The full version of this tutorial, along with many others is available in the Systematic Mixing Guide, available here: http://www.systematicproductions.com/mixing-guide.htm
-----------------------------
Originally, I had hoped to sequence these tutorials in the same way I normally progress through mixes. That is, to start with the drums, move through to bass, then guitars, then finally vocals. The reason that guitars got bumped up the ladder is simply because I'm mixing them right now, and these ideas are all very much fresh in my head. So, don't get a skewed perspective from this - the sequence of tutorials has no greater implication than mirroring whatever is fresh and relevant at that point in my own head.
As a mixing engineer, you need to develop an intense hatred of distorted guitars. They are your natural enemy. Your arch nemesis. Heck, they're your damned Van Helsing.
No other element you run into when mixing will want to, in its raw state, eat up so much frequency real estate. Guitars chew through vital midrange frequencies, they resonate over the top of bass guitars, and their very high frequencies sound extremely harsh and unmusical.
You can think of mixing guitars as a street fight. The weaker your opponent, the more easily you can subdue them. However, the bastard always enters with a glass bottle, and no matter how you go about it, you're still going to come off with some injuries. How substantial they potentially are will depend on three things:
#1 How weak the opponent enters the fight (how balanced your raws are)
#2: How cleanly you're able to subdue them (how well you can mix them into your arrangement)
#3: Whose turf you're on (how well balanced your mix is to accommodate them)
Since this is a mixing tutorial, we will already assume that you've started with a great player, who has recorded with a great instrument, which has been captured with high quality gear. If any of this isn't the case, you need to give up your search for 'perfection' now, because it won't happen. The most you can hope for is to salvage what you can using the following mix techniques, or wait for an opportunity where the aforementioned is the case.
The whole process of recording, and then subsequently mixing guitars, is somewhat contradictory and ironic, as we'll find out later. Your primary goal when tracking guitars - let's assume for a moment we actually care, and are using real tube amps with a real mic or two - is to get as large a sound as you can. What you want for most projects is something with a fine, even midrange balance. Anything out of the ordinary, such as excessive tight peaks (fizz or midrange resonance nodes) will immediately play havoc with the clarity of the mix. So you're looking for a finely appropriated midrange as your #1 requirement. The midrange is where guitars live on records, after all.
The second element is a balanced low-end. Guitars are naturally adversarial here, as they will have cab resonances that commonly lie in areas that excite nasty room modes in smaller spaces. One of the reasons that oversize 4x12 cabs are so popular for distorted guitar sounds is that they shift these resonances lower down the spectrum, where they are perceived as more musical and 'punchy', rather than outright muddy. If you're recording in a small room, then you need to rethink your circumstances. You may well be able to get a good midrange if you isolate your microphone and cab enough from the surrounding environment, but your low-end will never be clear. No matter the amount of treatment you put into that space, there will always be nodes in the low-end that get excited and cause the vital bass/guitar crossover region of your mix to be an absolute nightmare.
Here is where we get into the contradictory stuff.
So, you've loaded your raw guitars into your mix. What alone sounded quite large and nice is now completely clouding your bass guitar, vocals, and generally making the mix sound like a mess, right? Perfect; this is exactly where we want to be. The whole idea behind tracking a 'large' guitar sound, was that so you could downsize it in mixing. This is where the poking of holes begins.
Think subtractively with guitars. For the last couple of years I have not increased the gain on a single frequency band on guitars at all. Picture the ideal tone that's hiding within that mess, obscured by all that junk. Ultimately all you can do in mixing with guitars is to reveal that inherent character in its pristine form. You do this by subtracting the junk. If the raw sound is bad, then your final sound will also be bad. Guitars are notoriously poor at taking processing. Second perhaps only to vocals in this area, they absolutely need to have a good character when raw, otherwise this entire exercise is pointless.
The main issues you will find with guitars is how much parking space they want. They're like that fat asshole with the hummer who gets two parking spots, and gets away with it simply because he knows he's the CEO's cousin. Likewise guitars know they are vital to driving the energy in a rock or metal mix, and take up a whole heap of real estate. They will likely be conflicting with many major, vital elements of your mix. Your process begins by letting these other elements through.
Start with filtering.
High and low passes are commonly the first stage in the sequence of processing done to guitars. This is because the filters reject the most useless of frequencies. Everything that is entirely unmusical and messy can be rejected significantly with these.
