Last year I attended a session on mastering during the Canadian Music Week conference in Toronto, Canada. The room was packed, filled with many independent artists who were eager to learn some tips and tricks for mastering their own music (as self-recording is becoming more and more possible and affordable with modern advances in technology).
The presentation was less than satisfying and I was (along with many other audience members) frustrated, but I know now that it was not at the fault of the panel. We had all gone in there expecting to receive advice, but from what I have learned since then, mastering is an art and a very individual process, so many mastering engineers are reluctant to give advice to a room of people with different ears and different projects with different purposes. The presentation was really about discussing the importance and value of mastering, it wasn’t a ‘how to’ session, like many of us had hoped.
Around the same time, I met Karl Machat of Mister’s Mastering House. Though I am keen to learn about mastering for simple home recordings that I can send out as freebies in my newsletter, no tips and tricks can replace the seasoned ears of a professional that an album needs, so I hired Karl to master my 2011 release [Lark] We Are All Born Naked.
In turn, Karl was kind enough to answer a few questions about the steps that I (and other independent artists) can follow to help improve the quality of our self-mastered home recordings and how to best prepare our tracks for sending in to a mastering engineer, if we choose to go that route.
Mastering Questions from Meghan Morrison:
- We all know that Mastering improves the quality of a recorded track/album, but what exactly is “mastering”?
Technically we should call it “pre-mastering” because mastering is what’s done at the CD plant when the glass masters are made. But it’s understood, so everyone just calls it mastering! Just to get that bit out of the way!
Mastering takes a collection of songs (eg. for an album) and makes them sound like they really belong together with each other – making sure levels, tone, and dynamics are even and consistent from song to song while keeping in mind the style of the tracks. Mastering aims for a balance between cohesiveness and individuality.
Mastering (when you hire someone to do it) is a chance to get a sonic critique of your mix before you unleash it to the world, by a fresh set of ears. It’s almost impossible to have a clear idea of how you sound after weeks and months of work, unless you don’t listen to it for, say, 6 months to a year!
Another consideration for mastering is making sure the genre is competitive with similar music on the marketplace. This is more of a concern when mastering just a single song.
There is no one correct way to master a track: it’s sometimes as much creative as technical.
Mastering techniques involve equalization, compression and limiting – a lot like mixing but generally more subtle in their range. Sometimes enhancing processes are used such as tubes and tape saturation. Other things include enhancing the stereo image, such as mid-side techniques - a speciality of mine. Removing the odd click, hum, buzz or hiss is also often needed. Editing is involved too: trimming the heads and tails of a song file, and creating the proper fade-outs and fade-ins.
The goal is to preserve the best elements of a mix. Mastering engineers do only what needs to be done. This is an important point to remember and understand: The better a mix, the less ‘mastering’ there will be in the end result.
However, there are times when an OK mix can be recovered in the mastering stage – but don’t bet on it!
I want to point out what mastering is not! It should not be looked at to fix or salvage anything. It can’t fix a bad mix, or poor recording technique. It can’t change the individual levels of instruments or voices, though sometimes – and only sometimes – you can create that impression with careful EQ, and maybe, multiband compression.
If I happen to get a mix that’s ‘borderline’ I’ll discuss it with my artist first. I will try to save them the time/cost of a remix whenever possible as I’m willing to do a little beyond what’s considered standard mastering practice if necessary to save a track.
When you get tracks mastered the artist does have a final say in the results BTW!
For artists who want to self-record, what tips can you give them to use during mixing so that when they send it to you to be mastered your work will be more effective or easier?
Use the best audio interface possible.
Use the best speaker monitors possible and position them correctly. The listening position and the two monitors should form an equilateral triangle.
Be familiar with your listening environment.
It can be very frustrating for the mastering engineer if the artist client is auditioning samples on an inferior playback system/environment and criticizing what might actually be a weakness in their own monitoring setup.
Also: your deadlines are – your deadlines. Give about 2-3 weeks planning for a full album mastering project. Be respectful of the mastering engineer’s schedule. Otherwise you’ll most likely be shown the exit door!
Listen carefully for distortion. This (like reverb) can not be taken away. Re-recording is the only option to correct this. Just because you’re not hitting 0 dB on your DAW’s meters doesn’t mean you might not have overloaded a component somewhere earlier in your recording chain: a preamp, for example; or maybe there’s a faulty patch cable.
