, therefore we should be quite safe not to raise too much the volume of our single tracks (watch out for the group tracks and the mix buss though!).
Merry Christmas and Happy new Year from Guitar Nerding Blog!
Hello and welcome to this week's article! Today we are going to take a look at 5 ways to achieve a better mix separation, and this article will be particularly useful to those who feel when mixing that they can't separate exactly the various instruments, letting them mask one another and creating muddiness.
1) Choose carefully what to leave in the top end area: especially in crowded mixes (those with a lot of high pitched instruments like cymbals, shakers, high vocals etc), having a high end area full of eq masking and conflicts can lead very easily to a bad mix. We don't want our mix to be confused, so we must keep it clean by using a low pass filter on almost every instrument to free up frequency space for the chosen 2 or 3 elements that we can leave to roam in this area. This way the selected elements of the mix will be much easier to understand and in the overall sound will result less harsh.
2) Achieving separation in the low end area: similarly to the high end area, we cannot leave each element of the mix to roam free in the lower area, therefore we will have to high pass almost every track to move them further in the mix and again avoid eq masking: we should leave in this area mainly bass and kick drum, and anyway also the kick drum doesn't need to go too much low, where the bass instead could. Definitely guitars should free up some space here.
3) The most sensitive frequences to the human ear: Top end and low end are crowded, but no area of the frequency spectrum is crowded as the most sensitive one to the human ear, the one around 2000hz. This area is the one in which we should find 2 or 3 elements that will be in the front, and lower or carve the eq of the other elements of the mix. Usually the elements of the mix that must stay in front are vocals, snare and kick, but this rule can change from genre to genre.
4) Route everything in groups/busses: creating groups of tracks is good to process them together, for example routing all the vocal tracks in one buss, all the drum skins in another one, all the rhythm guitars in a third one and so on. The purpose is yes to free up computer resources, but also to give a sense of homogeneity to all the tracks inside a group processing them together, and to separate them better: it is easier to have only 5 or 6 groups to manage with eq and compression when finding the right place to every mix part, rather than 30 single tracks.
5) Don't boost too hard: if you need any boost harder than +6db you must or reconsider your starting sound, or free up place among the other instruments in that area: when you boost a sound the computer creates a digital reconstruction of what it imagines that sound would be if it was louder in that area, but the more you boost the farthest it gets from reality and it becomes twisted and unnatural. The ideal boosts should be 2-3db if we want to obtain a pleasant, natural sound that is representative of the source sound.
I know, I have said in the past that in theory the mix should be perfect and arrive to the mastering phase with such a perfect balance to the point that there will not be need to eq during mastering, because at that point one should correct the mix problems, rather than actually enhancing the sound. If the balance needs to be enhanced, it's better to go back to the mixing phase and make the corrections in the single tracks.
What I meant was that an eq in the mastering buss affects the whole track, therefore we cannot for example tweak the mids out of the guitar without affecting the body of the snare or the vocals, and so on. It is really a castle of cards. A solution in these cases comes with the mid side equalization (click here for an in depth article, with a list of free plugins): an equalizer that lets you choose what part of the sound process, the center of the sides. This way you can work on the bass frequences, like bass or kick drum, tweaking the bass as you normally would, and filter the bass freq out of the sides (for example from 100hz down), where guitars and drum overhead lays, and here we can also add some sparkle in the high end (obviously, since we are talking about mastering eq these correction must be subtle, not more than a couple of db): this way the bass will sound more focused in the mix, while the sides will sound more bright and open, all without touching the precious balance in the mids and upper mids of the center of the mix (vocals and snare).
Hello everyone and welcome to this week's article! Today we will say something about the history of Equalization, in a way similar to the one we did for Compression.
An equalizer, as we have already seen, is a tool that allows us to boost or cut a certain frequency area. In the case of a Graphic equalizer you have the frequency range divided in "slices" and you can raise or lower the single band, and this system is used usually for the whole track.
A parametric equalizer instead lets you decide with one knob the frequency area to cut or boost, with another knob the amount of db to add or subtract, and with a last knob how narrow the cut or boost should be.
This kind of eq is used both for mixing the single tracks, and for mastering.
