This Cyber Monday Tuts+ courses will be reduced to just $3 (usually $15). Don't miss out.
Many of you readers here at Tuts+ Premium are struggling to make your home recordings as good as possible. Although you might not achieve the big boy sounds you hear on the radio, you can actually get pretty darned close. The mixing phase plays a big part in making your records sound professional, but the production and recording phase obviously plays a big part of it.
In the following Premium tutorial I'm going to take you on a troubleshooting journey on how to make the most of your rock recording. I'll show you how to make a simple drum recording sound punchy and powerful, how to make the guitars and bass play nice, how to create an automatic fluidity between distorted guitars and the vocals and finally, how to make that terrible closet-sounding vocal sound a little better.
The most useful thing you can do when recording your tracks is to make up your mind. Production plays a big part of your workflow, and if you have thought the arrangements through and recorded your source sounds the way you want them, then half the battle is won. Sometimes you need to produce in the box, constantly changing your mind and figuring out what to put next. And it doesn't get easier when you have hundreds of drum loops, synthesizer patches and guitar amps lying around that can change your production a million times over.
Look at your production from a live, rock-band perspective. If everyone has their sound down, knows their parts and can play through it live, then that's what you need to record. Nothing more, nothing less. Sure, you can add overdubs if need be, but make sure that you don't get carried away by options. Options hinder productivity and your production.
I like using this song as an example because it teaches you how to make decisions and stick by them. In the following examples, the song was recorded like we wanted on the way in. We found our desired source sounds and stuck with them. The guitars are heavily processed with delay and flange, and that's fine because that's what we wanted the guitars to sound like.
All right, let's jump into some advanced examples on how to make your rock mix sound polished.
Here is the mix before we've done the processing in this tutorial.
In hard rock, those drums need to be punchy and tight. I won't go into the specifics of how I mixed each drum track. I follow the simple procedure of compressing to tame the peaks, EQing the mud and boxiness out and then boost the highs and lows in every drum.
Some things to think about:
- Attack and release on the kick - Attack is medium. I want the snap of the beater to come through and not get squashed by the compressor reacting to the volume of the kick drum. Release is medium/fast, just fast enough to open up before the next kick hit. There are a lot of double kick drum hits and such so I want the compressor to be reacting to each hit and not pumping over the whole track.
- The importance of filtering - Filter first. Always filter what you don't need. I'm filtering all the way up to around 50 Hz and I'm also low-pass filtering down to 7 kHz on the kick drum, since there's nothing except bleed from other drums dominating the higher frequencies.
- Double compression – A good compression technique is to insert one compressor before EQ that tames the peaks coming in, and then inserting another after the EQ to raise the overall level. In Logic you can choose either Peak or RMS level. Use Peak for the first compressor and then RMS for the second one.
To give the drums some extra punch I've routed all the tracks via an aux send to a heavily compressed bus.
Using the Multipressor in Logic, with some heavy settings I'm squashing the drums all together. Heavy thresholds, ratio and fast attacks are the name of the game when you're doing parallel compression on drums. Especially for a rock song like this the more compressed the drum bus, the punchier it will sound underneath the original drums.
I'm using a multiband compressor but a normal single band compressor would work just as well. The only thing that's noticeably different is that it squashes the higher frequencies more than if you separate them from the low end. The multiband compressor can squash the kick and snare while leaving the cymbals relatively uncompressed, which I think sounds better.
On the aux track, after compressing, I've boosted the low and high end considerably, as well as cutting the obligatory middle frequencies. They just don't sound good on drums, so I try to get rid of as much of the mids as possible without compromising the overall frequency balance of the drum kit.
In order to move the drum kit out of the forefront it's always good to add a nice reverb to the overall kit. I've sent each drum channel to a warm wooden studio impulse response in Space Designer. This is basically a small room preset that's not overly reflective. It adds a nice ambience to the drums, pushes them a little bit back without drowning them in reverb. In fact, I liked the reverb so much that I ended up adding a bunch of instruments to the same reverb, effectively making it the overall main reverb to glue the instruments together.
It's crazy how much of a difference the reverb and the parallel compression can make to the drum sound. Before it's just a bland sounding drum track, but with the NY Trick I make it way punchier, maybe even too much so. But with the reverb around it there's some space that pushes things back and makes everything sit a little better.
These things might sound basic, but they are not. If you neglect doing these things you won't end up with a punchy enough drum kit. Compared to the rest of the track it will sound weak and flabby.
Check out the mixed drum kit without any parallel processing:
There's nothing really wrong with this mix. All the drums sound really good, but you wouldn't know how much better it sounded unless you used parallel processing.
It's the extra mile that counts.
Bass is easy to record DIed. You get more control over the sound that way, but you can also lose some of the “realness” you would get from recording an amplifier. Amp emulators can fake this process to make them sound a little less DIed.
The bass sound before any mixing is already sounding pretty processed. That's because we decided on that sound during the production phase and stuck to it. It was recorded DIed with some distortion and into a digital pre-amp emulator.
Even though it really growls, it still lacks something. If you recorded DI then you get a very dry DI sound, and amp modelers can change that. They seem to give just a little air to the signal and warms up the otherwise cold DI signal.
