D.I.Y. Mixing… Well Almost
Blog/April 7, 2016
The trend towards D.I.Y. mixing, followed by ‘a bit of polishing’ from an audio engineer, is growing rapidly. But is this two-stage process really worth it? Does ‘doing it yourself’ really save you time and money or only cause grief, as you later try to ‘fix’ the mixes with a professional?
I see it a lot: a band has an album to mix that they may or may not have recorded themselves, and after some contemplation one of the members decides to mix it on their own.
Fair enough. If you know what you want your album to sound like and you think you can achieve that without any outside involvement, by all means, go for it.
But what if your mixing skills aren’t quite up to the challenge? What happens when, after weeks of wrangling the DAW session trying to land the mixes, you hit a wall, unable to find solutions to your problems?
That’s when most people call in a professional to (hopefully) fix their mixes.
But in some respects this is false economy. After reaching a mix impasse, often creating messy session files along the way, it’s not necessarily simple or time-efficient to expect (often for a pittance) someone to come in and resolve the mixes for you. Often it’s more time consuming and costly to decipher an average mix file than simply starting again from scratch.
But when you’ve put in all that hard work you don’t really want someone telling you it would be better to just scrap the mixes and start again, do you? Therein lies the problem.
These hybridised two-stage mixes are often beset with technical and emotional dilemmas as you try to preserve and protect the work you’ve done, even though it presumably hasn’t produced great results. So why do it?
‘NO-ONE UNDERSTANDS MY MUSIC BETTER THAN ME’
There are lots of reasons why people choose to mix their own music. Some just like the challenge. Others do it because they anticipate it’s far easier to mix the work themselves than elucidate their vision to someone else. Then there’s the semi-professional or professional engineering brigade who possess more than enough skill to work on their own stuff.
But the vast majority of people who opt for the D.I.Y. approach simply can’t afford to spend thousands of dollars getting someone else to do it for them – all fair enough and perfectly reasonable scenarios of course.
So if you’re heading down this path, expecting to do ‘most’ of the mixing yourself, perhaps only getting a professional involved to tinker around the edges with tone and levels before ‘signing off’ on the mixes, here are a few things to consider.
Lost in translation: Don’t expect things to translate easily if you’re transferring a complex mix session file into another computer. Most engineers like to use their own computer, but very few will always have every single plug-in you used in your sessions. Things like soft-synths and MIDI instruments may be rendered mute (or even causes crashes) in another computer.
If you’re expecting to get straight to work ‘finalising’ your mixes with someone else on their ‘foreign’ computer, where possible, convert all your MIDI instruments into audio tracks by bouncing them out, remembering to make sure the original tracks are inactive (or absent altogether) from the clearly labelled fresh session file you create specifically for the task. Alternatively, make sure you have a discussion with the engineer well in advance of the session about what they may or may not have onboard their system.
To stem or not to stem: This is possibly the trickiest question to answer in relation to the hybridised mix process. To the unfamiliar, a ‘stem’ is simply a single or group of processed (or unprocessed) channels sub-mixed together.
A stem mix session file might consist of stereo audio channels of say all the guitars (processed), the main vocal (dry), its effects in stereo (maybe), drums in stereo (and maybe kick and snare on their own), bass on its own, keyboards in stereo (maybe), percussion in stereo and so on…
If that all sounds pretty messy and vague already, you’d be right.
In short, stems are a total can of worms with the main dilemma always centring around questions like: “What should I group together as a stereo stem, what should I leave separate, what should I edit or leave for the engineer to edit, which channels should I leave plug-ins on, which MIDI instruments should I print out?” and so on.
It’s all about anticipating what ought to be ‘locked out’ of the final mixing session, and which ingredients should remain ‘up for grabs’.
Some would argue that stems are the way to go because they remove all the issues pertaining to plug-in and soft-instrument translation, and take out all the so-called secondary roles of editing and file cleanup that might otherwise bog down the session or tempt the engineer with too much scope for change. That way, things presumably move faster and more concisely, cost less, and prevent the possibility of a remix.
Presenting the mix engineer with an ‘almost finished’ stems mix file means they can tweak but not really reinvent things.
Problem is, deciding what’s locked off and up for grabs about the ingredients of a mix is fraught at best, particularly when the person making these decisions is typically the same individual who couldn’t land the mix in the first place. As a result, often more time is wasted trying to get around the limitations of the stems than would otherwise have been spent manipulating the original multitrack file.
This goes to the heart of the matter. Stem mixing is a compromise most of the time. Personally, I would argue that it’s largely an ineffective, hamstrung process. But if it’s unavoidable remember this: less control means less effective change. If you want the engineer to have more control, offer them a more open session where things like fades are long (or even raw), most instruments aren’t grouped and things aren’t printed with their effects, especially the vocal. If you love your mix, convinced that it’s excellent but just needs a few levels checked, by all means offer the engineer less scope for change.
Working for peanuts: Then there’s the issue of payment…
If money – or indeed a lack of funds – is the main driving force behind doing the bulk of the work yourself, don’t let the mixes get to the stage where you’re desperate for professional help, then go cap-in-hand to a mix engineer with virtually no budget. Remember, it’s not the engineer’s fault you have mix problems and very little funding with which to resolve them. Music is an emotional experience and sometimes this can cloud your judgement about what’s fair and reasonable with respect to employing others on your project.
Some but not others: The other issue to consider is whether or not to involve an engineer in some, but not all, the music on your album. This can have the desired effect of making the mix process cost less but perhaps not the album overall. By this I mean, inconsistency and multiple-environment mixing can have the knock-on effect of adding more time and therefore cost to the mastering process. It can also create undesirable inconsistency across the tracks, where some songs on the album sound finished while others seem relatively underdone.
In many respects you’re better off giving all the songs to a mix engineer, even if that person only spends a small amount of time on the less critical tracks.
Working on all, rather than just some, improves everyone’s ability to make clear judgements.
And finally… if you think you’re immune from making the wrong judgement call about whether to take a ‘full’ or ‘stem’ mix session file to an engineer, think again.
The best thing you can do to prevent this potentially grievous error is communicate thoroughly with the engineer about what you anticipate the problems to be, and solve them before you walk through their door.