The points where you set these filters, and their orders, are entirely dependent upon the mix you're creating. A common starting point is 3rd order (18 dB/oct) filters set at 70hz and 12kHz. Bear in mind that you can be significantly more aggressive than this. It's simply a starting point from where you can jog the filters around until the guitars start to sound the most correct in context with the mix. My personal extremes lie with a HPF at 130hz and LPF at 7.1kHz. Sometimes you just need a tone that fits within a very dense arrangement, and you need to reduce its bandwidth appropriately. Don't be afraid of doing this, if necessary. Get extreme if you need to.
Move onto basic EQ
Think of this as the sculpting portion of the process.
The idea is to do whatever you need to, in order for the guitars to take on a character that's more pleasing toward the mix balance. It's immensely difficult to talk about this in general terms, as each guitar sound is intrinsically unique.
Always remember, guitars take processing very poorly. So get to where you need to go in as few moves as humanly possible. Always have that in the back of your mind, reigning you in. Do more with less.
Referencing guitar sounds can be a dangerous affair. You might well be able to take on the same overall shape and balance of that tone you're referencing to, but in the process you will lose what's unique and powerful about your own tone. It's always a better idea to uncover the potential within your sound on its own terms, before referencing. Reference only for overall balance, if possible - not for the finest of little things. If you are having trouble hearing the rest of your mix in specific areas, then referencing may uncover where to look, but beyond that you should still treat your sound as its own entity.
Common problem zones to watch
Guitars are essentially shaped broadband noise. I like to use them as a starting point for frequency training, because they contain so many frequencies in such abundance. It's because of this specific reason that you will find yourself EQing them all over the spectrum, and will find them causing problems for your mix all over the spectrum.
Take a look at the low-end first. Chances are there will be a peak, or a few of them, anywhere from 100hz all the way through to 180hz. Smaller rooms may even induce ringing and resonance higher up, between 180 and 300hz. Wherever you find these peaks, suck them out! They are exceptionally detrimental to the clarity and tone of your bass guitar. Having some minor 'thump' to the guitars between 70 and 90hz can be beneficial in creating movement and size, but tread carefully. Always remember that the bass guitar has a role - and that is to create tight, expansive low-end which in turn creates the illusion of the guitars being bigger.
After this, consider looking at the low-mid frequencies. Many smaller rooms, or mid-focused amps will leave some junk down here. Usually a wide scoop of sorts can be in order to clean them up. We're primarily talking about 200 to 500hz here.
The next is one of my favourites. I call it the 'cardboard zone'. It essentially ranges from 600hz to 800hz (but on some amps can even be centered as low as 500hz). This is the zone that contributes to what you hear as a 'flat' character on guitars. Generally subduing this range on most raw guitar sounds will lead to an immediate increase in air and dimension for the entire mix.
The next zone is absolutely vital. It commonly ranges between 2kHz and 4kHz, and many times is centered close to 3kHz. This is the vital vocal presence range. This is an area that you need to balance with utmost precision if you plan to have both guitars and vocals intelligible at the same time. There are no specifics to be given here, as it really depends on the excursion of the amp. There will generally be something undesirable going on in this region with a raw amp tone, however, and it's your job to get it under control so that vocals, the most vital point of your entire mix, can poke through unobstructed.
Beyond this area lie mostly isolated fizz spikes. This is extremely amp/cab/speaker-dependent and no specifics can be given. Feel free to sweep a tight EQ boost beyond this region. Listen to hear if anything stands out to you as particularly unpleasant or detrimental to the mix. You can sometimes end up with a few to several tight EQ cuts in the highs to tame fizz. Each cut will bring increased clarity to your mix, but decreased clarity to the guitars. This is the tight rope that you are walking all the time with guitars.
Consider multiband compression
Taking us back to the lows and low-mids, we find many times that sucking out too much of the resonance down here makes the guitars very thin and powerless. This is largely because there are so many non-linear peaks happening in this region that we'd need multiple EQs to get them all controlled in proper proportion - after which there'd be no tone left at all from all the phase shift crazyness.
So what certain crafty engineers have pioneered is the use of multi-band compression on the mid-bass (and some low-mid) frequencies of the guitar. This allows you to tame only resonances that have crossed a certain threshold, and keep the rest of the low-end intact. This is a very good solution for particular cab and room resonances that tend to get excited by particular notes. It keeps the rest of the low-end relatively untouched and allows the guitars to stay meaty as well as controlled.
The fabled Sneap C4 preset is a great starting point here. I personally run the multiband compressor between 70hz and 250hz a lot of the time. Possibly higher if the room was small and the resonances extend higher. I also keep a shorter attack time than the Sneap default, because I like to have my bass guitar control more of the low-end, and don't necessarily want to let much cab resonance through.