The story of equalization starts around 1920, and it was made mainly to process the voice of a radio host; equalizers were initially tube driven, then, the more mainstream they became, the more the analog version became popular, to the point that around 60s and 70s every mixing console had a separate analog parametric eq section for each track.
Around the '90s the rack eq became back popular, and today we can choose among tube driven, analog and digital eq, in passive (so just made to filter out frequences) or active version.
All this different versions in the market are made to produce different results: a tube eq will give to the mix a certain amount of harmonic enhancement, an analog eq will have a certain sonic property that can be heard in almost every record of the 70s (a very subtle saturation and a certain eq cut), while a digital eq has the undeniable advatage to let us sometimes see graphically (with a spectrum analizer).
The spectrum analizer has been a revolutionary tool: it lets you see the actual wave and to operate surgically in the specific frequences with great precision, letting you using not only your ears, but also your eyes.
Today we have Vst emulations of all kinds of hardware eq, and especially lately we are starting to see again the arrival in the market (both free and paid) of vintage parametric equalizers, with pieces of software that tries to recreate not only the way those processors used to work, but also the collateral properties, as the saturation, for example.
Usually parametric equalizers have one gain knob that at unity position leaves the signal unaltered, and if you turn it up or down it adds or subtracts dbs in the assigned frequency area, but there is a very particular hardware processor, called Pultec EQP-1A that works differently (see picture above).
This equalizer, lauched originally in 1951, is one of the most used of all times in the music industry, and its latest incarnation is still on sale today.
The original Pultec is a tube eq which has the particularity to have for each one of the 2 bands, 2 gain knob: one for the amount of boost to apply and one for the amount of attenuation. Said this way it sounds like those two knobs cancels out each other, but the truth is that the two knobs which works in the same frequency range doesn't actually affect exactly the same identical frequences: the attenuation in facts happens in a slighly higher area than the boost, and this makes possible to achieve an attenuation that has for example a boost in the area before the slope starts. Is it also possible to control the Q, which is how narrow is the boost or cut.
Why to choose today a Vst that emulates an old piece of hardware with very limited controls like the Pultec instead of using a very flexible new Vst equalizer, with a lot of functions and a built in spectrum analizer? Because often, if the programmers are good, it recreates the coloring that this device used to give to the tracks, and because it forces you also to use softer settings, which leads to less drastic and surgical changes in the sound, leaving a more natural result which is proper of the '60s and '70s records. My suggestion is to give it a shot and decide with your ears.
Here's 2 versions, one free and one paid of the Pultec EQP-1A Vst emulations:
Hello and welcome to this week's article! Today we're going to focus on a topic that we have mentioned in many occasions, but that we never really explained yet: what does the term Linear Phase Eq and Regular Eq mean.
Let's start with this assumption: Regular EQ’s have phase shift between the different bands. When we boost or cut a certain area, that frequency range is technically delayed by a tiny amount, due to the way the equalizer itself works.
Linear-phase EQ, instead, compensates this delay keeping all the frequencies 100% in phase (it basically anticipates the wave of a certain time to correct the phase shift). In other words, all the frequencies pass through the EQ at the same speed, resulting in zero phase issues between bands.
The basic question is: "if they have created a way to prevent the phase to shift, why don't everyone uses just linear phase eq?"
the answer is because it consumes more cpu resources and, especially, because it creates a "pre-ring", which means that by correcting the phase variation it creates a part of the wave that actually starts BEFORE the peak of the transient, and this, for certain types of sounds as for example percussive drums, it can ruin them.
The conclusion is that linear phase eq is best suited for certain instruments in which this problem is less apparent (like acoustic guitars) and less for others, and we must try it and compare it with the regular one in order to understand on which instrument it brings a benefit and on which one not.
About the mastering phase, finally, there are different schools of thought, for example many producers likes the result obtained with the Waves Lin-Mb, a linear phase multiband compressor, because they think linear phase sounds more transparent, while other engineers think that the "pre-ring" downside can ruin the mastering, especially if we are speaking of a heavily percussive track.
