Tips for preparing to send your tracks for mastering:
from Francis at Echoplex mastering services
"As a mastering engineer I hear a lot of mixes. Some are great and some really aren’t so great. While it’s now true that anyone can create a good mix with almost any gear it’s still easy to mess up a mix and betray the fact that you didn't spend thousands at a top studio.
Frustratingly, the same errors crop up time and time again and it’s these that separate the greats from the not-so-greats. Mastering can be the stage where the final layer of magic is applied, but mastering is always the stage where any mix problems will be ruthlessly exposed.
Think of mastering like holding up a big magnifying glass to your tracks… there is nowhere to hide! What follows is a small list of the most common mix errors I see, how you can avoid them and give your mastering engineer the best possible chance of creating an awesome final product. So let’s look at some useful music mastering tips and common mixing mistakes to avoid."
1) The Bass and Kick are too loud
Without doubt the biggest error I see is the low frequencies being pushed to a needlessly loud level. This is a misguided practice for one main reason – overly loud bass eats up headroom and makes it harder to get a loud master.
At the mastering stage an overly loud bass can make it very difficult to achieve the holy grail of “LOUDNESS” as the highs and mids won’t get to be that loud before the bass or kick starts clipping or distorting. Not cool if you want your track to be as loud as the vast majority of commercial releases. So why do so many mixes have this issue? When speaking to people who have submitted tracks where the bass has been mixed too loud the answer is either, “We wanted the bass to sound really big in the club” or “We didn’t realise it was so loud”. To deal with the first point, if you are making music that will be played in clubs remember that club sound systems will already be designed and calibrated to make sure that any bass frequencies are amply represented. Mixing should be a time to balance levels, sculpt tones and get your music sounding good, not loud. The second point, however, needs a slightly more complex answer…
How to avoid it:
Bass frequencies are hard to accurately reproduce in listening environments and if it was easy to fix then this mistake wouldn’t be a common one! However, the best thing you can do is ensure that you have the most acoustically neutral mixing environment that you can get. It is not possible to overstate the importance of this. You can have the latest plugins or most expensive hardware but if you’re using them in an untreated room then you’re mixes are always going to struggle. You can achieve this through use of acoustic products such as diffusers, absorbers, and bass traps, and you can even use well placed blankets and duvets if money is tight. Get used to your space and listen to plenty of reference tracks.
Using acoustic products will not only improve bass reproduction but, done properly, you will notice improvements in stereo imaging and reproduction in the highs and mids. Another way to improve here is to periodically check your mix through a frequency/spectrum analyser. It may take a bit of time to get used to an analyser and learn what you should be looking for but the rewards can be huge. Again, listen to plenty of reference tracks while watching the analyser and start to piece together how the producers and mixers got their track to “look” like that.
2) The mix disappears when checked in mono
Hang on, mono? Didn’t that die in the 1970s?? No! Many places and items use mono – stadiums, clubs, shops, TVs, cell phones… the list goes on. Lots of electronic items use mono speakers as a convenience or for cost reasons, but systems in big spaces will use often use mono to ensure an even sound across a location. As part of the mastering process I’ll sometimes check what I’m doing in mono and when I hit that button elements of the mix just disappear! So why does this happen? The main reason I’ve encountered is misuse of stereo widening effects. Many widening effects work by adjusting the phase of a stereo signal (phase is too big a topic to try to explain in this article but there are good articles and resources all over the net) and when pushed to the extreme the signal disappears in mono.
How to avoid it:
This one is easy – check your mix in mono! Nearly all DAWs have a Mono switch on the master channel (if you lack this function then Brainworx do a great free tool called BX Solo which allows summing of all incoming signal to mono, amongst other things). Quickly checking in mono will immediately tell you if anything’s wrong and you can do something about it there and then. If the offending cancellation is coming from a part that has been recorded in from a real world stereo source then you can use your mixer’s polarity inversion switch to counter act. Another thing that may help is to grab one channel of the stereo recording and nudge it forwards or backwards until it starts sounding good. This approach can also help if you have a multimic’d source like a drum kit. If the offending cancellation is coming from an effect then ease it off until it’s sweet, or find an alternative effect. Bear in mind that not all stereo widening effects are equal and you may get results close to what you want without phase cancellation using a different plugin. Experiment and see which works best for your track. Phase monitoring plugins are also available to help give you a quick visual aid with any phase issues.