Consider broadband compression?
Broadband compression on distorted guitars is an interesting thing. They tend to naturally be so compressed because of their high distortion, but at the same time still have peaks which are caused by the exciting of resonances within the guitar, cab, room etc.
Broadband compressors on guitars mostly act on low-end spikes. That is to say, they will conversely make your guitars more fizzy and edgy. They will feel more upfront and controlled, but also gain a harsh, present character.
I'm personally still experimenting in this area to find a balance that works, but I would advise that unless you are very experienced in mixing, avoid broadband compression on high gain guitars, as it can cause more problems for you than it solves.
Small amounts of limiting can help in the avoidance of clipping at very high master RMS levels. The limiter will act very aggressively on errant micropeaks on the guitars, and allow you to get a little bit of extra headroom. This, once again, comes at the expense of punch, and an increase in presence.
To summarize
You may notice that our 'common problem zones' effectively covered the entire frequency spectrum. This is why distorted guitars are our natural enemy. To repeat: they are essentially shaped, broadband noise which we are trying to get working in concert with other vital, midrange-centric elements in our mixes. It is notoriously difficult to create a mix that breathes, and has a guitar tone which is perceptibly 'strong' as well. The one counteracts the other, and the two goals lie at odds all the way throughout the mix. If you master the art of powerful guitar sounds that don't intrude upon your mix in significant ways, then you've got a huge leg up on the vast majority of the engineering population out there.
Frequency spectrum summary:
With that, we conclude the guitar part of the mixing series. In the future I may come back and make amendments or additions to it, in order to provide more detail as needed, but for the moment we have covered the fundamentals and it should be enough to get most of you on your way.
Thanks,
Ermin.
-----------------------------
Originally, I had hoped to sequence these tutorials in the same way I normally progress through mixes. That is, to start with the drums, move through to bass, then guitars, then finally vocals. The reason that guitars got bumped up the ladder is simply because I'm mixing them right now, and these ideas are all very much fresh in my head. So, don't get a skewed perspective from this - the sequence of tutorials has no greater implication than mirroring whatever is fresh and relevant at that point in my own head.
As a mixing engineer, you need to develop an intense hatred of distorted guitars. They are your natural enemy. Your arch nemesis. Heck, they're your damned Van Helsing.
No other element you run into when mixing will want to, in its raw state, eat up so much frequency real estate. Guitars chew through vital midrange frequencies, they resonate over the top of bass guitars, and their very high frequencies sound extremely harsh and unmusical.
You can think of mixing guitars as a street fight. The weaker your opponent, the more easily you can subdue them. However, the bastard always enters with a glass bottle, and no matter how you go about it, you're still going to come off with some injuries. How substantial they potentially are will depend on three things:
#1 How weak the opponent enters the fight (how balanced your raws are)
#2: How cleanly you're able to subdue them (how well you can mix them into your arrangement)
#3: Whose turf you're on (how well balanced your mix is to accommodate them)
Since this is a mixing tutorial, we will already assume that you've started with a great player, who has recorded with a great instrument, which has been captured with high quality gear. If any of this isn't the case, you need to give up your search for 'perfection' now, because it won't happen. The most you can hope for is to salvage what you can using the following mix techniques, or wait for an opportunity where the aforementioned is the case.
The whole process of recording, and then subsequently mixing guitars, is somewhat contradictory and ironic, as we'll find out later. Your primary goal when tracking guitars - let's assume for a moment we actually care, and are using real tube amps with a real mic or two - is to get as large a sound as you can. What you want for most projects is something with a fine, even midrange balance. Anything out of the ordinary, such as excessive tight peaks (fizz or midrange resonance nodes) will immediately play havoc with the clarity of the mix. So you're looking for a finely appropriated midrange as your #1 requirement. The midrange is where guitars live on records, after all.
The second element is a balanced low-end. Guitars are naturally adversarial here, as they will have cab resonances that commonly lie in areas that excite nasty room modes in smaller spaces. One of the reasons that oversize 4x12 cabs are so popular for distorted guitar sounds is that they shift these resonances lower down the spectrum, where they are perceived as more musical and 'punchy', rather than outright muddy. If you're recording in a small room, then you need to rethink your circumstances. You may well be able to get a good midrange if you isolate your microphone and cab enough from the surrounding environment, but your low-end will never be clear. No matter the amount of treatment you put into that space, there will always be nodes in the low-end that get excited and cause the vital bass/guitar crossover region of your mix to be an absolute nightmare.
Here is where we get into the contradictory stuff.