Hello and welcome to this week's article! Today we are going to talk about de-essers! A deesser is basically a compressor which is set to operate only on a very narrow range of frequences that corresponds to the sibilants of the human voice, which can be captured with high peaks of volume by the microphones. Usually these sibilant "S" and "T" sounds are located in an area between the 9khz and the 15khz, and often the deessers have presets for male or female vocals, but the best ones sometimes features also a "learn" mode in which they automatically detects the right area to attenuate: the more the area is narrow, the more the deesser will be effective without processing unneeded areas of the mix.
There are also other, more creative ways to use a deesser: someone uses them on a cymbals track or on a distorted guitar track to tame the harsh frequences that sometimes can get a bit too prominent, giving a more smooth result, similar to the one you'd get from a Low-pass filter, but less invasive.
There are many Deessers on the market, and almost every DAW has one bundled, but if you want to try some free Vst, here are best ones the web has to offer:
Hello and welcome to this week's article! Today we're going to talk about my to-go Eq Vst plugin, the Fabfilter Pro Q 2, which is basically nowadays the only Equalizer I use, both in the Mixing and in the Mastering phase.
Why do I think this Eq is better than the others? Apart from the fact that it sounds good and it doesn't colour my source sounds too much (but this is a characteristic that many good quality Vsts have, for example some of the Waves ones), I love this Eq and I use it on every project because of its unique tools, that makes my job so much easier:
- Built in Spectrum Analizer: it lets you see the original curve and the Post Eq one, even at the same time, to see if there are peaks or resonances to tame. I don't know of any other Equalizer that gives that much flexibility, so far I had to use at least a Spectrum analizer on each chain, and it has been a pleasure to take 'em out.
- The fact that if you're working on a stereo track (i.e. Mastering), you can have each band to work on the whole track or just on the left or right channel.
- It has an Eq Matching function, much like the Voxengo Curve Eq: you can make it analyze another track via the Sidechain Input, and it will apply the same curve to yours.
- Solo mode: this function allows you to isolate just one "slice" of the spectrum and hear it. I find this function, which is quite rare, to be the most useful feature in the whole plugin, and it works by clicking and holding down the headophone icon that appears whenever you highlight a single band. You can move left and right to move the "slice" of the curve you're listening to, and use the mouse wheel to make the audible part more wide or more narrow. This is the best and most precise way to point out a single area of the sound to intervene.
All in all, this plugin makes my tracks sound better and my signal chain shorter, which is very important since the more processors you need to stack on a track to make it sound good, the more problems you will have, and ultimately the more "unnatural" the final result will sound (plus the project will be heavier on the cpu).
Hello and welcome to this week's article! This article links our in-depth Mastering article, which covers pretty much everything that can be done in the mastering phase, whith our "Mixing with stock plugins" article, which shows us how to mix a song only using the basic stock plugins featured in our daw.
The idea is to try to separate what's essential when mixing and mastering from what's just "the icing on the cake", and to focus on the essential things with just the tools provided in our Daw. The second objective is to free ourself as much as we can from the slavery of the presets: we must understand how does a tool works and how to adapt it to our song; the presets that we can find on the plugins are always a caricature, made to let us understand the direction of the processing, but they're always exaggerated. The sooner we'll stop using presets without knowing what the hell is going on, the better our songs will sound, trust me.
Speaking of the essential mastering chain, once we have a good mix we just need to give our tracks the final boost in volume and the final touches to make them sound in line with the commercial products.
Let's start with a compressor. This is needed to reduce the dynamic range a bit, letting us push more with the limiter. We need to check out the average level of our song (for example -10db) and set the threshold of our compressor a bit higher (for example -8db), so that the compressor only lowers the big peaks. The Ratio should be somewhere between 2:1 and 8:1, we need to limit the peaks but not taking away the life from our songs so find the right compromise. The attack and release should be set to a speed that lets the compressor to kick in and go back to inactivity, we don't want to compress everything all the time (ideally, the meter should move in time with the song).
Now we need to use an Equalizer. In theory, in the mixing phase we should have already done all the equalization we need, and if there are problems it would be better to re open the mix and to fix it from there, because now we're affecting the whole mix, so we must be very careful. A good thing to start with the mastering eq is to take away everything from 45/50hz with a high pass filter, then we can check out the most problematic areas in the mastering phase, which are usually the lower frequences: sometimes our master needs a little boost in that area, sometimes a little cut, sometimes nothing (and we often can notice it only using a reference track), but probably the best way to eq a track during mastering is using a Mid/Side processor, in the way described HERE.