3) Mix is too bright
Making audio sound bright is a big temptation when mixing. Lifting up the high frequencies can be a great way of adding extra sheen to a production and making your music sound “big money”. Also, human hearing is particularly sensitive to frequencies between 2kHz and 5kHz, and it is a great temptation to load up this area with information in an effort to make parts stand out or have presence. But what can happen is the highs are inundated with extra additive EQ, extra energy and extra bite and it all turns into one big spiked brick of harshness.
How to avoid it:
Subtractive EQing is your friend here. If you have a part that you want to stand out carve out some space in the other tracks that occupy the same frequencies. Automate your EQ to give space to the main lines when they appear and let that shine. You can achieve the same effect with a dynamic EQ if you have one available. Compression can also help if the offending part is percussive in nature or has a spiky edge so experiment with faster attack times, or even with a transient processor, to tame high energy spikes. Try to avoid EQing parts in solo as it’s all too easy to mix a part to sound great on its own and lose sight of the overall picture. If you want to add more to a part remember that with additive EQing less is more. Use your ears and don’t be afraid to back off if you’re unsure – at the mastering stage it’s much easier to add a touch of “something special” than to remove a problem.
4) Mix is muddy or excessively thick
Odd rumbles, lack of definition, bass fighting kick, kick fighting bass, bass and kick fighting a weird energy that’s pumping in time with the pad or lead or high hats… Low end can come from anything, and from many places you don’t expect. If there’s an excess of low frequency energy it can cause all sorts of problems at mixdown and mastering by eating up headroom, interfering with compression, and generally being a PITA. This is a big problem if you’re working with audio that’s been recorded through microphones as they tend to pick up all manner of bumps, knocks and rumbles. Synths and samples also contribute to muddying mixes as their sounds can be very full sounding which, in the context of a mix, can be inappropriate as you may only need a section of their frequency content to get the effect you’re after.
How to avoid it:
One of the first things you can do is identify which tracks should contain the low frequencies (e.g. kick, bass, that tuba freakout in the breakdown, etc…) and then put a high pass filter on every other track to completely remove any rogue frequencies that may otherwise be free to roam. This will free up all of the low end space for the parts that actually need it. Then, with the parts that you’ve identified as needing the low end, sculpt individual pockets in the mix for them all to live in. An obvious technique is sidechaining the kick to duck the other parts when needed. This doesn’t need be an obvious “Eric Prydz” effect but can be as subtle as you like. Don’t be afraid to get really surgical with cutting out low frequencies, as many instruments can be brutally hacked at before sounding odd in the mix. Additionally, and maybe counterintuitively, setting a high pass filter on these low end parts can also clear up a mix in certain situations. I’m not talking a high value here, but cutting out the super low end can give extra room for the vital low frequencies to operate. As always, experimentation is key as techniques and final settings will always depend on the source material.
5) Too much processing on the master channel
For better or for worse it’s become very common to mix with lots of processing on the master channel. Nearly all modern music demands a certain sound that can only be achieved through master channel processing but care must be taken as it can be a double edged sword. Too much eq and you skew the tonal balance of the whole track, too much compression and you squeeze the life out of a track, and too many “audio candy” processors will drag in tonal issues, phase problems and all manner of oddities. Many times I’ve had to send a mix back and ask for a version without master channel processing as it’s been used in an inappropriate way and ruined most chances of getting a good master.
How to avoid it:
When approaching a mix resist applying master channel processing until you’ve achieved a mix that’s tonally balanced across the frequency spectrum, has an appropriate dynamic range and, put simply, sounds good. Once you’ve got your mix structurally developed then start adding small amounts of processing on the master channel and tweak your individual channels accordingly from there.
Special Note – Limiters
Many people will have different ideas as to whether limiters have any place at all on the master channel when mixing. There’s definitely a spot for them when you’re referencing your mixes against commercial releases or if you need to give a preview of your un-mastered track to anyone but it’s probably a good idea to avoid them during mixdown and there are two reasons for this.
1) If you’re using a limiter on the master channel to increase loudness and energy take it off immediately. Mixing is about getting your track to sound good, not loud. Mixing into a limiter used in this way will rob you of a chance to hear how your track actually sounds as you’ll be a getting a wonky image of everything that’s going on – dynamically and tonally. Loudness should come at the mastering stage – if you need more volume when mixing turn up your speakers!