So, you've loaded your raw guitars into your mix. What alone sounded quite large and nice is now completely clouding your bass guitar, vocals, and generally making the mix sound like a mess, right? Perfect; this is exactly where we want to be. The whole idea behind tracking a 'large' guitar sound, was that so you could downsize it in mixing. This is where the poking of holes begins.
Think subtractively with guitars. For the last couple of years I have not increased the gain on a single frequency band on guitars at all. Picture the ideal tone that's hiding within that mess, obscured by all that junk. Ultimately all you can do in mixing with guitars is to reveal that inherent character in its pristine form. You do this by subtracting the junk. If the raw sound is bad, then your final sound will also be bad. Guitars are notoriously poor at taking processing. Second perhaps only to vocals in this area, they absolutely need to have a good character when raw, otherwise this entire exercise is pointless.
The main issues you will find with guitars is how much parking space they want. They're like that fat asshole with the hummer who gets two parking spots, and gets away with it simply because he knows he's the CEO's cousin. Likewise guitars know they are vital to driving the energy in a rock or metal mix, and take up a whole heap of real estate. They will likely be conflicting with many major, vital elements of your mix. Your process begins by letting these other elements through.
Start with filtering.
High and low passes are commonly the first stage in the sequence of processing done to guitars. This is because the filters reject the most useless of frequencies. Everything that is entirely unmusical and messy can be rejected significantly with these.
The points where you set these filters, and their orders, are entirely dependent upon the mix you're creating. A common starting point is 3rd order (18 dB/oct) filters set at 70hz and 12kHz. Bear in mind that you can be significantly more aggressive than this. It's simply a starting point from where you can jog the filters around until the guitars start to sound the most correct in context with the mix. My personal extremes lie with a HPF at 130hz and LPF at 7.1kHz. Sometimes you just need a tone that fits within a very dense arrangement, and you need to reduce its bandwidth appropriately. Don't be afraid of doing this, if necessary. Get extreme if you need to.
Move onto basic EQ
Think of this as the sculpting portion of the process.
The idea is to do whatever you need to, in order for the guitars to take on a character that's more pleasing toward the mix balance. It's immensely difficult to talk about this in general terms, as each guitar sound is intrinsically unique.
Always remember, guitars take processing very poorly. So get to where you need to go in as few moves as humanly possible. Always have that in the back of your mind, reigning you in. Do more with less.
Referencing guitar sounds can be a dangerous affair. You might well be able to take on the same overall shape and balance of that tone you're referencing to, but in the process you will lose what's unique and powerful about your own tone. It's always a better idea to uncover the potential within your sound on its own terms, before referencing. Reference only for overall balance, if possible - not for the finest of little things. If you are having trouble hearing the rest of your mix in specific areas, then referencing may uncover where to look, but beyond that you should still treat your sound as its own entity.
Common problem zones to watch
Guitars are essentially shaped broadband noise. I like to use them as a starting point for frequency training, because they contain so many frequencies in such abundance. It's because of this specific reason that you will find yourself EQing them all over the spectrum, and will find them causing problems for your mix all over the spectrum.
Take a look at the low-end first. Chances are there will be a peak, or a few of them, anywhere from 100hz all the way through to 180hz. Smaller rooms may even induce ringing and resonance higher up, between 180 and 300hz. Wherever you find these peaks, suck them out! They are exceptionally detrimental to the clarity and tone of your bass guitar. Having some minor 'thump' to the guitars between 70 and 90hz can be beneficial in creating movement and size, but tread carefully. Always remember that the bass guitar has a role - and that is to create tight, expansive low-end which in turn creates the illusion of the guitars being bigger.
After this, consider looking at the low-mid frequencies. Many smaller rooms, or mid-focused amps will leave some junk down here. Usually a wide scoop of sorts can be in order to clean them up. We're primarily talking about 200 to 500hz here.
The next is one of my favourites. I call it the 'cardboard zone'. It essentially ranges from 600hz to 800hz (but on some amps can even be centered as low as 500hz). This is the zone that contributes to what you hear as a 'flat' character on guitars. Generally subduing this range on most raw guitar sounds will lead to an immediate increase in air and dimension for the entire mix.
The next zone is absolutely vital. It commonly ranges between 2kHz and 4kHz, and many times is centered close to 3kHz. This is the vital vocal presence range. This is an area that you need to balance with utmost precision if you plan to have both guitars and vocals intelligible at the same time. There are no specifics to be given here, as it really depends on the excursion of the amp. There will generally be something undesirable going on in this region with a raw amp tone, however, and it's your job to get it under control so that vocals, the most vital point of your entire mix, can poke through unobstructed.