Now that we have tamed the most problematic frequences we can try to add a little sparkle to our sound, with some harmonic exciter or saturation tool. The ideal would be to have a multiband processor, since in my opinion we should excite a bit more the higher frequences and much less the lower ones (or just bypass them), otherwise the delicate equilibrium we've tried not to ruin until now will be damaged. These tools are usually needed to add some nice ringing to the snare and cymbals, and to add some bite to vocals and guitars, but if we see that we cannot obtain a sensible enhancement, we can just skip this whole step and pass to the limiters.
The limiter is the single most important part of the mastering chain, as we have already seen in many other articles of this blog. What we need to do is to see how much headroom we have and to raise the final volume of our song to a point that it is competitive with the volume of the commercial songs, but at the same time it must not sound excessively squashed and distorted. Our job here is to find the right compromise between power and clarity, without cutting too much the transients. We should set the ceiling somewhere between -0.1db (if our final track will be played mostly from a cd) and -1.0db (if we are planning to distribute it mainly through internet streaming) and to lower our threshold until we see some gain reduction: we should let the limiter kick in only with the highest peaks, and never let it limit more than 3 or 4dbs, otherwise we will damage our song.
In order to know if the overall level of the song is in line with the industry standards and if there is enough headroom before exporting, we could use some metering tool. Those tools just tell us visually if the song volume is right, or if we could raise it or lower it, and some tool like the TT Metering tool also tell us if the song is too compressed, so that we can limit it less and make it less ear fatiguing. Once all the meters says that the levels are right and there are no distortions, we can consider our mastering done!
Hope this was helpful! By the way, the song on the video is a song taken by Wisteria's full lenght "8-Bit Nightmare", played and mixed by me. Buy one copy to support us by clicking on the banner on the right side of this blog!
Hello and welcome to this week's article! We have finally approached the end of the year and to the end of this long tutorial on the many ways to mix a rock / metal drumset! The snare is the focus of the album, along with vocals, and the equilibrium among vocals, snare and kick it's the axis of the mix, the one that sets the mood of the song (higher kick and snare and lower vocals for death metal, higher vocals and lower drums for soft rock): long story short, we need to get the volume and the tone RIGHT, otherwise the whole album will sound cheesy.
If we have just the snare top track, we can start by using a gate, to take out at least part of the cymbal bleed (not too strong, or we'll take away also some of the good harmonics of the snare), then a hi pass filter up to around 70hz and finally a low pass one down to 12khz, to cut away some of the left cymbal bleed. Now we need to take away the low-mid "mud", which lies among 200 and 500hz: just find the right frequences and take them slighly down, without taking away too much body from the sound (you can use a frequency analyzer to locate exactly where most of the energy lies). Now if the snare needs some low-end boost we can lift with a medium-width boost the area around 125hz, but beware because we risk to end up by adding again the "mud" we have just taken out. The last and trickiest thing to do to enhance the tone with an equalizer it's to boost the highs: if we feel that the snare sounds too dark we can try boosting somewhere between 2'000 and 6'000hz: this is the most delicate moment of all, because now we are defining the colour of the tone that will hit the ear of the listener first, so we must not overdo and be aware that the higher the frequences that we're boosting are, the louder will be the cymbal bleed.
Once we have found the right tone for our snare top track we can compress it at a ratio up to 8:1 (we must choose the ratio according to the dynamic range of the track: if we have many low energy snare parts like press rolls, flams etc, we need more compression or they will disappear in the mix, while if there are not and the strenght of the hits is consistent, we can lower the ratio). The attack should be around 10ms, and as always we should set the release at a level that the compression can go back to zero or almost, between a hit and the other. If we feel that by compressing we have lost part of the transient we can use a Transient Shaper to bring back some of the snap of the snare, usually avoiding cymbal bleed, but don't overdo because too much transient enhancement can make the snare too much "in your face" and unpleasant: we must always look for that euphonic "sweet spot" :)
If we need to transform the snare sound too much to get a decent tone, then we have a problem: an excessively processed snare it's a snare that sounds fake, therefore it would be a good idea to record it again. If we can't record again and we need a way to add some punch on the snare we can mix one or more samples with the original sound, and if we're good, we will find a sample that compensates the lack of frequences of our original sound.