2) If you’re using a limiter on the master channel to catch peaks or stray transients that are making your track clip take it off immediately as your individual channels are too loud and they should be adjusted. In this wonderful world of computer based DAWs we don’t need to worry about noise floors or signal to noise ratios in our mixers like they did in the days of analogue desks so there’s no excuse to be mixing your channels at levels where they might be clipping, or clipping the master channel when summed together.
It’s important to stress that none of these tips on their own will propel your tracks on to MTV and they’re definitely not rules that can’t be broken, but being aware of common pitfalls is half the battle already won. Keep experimenting, keep cranking out the tunes and practising your mixing and, as you may have already guessed, the best thing you can do to make great master is have a great mix. Get mixing!
Make sure your track doesn’t peak above 0dB
Nearly all mastering houses will ask for a file that peaks somewhere between -12dB and -3dB. This is to ensure that all files that come in to them don’t clip at any point. Rendering any audio that goes above 0dB will result in a clipped file and distorted audio, and once you lose those peaks you cannot get them back so your audio will suffer come mastering.
Before you render your finished piece, have a play through and watch your master channel like a hawk, making sure that at that no point does it register 0.0db or more on either the L, R or combined channel meter. If it does nudge into the red then you need to turn down the master fader until it stops going over. “But,” I hear you say, “Wouldn’t a limiter do this job?” If you mix with a limiter on your master channel then allow me to suggest that you’re doing it wrong and you should go back and read my previous article on why this is a bad idea.
Although it’s nice for us mastering-types if your audio peaks between -12dB and -3dB it’s not absolutely necessary. Just make sure it doesn’t go over 0dB and your mastering engineer can turn it up or down if needed.
Don’t cut your end short
If your track finishes with a reverb or delay tail let it end naturally and, even better, add a few seconds at the end to be absolutely sure you’ve captured it all.
Have you ever heard a track with an effect tail off that ends abruptly? Sounds cheap and amateur, doesn’t it? Don’t be that guy. Place your render markers appropriately or, if your DAW has this option or an equivalent, choose “Render all project”. That way you can be sure that you capture all of your audio and your mastering engineer doesn’t need to resort to any awkward fade outs.
If you want a fade in or fade out on your tracks should you do it yourself or leave it for the mastering engineer? There are no rules here and sometimes it’s not possible to remove fade ins and outs that have been part of the creative process but if in doubt then please ask. Some prefer to do any fading as it leaves more scope to get the “right” result, especially if your tracks are part of a fully sequenced release where the mastering engineer is responsible for the gaps and crossfades between the tracks.
Use a high resolution audio file
This may seem like a simple thing to point out but it still happens… Don’t use an mp3. Use a .wav or .aif at the bit rate and sample rate you used when writing or mixing the project.
There can be certain pros (and cons) from using a higher bit and sample rates when writing and producing but should you increase the bit rate and sample rate at the point of rendering? The jury is still out on whether this makes any difference and is too deep and boringly technical a subject to go into here. if it already sounds good then probably best to leave it as it is. Making last minute changes to something that is already good can be a risky business and I guarantee you nobody else that listens to your track will notice if it’s at 16/44.1 rather than 24/48, 32/96, or any other variants.
Please, though, don’t send mp3s.
Give your engineer as much accompanying info as possible
Information about tracks sent to us makes our lives so much easier you would not believe it. You would also not believe how much info is not provided with the tracks. My number one crime in this area is not naming the file properly. By “properly” I mean using the song name. Unless you are, quite literally, naming your tracks numerically then song names are vital for our sanity and the smooth flow of a project. Don’t feel as if you have to strip away any other info in a file name (you might be numbering your mixes of a particular track, or some other identifier) but please include a song name. Also, if there are remixes of a specific track then please clearly label the remixes with an identifier, ideally the remixer. If there is more than one original artist on a release then please use the artist name in the file name too.
Moving away from file names, don’t be afraid to include a little text file containing any other information that you might think is useful. This could include ISRC codes; full track, artist and remixer names; names of reference tracks; your thoughts on what direction you want the mastering to go; running order and notes on sequencing (if you’re making an EP or an album)… Although it’s entirely possible to provide information that we may not need you can never provide too much!
Last edited by Mark Bliss
on Sat Mar 21, 2015 5:42 am, edited 1 time in total.