You can also re-amp your DIed bass for a different signal, either with a bass amp or just using your studio monitor to get a little air moving in your bass sound. But that can take a while, and won't give you good results if you don't have a good place to record bass.
Notice how the bass amp gives the track that hint of depth?
A Note On EQ
If you want a really aggressive bass, and in this case we do, boosting the middle frequencies will bring out the attack. On this track I boosted pretty heavily in the 700 Hz and 2 kHz range because that's where the growl from the bass and attack from the strings lie, respectively.
The guitar is probably the most processed track on this entire production. The phaser and delay sound is a pivotal part of the track, and is really the underlying part that drives the song forward.
It seems counter-intuitive to say that every part of the mix needs to be punchy and in your face. Because mixing logic would dictate that if everything is screaming for your attention, nothing is going to stand out. However, that's not really true because using EQ and creative compression you can make all the elements of your mix play nicely together.
Guitar and Bass
It would seem like I'm beating a dead horse here but don't underestimate the power of filtering. Especially when it comes to making a really thick guitar part fit with a really thick bass part.
The guitar sounds super beefy and awesome in the low end, but with everything else it clutters up the mix. I'm filtering all the way up to 200 Hz to clear it up. A heavily distorted track like this also sounds noisy and hissy in the high-end so I've filtered out everything above 10 kHz. To compensate for the heavy low-end filtering I've added some fullness in the 500 Hz area, boosting the mids instead of the bass so that both the bass and guitar have enough room to play.
Listen to the guitar gets in the way of the bass track. The end result is a cluttered low end that lacks definition.
After filtering and some extra EQ there is more clarity to both tracks. The guitar doesn't get in the way of the bass guitar as much since they both have their respective EQ areas.
I think this is the coolest mixing trick in this tutorial. There are two compressor on the guitar track. The first one is only lightly compressing the peaks, but it still gives us a punchier guitar sound.
What's so cool about the other compressor is that it is side-chained to the vocal so that whenever the singer starts singing the guitar ducks behind it. It's compressing around 4 dBs every time the singer sings, but I'm only using a ratio of 2:1 so the guitar doesn't sound compressed. The singer just lowers the level of the guitar automatically, and once he stops singing the guitar swells back into the mix. It's actually pretty cool because it give the phaser effect an added volume swell as well. Think of it like the phaser modulating the frequencies from left to right while the singer makes the amplitude go up and down.
Ducking is used in a lot of things, ducking the bass behind the kick drum every time the drummer hits the kick is commonly used. And ducking is also used in radio broadcast a lot, as you can hear that whenever the announcer starts speaking the volume of the song is automatically lowered.
It sounds good in the context of the mix, but if you solo just the guitar and vocal you can easily hear how the vocal controls the volume of the guitar.
The vocals, not surprisingly, were the hardest part to get right. The main problem was that the source recording is just plain horrible. It was recorded in a closet without any acoustic treatment. It doesn't have a big room sound, but it definitely could have been done better. So in the end I decided to go for a more effected, processed rock vocal than a clean and clear one.
Many of you might have to deal with vocal recordings that sound like this:
EQing out the room
My normal process when EQing vocals is to find offending frequencies, cut them and then boost where I think is needed. In this case I did the normal filter at 100 Hz, reducing rumble and unnecessary low end. I was having some problems with him sounding telephone-like and nasal so I tried to cut those frequencies at two places, 760 Hz and 1 kHz respectively.
This cleared up some of the nasal character that was apparent in the recording. That didn't clear up the nasal sound as much so I resorted to masking it by boosting the 5 kHz area.
Masking is an EQ effect that happens when you boost a higher frequency than the one you are having trouble with. Since your ears will more likely hear the boosted frequency you are essentially masking the problematic frequency from being heard.
I'm using two sends on the vocal, a delay for depth and space and the other is a doubling effect.
- Delay – The delay effect is compressed on the way to the effects, just to control the peaks. From there it goes into the Space Designer plug-in. The Space Designer is actually a reverb but there is an interesting Tape saturation impulse response that gives the delay a gritty character. I wanted to over-effect the vocals to make it sound aggressive and ugly, and the tape saturation is a nice addition. From there it goes into the Mono tape delay plug-in. I have it set to a medium delay with low feedback. Sort of like a a slap-back echo on steroids. From the delay it goes into an EQ that is a little different to the original vocal to get a little separation from the two.
- Artificial Double Tracking – Artificial double tracking is achieved by delaying the signal by 20–30 milliseconds, just so that the ears perceive another signal being present. In addition to the stereo delay, which has 20 ms on the left side and 30 on the right, I've also added a pitch shifter to lightly pitch the signal up by a few cents. It's not enough to make it sound detuned or chorused, it's just enough to make it sound distinct. And finally, in order to get a different sound and separate the doubled vocal from the original I've EQed it a little differently.
So we end up with a grittier, aggressive and in-your-face vocal. Sometimes you need to go overboard to get rid of your home recording sounds.
After all this extra processing we've ended up with a much heavier and punchier mix.
The extra mile is what makes a mix better. Yes, EQ and compression will take you far but creative thinking and extra processing will make your mix great.