Beyond this area lie mostly isolated fizz spikes. This is extremely amp/cab/speaker-dependent and no specifics can be given. Feel free to sweep a tight EQ boost beyond this region. Listen to hear if anything stands out to you as particularly unpleasant or detrimental to the mix. You can sometimes end up with a few to several tight EQ cuts in the highs to tame fizz. Each cut will bring increased clarity to your mix, but decreased clarity to the guitars. This is the tight rope that you are walking all the time with guitars.
Consider multiband compression
Taking us back to the lows and low-mids, we find many times that sucking out too much of the resonance down here makes the guitars very thin and powerless. This is largely because there are so many non-linear peaks happening in this region that we'd need multiple EQs to get them all controlled in proper proportion - after which there'd be no tone left at all from all the phase shift crazyness.
So what certain crafty engineers have pioneered is the use of multi-band compression on the mid-bass (and some low-mid) frequencies of the guitar. This allows you to tame only resonances that have crossed a certain threshold, and keep the rest of the low-end intact. This is a very good solution for particular cab and room resonances that tend to get excited by particular notes. It keeps the rest of the low-end relatively untouched and allows the guitars to stay meaty as well as controlled.
The fabled Sneap C4 preset is a great starting point here. I personally run the multiband compressor between 70hz and 250hz a lot of the time. Possibly higher if the room was small and the resonances extend higher. I also keep a shorter attack time than the Sneap default, because I like to have my bass guitar control more of the low-end, and don't necessarily want to let much cab resonance through.
Consider broadband compression?
Broadband compression on distorted guitars is an interesting thing. They tend to naturally be so compressed because of their high distortion, but at the same time still have peaks which are caused by the exciting of resonances within the guitar, cab, room etc.
Broadband compressors on guitars mostly act on low-end spikes. That is to say, they will conversely make your guitars more fizzy and edgy. They will feel more upfront and controlled, but also gain a harsh, present character.
I'm personally still experimenting in this area to find a balance that works, but I would advise that unless you are very experienced in mixing, avoid broadband compression on high gain guitars, as it can cause more problems for you than it solves.
Small amounts of limiting can help in the avoidance of clipping at very high master RMS levels. The limiter will act very aggressively on errant micropeaks on the guitars, and allow you to get a little bit of extra headroom. This, once again, comes at the expense of punch, and an increase in presence.
To summarize
You may notice that our 'common problem zones' effectively covered the entire frequency spectrum. This is why distorted guitars are our natural enemy. To repeat: they are essentially shaped, broadband noise which we are trying to get working in concert with other vital, midrange-centric elements in our mixes. It is notoriously difficult to create a mix that breathes, and has a guitar tone which is perceptibly 'strong' as well. The one counteracts the other, and the two goals lie at odds all the way throughout the mix. If you master the art of powerful guitar sounds that don't intrude upon your mix in significant ways, then you've got a huge leg up on the vast majority of the engineering population out there.
Frequency spectrum summary:
- 22 to 70hz: Can generally be filtered without issue.
- 70hz to 90hz: Can have some useful cab 'thump', but otherwise to be treated as dangerous.
- 90hz to 180hz: Cab and room resonance area. Clamp down heavily here in order to aid the low-end clarity of your mix. Clamping down too much will leave the guitars sounding weak. Consider multiband compression.
- 180hz to 300hz: Can contain smaller room resonances, and mud from the cab. Go easy controlling this, and only take out as much as is necessary, as it's very easy to lose power here. Consider extending the multiband compression to this region if necessary.
- Low-mids: This entire region can contribute to mud, depending on the playing style, guitar, amp, cab and room. Treat appropriately
- 500 to 800hz: The 'cardboard zone'. Decrease this to improve the dimensionality and depth in your guitars and mixes.
- 2kHz to 4kHz: Vocal presence region. Tread carefully here as it determines both the intelligibility of vocals and guitars. Finding the right balance here is a masterful technique which can take years to develop.
- 4kHz+: Generally isolated fizz spikes. Treat them with tight EQ cuts in general. As always, be careful, as sucking out too much will kill the edge and intelligibility of the guitars.
- 10kHz+: Can generally be filtered out for most guitar sounds. If you want some more air, you can extend the LPF to 12kHz.
With that, we conclude the guitar part of the mixing series. In the future I may come back and make amendments or additions to it, in order to provide more detail as needed, but for the moment we have covered the fundamentals and it should be enough to get most of you on your way.
Thanks,
Ermin.
