How to Make your Mix Louder

Patrick asked: I usually mix conservatively since my interface has lots of monitor volume range (so I can turn that up to compensate). What can I do to bring up the volume after? Limiting? A volume boost plugin then limiting? I'm wanting a super dirty master for my style.

This was something that confused me a lot when I started out. When mixing, you have volume controls all over the place, input gains on your sound card, the track volumes in your DAW, plugin volume controls, the master fader in your DAW, the master output on you audio interface, and finally the volume knob on your monitors. So how do you get your songs nice and loud to an appropriate level?

Now to start with, this article is about making the mix louder, despite the picture above, we will be discussing raising the RMS volume quite high. If you fully oppose the loudness wars this isn't the article for you, as we will be exploring trying to match the volumes of modern music.

So, the very first thing I'd recommend is that you grab a few songs that you want to match in volume and mixing. I tend to leave this reference track as the top track in my DAW so I can easily reference against it.

(If you have yet to record a track, do that first. You will not be able to match a mastered track in the recording stage so wait until you are at the mixing stage)

So once you have picked your reference track(s), leave their volume faders in the DAW at zero. Never touch them again. Listen to your track and set your monitor and headphone levels to a volume that you are comfortable to work with for a few hours. If possible don't touch the listening volumes again, if you feel like you are getting ear fatigue, take a break instead of dropping the volumes you will be more fresh to come back to the mix.

So let's get to the actual mixing part. I will assume you can mix to a sufficient level, if you can't don't worry, continue reading and practising but you may find it a bit difficult at first.

The very first thing I do in a mix is clean it up. There are so many things that hog up headroom and we want to get rid of them.

Use charts like this, if you are unfamiliar with frequencies.

Let's say you have a piano, a synth, and a vocal that are all really strong around 500Hz, they will all fight over that frequency and it will sound like a mess. You will need to figure out what comes first. Usually this would be the vocals so let's keep them in the front. 

Dip the 500Hz regions of the synth and piano, you should hear the vocals suddenly clearer. When solo'ed the piano and synth may sound a bit odd but remember that you are EQing in the context of the mix as a whole and so it won't be as noticeable.

Consider the bass frequencies now. These are the main culprits of taking up headroom. You want as much of the bass frequencies limited to the kick and bass as possible. So, using the chart above and your ears, start taking the bass out of as many instruments as possible. For example, the male vocals don't really extend below 100Hz, take everything below that out and then bring the EQ up even further until you can hear it eating away at the voice. Bring it back down to just below that point and you will have removed all redundant frequencies.

Synths tend to occupy a lot of space across the whole frequency range. While they will sound good by themselves, you can usually take out a lot of the low end for the benefit of the mix as a whole. If you need more advice on using EQ, read this article here.

Once you have removed all the bass from the instruments that don't need it, look towards the kick and bass. A lot of electronic kicks and basses have content that we don't need. Consider our limit of hearing, we can't hear anything below 20Hz and most speakers cannot even reproduce that frequency.

Many bass synths have a lot of content below 20Hz so remove all that, as even though it is inaudible, it is still taking up headroom. No point using up space that will never be heard. As a rule, I also take out everything below 20Hz on the master track as effects added after the EQ can add some very low end content.

Now we can look at compressing the signal. A point to consider is whether to EQ or Compress first, I'll let you read up and decide how you want to order it but I tend to use EQ before compression.

Browse along each track and look for peaks that really stand out. Obviously the drums will have some but look especially at other instruments first, such as vocals. 

ReaComp is my go to compressor for the first level of compression
it comes with REAPER and is also available to download for other 

I like to have a fairly even volume for vocals as you don't want the quiet bits drowned out by the rest of the song. It is always more transparent to add a few low ratio compressors in series than to compress a signal really hard once. Try both and see what you prefer in the context of your mix.

One brilliant compression technique is parallel, or New York, compression this is where you mix a hard compressed signal with an uncompressed version of the same signal. This can be achieved with ReaComp by having the Dry signal at 100% and mixing in the Wet compressed signal to taste. This keeps some of the perceived dynamic range while boosting the RMS volume. Parallel compression works great on the drum and vocal bus. 

For good overall track volume, I compress in several stages. I compress individual tracks that need it and try and put as many tracks into busses as groups. For example, I will have a drum buss for all the drums, a vocal bus for the vocals and a synth bus for the synths and pads. Sound effects and other instruments can be added to particular busses or left as their own tracks. 

