If you want to make sure you get the best out of your mixes, there’s a few things I recommend you check before rendering it to audio. Here are 9 I find the most important. When trying to tick these boxes, it’s a good idea to focus on one at a time and try to identify these issues one by one, dedicating one complete listen of the track to each of them.
To make it all easier to remember, I made an infographic that you can print and stick on your wall if you like.Subscribe to Mailing List and Download in High Resolution
1. No Clipping
Make sure there’s no clipping on your channels. It’s easy to lose sight of the volume meters when your creative juices are flowing but I always try to keep an eye on them and make sure I don’t get red lights on any channel. It’s probably a good idea to bear that in mind from the word go – setting the peak of your kick and snare around -7 dB is usually a good starting point.
Aim for 3-6 dB of headroom with your mixes. That gives you plenty of room to work with, especially if you add more instruments to your mix, and you keep your mastering engineer happy, too. That’s assuming you don’t use mastering plugins (maximizer, limiter, compressor) on your master channel. I personally don’t and normally advise against it but I wouldn’t say it doesn’t have its benefits. I will try and make a separate post on the topic, there’s a lot to it and there are certain scenarios when I actually would recommend using certain mastering plugins before the mastering phase.
Check every channel to see if there’s any sudden level changes. If there is, the question is, is it intentional, or perhaps so quick that your ears can’t quite catch it – why you missed it in the first place. A typical case, for instance, would be a bass sound with a very short attack and with Ableton Live’s Amp device on, that might generate a very short but quite loud peak at the attack that you can normally do without. You probably wouldn’t want to remove it entirely, but often reducing 4-6 dB is almost unnoticable and frees up a lot of headroom, so you can make your track louder in RMS. A great tool I use for spotting short, unwanted peaks is Vengeance Sound’s Scope, a free analyser plugin with scope, sprectral and stereo field analyse capabilities. It shows a real-time waveform of what ever you put it on.
Mixing issues are easier to spot when mixing or mastering other people’s music, because you’re hearing the track for the first time. Our own tracks can often have masking problems but they are more difficult to notice, because we made it. So what’s interesting is, the person who made the track may not realise if a sound gets washed by the others for a moment, actually, they probably hear it’s there even if it isn’t. Spotting the absence of something, especially when it only occurs unexpectedly and rarely, is very difficult. Our eyes see the sound in our DAW, and we “know” it’s there, too, we put it there after all, so our ears correct the masking. A technique I recommend using is listening through the entire track a few times, only focusing on one particular sound at a time, with one question in mind: “is every single note of the sound 100% intact?“. A kick drum, for example, might lose some of its transient a few times throughout the track. A good fix for that can be to try and get things out of its way for that short amount of time.
The truth is, depending on how important a role a particular sound plays, it can be fine for it to get secondary. You might not hear every detail of a guitar sound, for example, because the vocals might mask it at one point, but that’s fine because the vocals will often have higher priority than the guitars. So a question to ask is: “can I afford this sound getting masked by a more important one?”. With the most vital instruments, you can’t, of course.
I believe it is crucial to understand the priority level of each instrument in your music. If a sound is turned down in the mix, the person listening to it will think it’s not a very important one. Although, if it clearly is, like the vocals in rap music for example, there’s something obviously wrong. I believe it’s generally a good idea to have at least two or three different groups of priority levels, and a clear understanding to which instrument belonging to which. A pad will often go in the lowest importance bag, a lead will typically have higher priority. If you browse through pad presets you will hear softer sounds with less high frequency content, as if they were less significant, distant, definitely not close to the listener, whereas if you check lead presets, you will hear in-your-face, sometimes harsh sounds with typically a lot of mid and high frequencies. Sounds that act as decoration or sound design, like vinyl dirt for example, or some field recording to give a little texture to your drums, usually go in the third group. Again, it doesn’t mean they are not important, it only means that when it comes to staying on top, it is fine for them to make way for more dominant ones.
So what you want to ask yourself is: “are the most important sounds the loudest and are they clearly audible throughout the entire track?“.
Having one (or more) reference track can be very useful to making sure you set out and stay on the right track all the way through. Our ears get tired easily and quickly, so a reference track often acts as a sort of ear calibration. I recommend checking your mix towards your reference track(s) regularly.
I consider every sound for a high-pass filter, unless I have a good reason not to. Only very rarely do I feel that a hi-hat needs frequencies lower than 600 Hz. Pads, guitars, leads, vocals – I chop off below 2-300 Hz. Go through each track in your project and make sure that every instrument gives way to the kick and bass, especially below 150Hz. Often you might not expect a sound to have low frequency quality content, but checking with a spectrum analyser may prove otherwise.
When mixing for other people, the easiest issue to notice is when a sound needs more limiting. That’s simply because I get individual tracks rendered as audio and can identify sudden, unwanted peaks on the waveform. Of course not every sudden peak is unwanted, so after spotting one on the waveform, I listen to it to see whether I can back up my suspicion with my ears, too. A good example would be the kick drum channel getting very loud when two beats overlap (people sometimes don’t remember to set the sampler to monophonic when making a kick), or a bass sound with a distorting high resonance filter at one point. Of course, limiting is not the always the only or best solution to every peak-related issue, in the latter example, reducing certain frequencies with an EQ, or applying a dynamic EQ, or multi-band compression might make more sense. Check every channel in your project and make sure all of them are under your control.
9. Signal Path
Teaching students one-on-one and opening up their projects have showed me many times how common conflicting mixing techniques are. For example, someone in one studio session might use an exciter to make a sound sharper and crispier, but in a later session use a low-pass filter to remove high frequencies of the same sound. If you don’t consciously look for these incoherencies, you may never spot them. Every mixing decision should support your mixing vision. A good question to ask with every track is “what is my aim with this instrument”, and make sure that that aim is followed throughout the entire signal path.
Once you’re happy with your mix, and ticked all of the boxes above, you can render it to an audio file. I highly recommend opening it up in an empty project in your DAW or in an audio editor, to see what the waveform tells you. Like I said earlier, some issues may get hidden from your ears but can get very obvious when looking at the waveform. An example could be a very high peak that you see occuring regularly, maybe 8 times during the song in total. Maybe you find out that it’s a moment in the song when, let’s say, three sounds overlap. You might want to go back to your mix then, and see if there’s a good fix for the problem. One time, in a similar situation, I went back to the mix, phase inverted one of the overlapping sounds, and bang, problem solved. A much more intelligent solution than smacking a limiter on the mix, it actually makes the problem go away, whereas a limiter would not.
I hope this post will help you get more confident with your mixes. As always, if you have any questions or tips, please feel free to post them in the comments. If you find these production tips useful, please give them a ‘like’ or share them or support which ever way you see fit. Thank you.