Another way is to Double the snare track: we can process the first one like explained above, and use the second one just to add the "snap". We can gate the sound until we have only the snare snap, then we compress it very heavily, use some frequency exciter and eq to make the "snap" very strong, similar to a fingersnap, and then we can mix this sound with the regular snare to get the same effect we would obtain by using a sample, with the difference that this way the sound comes all from our drumset.
Finally, if we have recorded the snare with 2 microphones, top and bottom, we're gonna have a second track recorded pointing to the snare wires in order to give some sizzle to complement the top snare tone. This track can be also very useful to add some body to the top snare tone: instead of boosting around 125hz on the top track, we can boost the same area in the snare bottom track and see if it sounds better. About the hi pass filter, we can take out a bit more, even up to 90hz if we want, the important thing is to take a look at the phase: we must try inverting the phase to see the version that produces less frequency canceling, and use that one.
On the compression side, if we use more than one snare track (top + sample, top + bottom, top + top copy), my suggestion is to route them into a group track and compress that one, with the same settings suggested for the snare top. If we need it, we can also set a clipper at the end of the snare chain: this is sometimes useful because, during the mastering phase, the limiting traditionally takes out part of the snare transient, but if we use a clipper (with very soft settings) we can recover part of that lost transient (it's not easy to explain how a clipped sound helps recovering its transient while it should actually work exactly at the opposite, but it works: the software recreates part of the transient). It's interesting to notice that unlike for other instruments (e.g. hi gain rhythm guitar), in which the compressor should be as transparent as possible and the moment you notice it, it means that is damaging your sound, compression for the snare is a real tone shaping tool, and if used without exceeding it can bring up the wires of the snare bottom, and enhance the sound with an explosivity that cannot be obtained otherwise.
About the Reverb: if our drumset sounds a bit dry (or we like a big '80s style snare) we should create an fx track with the best reverb that we have and set it in order to make it sound similar to a studio room. Then we must equalize it with a high pass filter that takes out everything up to around 200hz, and send it to our snare track (or group), to our tom group and to our cymbalgroup track, from there we can decide the amount of effect sent to each track according to our taste.
I hope that this long drum mixing tutorial was helpful, and by the way the song is mixed and mastered by me for my band, Strider. Contact us if you want a copy of our latest Ep "The Black Lotus", also to support my work with this blog! Happy New Year from Guitar Nerding Blog! CLICK HERE TO READ PART 1/4
Let's start by saying that there are different ways to work a kick effectively, according on if we want to preserve the acoustic sound or, at least, part of it:
Assuming that we have tracked a good sound, we must first do a high pass filter, cutting everything below about 35hz, then we can scoop an area that may vary from 150hz to 400hz: this is where the "mud" lies, but beware because if we take out too much energy from here, the kick will lose its body, especially on the cheapest speakers. Once we have tamed the frequences where most of the energy lies, we can boost if we need some more "thump" in the 50hz-80hz area, and some high end between 2'000hz and 10'000hz, depending on the type of coloring that we want to give to our kick. On the compression side we can compress at a ratio around 8:1, with a fast attack and a release speed that should be set according to the song's speed. Just remember that a too hard compression can ruin the sound's transient, so if we need to apply so much compression that the sound becomes dull, it's better to stack 2 compressors with lower settings, and/or try to reconstruct the lost transient with a transient shaper.
If we have a sub kick microphone too, we can consider our kick sound splitted in 2 tracks, one for the lower frequences, one (with a high pass filter at around 500hz) for the higher ones: this way we can compress the low end more and the higher end less (thus preserving the transient), and maybe do some sidechain compression with the bass, so that the bass sound is slighly compressed when the sub kick track kicks in,which may be crucial when dealing with death metal speeds, to preserve clarity and separation.
Even if we are tracking our drums just with one microphone, we can always double the track and do the same as if we had 2 microphones: the result will be less complete, but always more flexible than dealing with one single track.
Finally, we can obviously use a sample: usually pre-processed samples tends to be oriented to the high frequences, so when we blend them with our acoustic sound, it's often a good idea to treat the acoustic track as the "sub kick one", and use it to emphasize the lower region.