By compressing a bus, you are essentially glueing a sub mix of your track together. I tend to use gentle compression on the sub-mixes just to keep them nicely together. Alternatively, if the bus contains more dynamic content, parallel compression (as described above) can work really nicely.

Consider side-chain compression for the bass and kick so they sit nicely together. This can also be done on pads and synths to make more room for the kick. I often send the vocals to the synth bus as the trigger signal and gently side-chain compress the synths for a bit more vocal room. The trick with this is to do it so that it is inaudible. By combining the use of attacks and releases and good control of threshold and ratios, you can get a few dBs more room without it bouncing all over the place.

Refer to your reference track.

This is the stage where multi-band compression works very nicely. I tend to group all my tracks and busses under a "mix" bus and apply multi-band compression onto that. This is a style of compression that I can never get right (everyone sucks at something) but if you want a good compressor with awesome presets, try ReaXcomp. It is another Reaper stock plugin available for other DAWs and it's presets are so good in fact, that I tend to stick to them.

As with ReaComp, ReaXcomp is a fantastic multi-band 

Now, I feel this is a good place in the article to talk about Voxengo Span. It is an FFT (Fast Fourier Transform) spectrum analyser. At the bottom, it displays the RMS volumes and shows the frequencies that are expressed in real time. Use this to your advantage when seeing your average volumes and seeing how much you may want to boost them.

Span, by Voxengo is a free plugin that you should have.
It is an essential analysis tool which I've only scraped the 
surface of!

I should probably add a paragraph explaining the difference between volume types. You have peak volumes, which is the volume of the highest peak. This is a bad indicator of a tracks loudness as the peak may only last milliseconds. While it is good to know the peaks of a track so you know it isn't clipping, the rest of the track could be significantly quieter. Boosting the volume until the peaks hit 0dB may still mean that the average volume ends up at -25dB.

By looking at the RMS volume, you get a greater sense of the overall, average volume of the track. This will fall quite short of the peak volume and we need to keep it that way otherwise the whole track will just be one big, fat sausage and sound awful.

However if you want a loud track, you do want to raise the RMS volume. When I follow the stages above, I tend to get around -14dB RMS on my tracks, sometimes higher, sometimes lower. My general aim is to get between -8dB and -6dB RMS if I want a loud track. Some artists can get as high as -4dB RMS and still sound awesome but I'm not that good, that takes serious mixing indeed!

Good ways to increase the RMS volume of your mix is to add some more gentle compression on your mix bus. Parallel compression also works in some scenarios where you want a chunky sound but try it and be willing to remove it if need be.

Refer to your reference track.

Once your mix bus is sounding good, render it and have a look. It should already look loud and be quite sausage shaped. Feel free to leave it there. There is not always need to go further. But if you want to keep reading.

Once you have a rendered, mixed track, remember that you cannot see the effects you placed on the mix bus. Consider if you actually need another multiband compressor as there is on applied before the render. It just won't be visible in the chain any more.

Refer to your reference track.

So let's look at some cool mastering plugins to add for even more volume. First, add Span again to the track and keep it at the end of the chain every time you add an effect or processor.

M Horse P3 is a free mastering tool by Terry West and is a brilliant tool for improving your master. There are many effects and you will really need to read the manuals and tutorials online to get the most out of it. This does pretty much everything you'd want with some awesome classical sounds as well as modern mastering sounds too!

Ferric TDS is a free tape emulator which is really good for nice saturation and gentle compression, I tend to have this as one of my last plugins on the chain. It is really good as an indicator of a bad mix too. I use it with hard saturation settings and if it sounds shit, I should go back and fix the bass frequencies. If it doesn't, I'll slowly turn back the saturation until it sounds nice and warm.

Loudmax is a brilliant free limiter which I stick at the end of the chain to make sure nothing exceeds 0.1dB it sounds good, even when worked hard but I don't recommend it. Bring the threshold down as far as you want, but realise you are hard compressing your song with it. It is good for bringing up the volume just a little bit more, if you need to run it hard, consider a slightly more generous application of compression further back in the chain, or even go back to the mixing stage.

Refer to your reference track.

And there you have it. A nice loud track. Don't feel you need to do everything, and don't feel like you need to limit the shit out of your song. If you are producing a dance song for a DJ to play, remember that even if it is a few dB quieter than the others, the DJ has gain controls so the main thing is to keep your song sounding good. It is then up to the DJ to let it sound the same as everyone else.