Use high quality file formats: .wav or .aiff at 24 bits or higher. Use a minimum sampling rate of 44.1 kHz. Higher sampling rates are great..but(!)..if you’re going to a CD format use sampling rates in multiples of 44.1k. So 44.1 kHz; 88.2 kHz; or 176.4 kHz, etc. If you know you’re only ever going to a video medium then use 48 kHz, 96 kHz, etc.
Never, ever, use mp3 or any other data reduction format as a delivery medium: you’ll never recover what they take away and they sound awful anyway!
Get as much of the sound right at this stage as much as possible, rather than piling up effects in the mix stage.
Do – everything – you can to make your mix the way you want it to sound. Don’t decide later that the guitars or vocals weren’t loud enough and leave it to the mastering stage to try to fix it. Do not leave things like that for the mastering engineer to take care of. It adds to the time and, really, the final 2 track mixdown file is not the best place to repair things like that, even if it were possible.
Also don’t try to do any part of the mastering engineer’s job for whatever reason. I once had a client send mixes that were all maxed out level wise. They just wanted to make sure the EQ was consistent from track to track. Never mind the fact that if you’re EQ’ing after you’ve maxed out levels, you still need to limit again to make sure you’re not going over 0 dB, the additional problem here was the artist ended up creating distortion in every song because he couldn’t hear it on his monitoring system!
Use a high pass filter on every track.
- Do you mean each individual guitar, bass, vocal, etc. track? or on the master fader when bouncing each song?
I mean for each individual track/instrument in your mix – your guitars, bass, vocals, drums, etc. This removes the build up and accumulation of low level sonic mud which eats up precious loudness levels in the final stage. The technique for this is to start at 20 Hz and slowly move upward until you hear a loss of low frequencies. Stop at that point and lower slowly until you recover that frequency – that is the setting for that particular instrument.
Use low pass filters on every track. This helps with reducing the accumulation of noise levels from each track and also with any digital “edge-iness’. The technique is the same as above except in reverse: start at 20,000 Hz (20 kHz) and slowly move lower in frequency until you hear a loss of high end. Stop there and move up until you recover that loss.
- And how much of these low and high pass filters should we be using?
Use one for every track/instrument in your mix.
- Any recommended settings?
In both cases you need to use your ears to make sure you’re not removing any musical content you want to keep. The settings are determined by the frequency range of each instrument on each track.
Mute all and any areas that are not necessary for the mix. This helps minimize noise problems like hum, clicks or hiss.
Always check that your mix kicks ass in mono! :)
If you’re trying to emulate certain aspects of a commercial song, turn down the volume of that song in order to compare with your mix better. Commercial tracks are at a much hotter level than your final mix.
Avoid going overboard with effects: less is better. Use just what you need.
Take frequent breaks. Always listen with fresh ears. Avoid doing everything at one sitting or all night.
And again: don’t make the final mix loud. Give lots of headroom to work with: -3 to -6 dB is fine.
Avoid any effects/inserts on the master buss except maybe a good RMS meter.
- What is an RMS meter? and can you recommend a specific brand/product?
An RMS meter measures average levels. Their readings correspond more closely to the way the human ear hears. Aim for an RMS level from -20 to -14 dB and that should also take care of your peak levels not hitting 0 dB.
BrainworX, Blue Cat Audio, and Audio Pluggers are a just a few of the companies making metering software. I’ve ranked them according to my preference:
The people behind this dynamic range meter in the link below are part of a movement trying to regain dynamics and life back into music by asking professionals to not request such loud masters. It used to be free but now they’re asking for a donation. When I was on the site recently, though, there seemed to be in-appropriate or un-related remarks (spam?) left ‘un-attended’ in the comments section, so at the moment while I’m all for lower levels and more dynamic range, go to this site at your own risk:
There is a free meter available, but it has a vintage look so I don’t know how useful it will be to some people. When you use it, make sure you set it to the VU setting.
And once more: avoid hitting 0 dB on the final mix at all costs! That’s the biggie!
Why? Those clipped 0 dB points can and will be amplified during mastering and the result will be more distortion in your final results.
Creatively, keep the main elements of a mix – kick, bass, snare, voice – centred in a mix. Rhythm instruments should generally be centred but now you need to careful they don’t fight with the voice. (Remember about kicking ass in mono!)
Pan carefully – you want the left-to-right balance to be equal. Use a short delay (less than 25 ms) to send hard panned tracks to the opposite channel (at a lower, filtered level) to help balance and blend the stereo image better.
Also if you know the order you want your songs in, let the mastering engineer know from the start. One trick is to put the number of the track before the song title like this: 01 Song One; 02 Song Two: etc. Let him/her know the album title, the artist name, and the final titles for the songs.
Give your ISRC and UPC/EAN codes to the engineer. For ISRC codes you can apply here (for Canada):
Leave some silence at the start and end of your finished mix files. The mastering engineer can take care of trimming and fading. If you have a special fade, give him/her the start and end point of the fades.
Communicate with your engineer! Offer your thoughts, and be available when you’re needed to answer questions!
By giving the mastering engineer less to ‘fix’, you’re giving her/him more time to get ‘creative’ and actually concentrate on mastering.
- Most artists don’t have the time or desire to record/edit/master, but for those who ARE interested, where can they learn how to master? Do you teach people?
I don’t teach, but there is this online program:
Do you have any simple mastering tricks that artists can use to enhance the quality of the home recordings they may send to fans in newsletters or free downloads, but will never be pitched to labels, radio, or film (which you need professional mastering for).
In addition to the mixing tips:
I’m going to assume you want to know about the loudness aspect. Otherwise if you know you’re doing you’re own mastering, make the mix sound just like you want it to sound but without worrying about loudness, first, and render to a two track stereo file.
Then you can use some of these additional ideas.
And always use your ears as the final judge.
Parallel compression: it preserves the loud parts better and adds body and fullness to a mix when used correctly.
- How does one apply “parallel compression” … is it an option in a standard “compression” insert? where do I find it to use it?
Take your final 2-track mixdown and make a copy of it and line the copy up in your DAW directly in sync with, and underneath, the original. On the copy we’re going slap an enormous amount of compression: use a ratio of 2:1 or 2.5:1 at a threshold of -50 dB with the fastest possible attack time. Release time should be from 200 ms, to as much as 500 ms, or more. Mix this with your original un-processed file – the output level of the compressed file is the creative part. Be very careful that your original and compressed files are in perfect alignment or you’ll have major problems with phase cancellations.
You will need to apply limiting afterward simply to make sure you’re not getting ANY digital overs (going over 0 dB).
Most decently designed limiters and those in most major DAW’s can handle from 2-3 dB of limiting pretty transparently. A better option for mastering limiters is PSP’s Xenon or Voxengo’s Elephant. UAD’s Precision Limiter is excellent but does need an installed card. The new Ozone 5 Maximizer is amazing but like the Elephant, it does offer an almost bewildering amount of choices now.
(T-Racks 3 actually have a few decent presets that don’t suck and need minimal or sometimes no tweaking).
Limit the maximum peak output to -0.3 dB for .wav and.aiff files. For mp3 files and their equivalent, limit the maximum output to -2.0 dB. That’s because, when you render to low quality formats like that, they don’t “obey” the limiter settings and frequently go over the limit. The -2.0 db setting is a safety net to make sure you never hit 0 dB.
Also for mp3’s lowpass filter out to 15 kHz.
For RMS levels: I like to have my masters “hover” around -6 dB RMS for a maximum. That still leaves a bit of room for dynamics while still being competitive. Of course, that doesn’t mean you have stayed “pinned” at that level – if a song has quieter parts, let them be quiet.
Dither (don’t truncate) your final files after limiting to 16 bits.
For EQ try some of these ideas::
First though, because you’ve done the mixing yourself, if you find that you’re cutting/boosting more than 3 dB – STOP – go back and remix. You’ve done something wrong!
Cut out frequencies below 20 Hz if you’re doing a club mix. Otherwise it should often be safe to cut below 40 Hz BUT listen first! Like recording itself, there are no rules.
Human hearing is most sensitive in the 3 kHz – 6 kHz range, so unless you’ve already done so, a small boost of around 1 dB might help with loudness perception and maybe clarity as well
An alternative might be to cut around 1 dB in the region from 200 to 500 Hz. With one cut like this you can also think of it as boosting the bass and adding some clarity at the same time.
Before it gets too complex, some extra warmth might be had (if needed) by cutting one half to one dB at around 7-8 kHz; and finally one half to one dB boost around 14-15 kHz for some air.
It is very possible that all these suggestions could totally tank, too, because they just might be inappropriate for the mix. It is an ‘ear’ art!
Ultimately, one needs to hear the mix in question before making suggestions or recommendations!
It’s important to understand that it’s difficult or almost impossible to generalize because mastering is not done by formula. Every mix is different: what works great for one mix can be lousy in another. This is why a lot of mastering engineers won’t give their thoughts on what’s best to do because many people forget that each situation is unique. The big secret is there is no secret.
My Final Words:
The magic starts with the song. If that’s right, everything else falls into place!
Follow your inspiration and be true to it to the end!
- Thanks for doing this Karl! You rock! :)