I've been having a lot of fun with our new Mix Rescue series, not least because I get to spend more time in the studio and less time typing! One thing this series has really brought home is that without accurate monitors and a reasonable acoustic environment in which to use them, you've really no idea how your mix will sound once you unleash it upon an unsuspecting world. When it comes to mastering your monitoring accuracy needs to be better still — even my little studio probably isn't up to anything too serious on the mastering front. Anyone can have a great time trying out various mixing and mastering techniques though — and you'll certainly learn a lot more by doing this than you will as a passive student in a lecture.
Another thing I've realised is that human voices and acoustic instruments are comparatively straightforward to deal with because you know how they should sound. You can either fix them up to sound the way you want them or fail because of readily identifiable problems with the source material. Sometimes you can only rescue so much, but the tools of salvage are getting better all the time. We're also pretty familiar with how drum kits should sound on records, but where things get tricky is when somebody sends you a mix with electric guitar and bass sounds, because there is no definitive 'right sound' to aim for. The electric guitar sound is a mixture of the instrument, the amplifier and the playing style of the performer, and making it sit correctly with the rest of the mix when the sound isn't working properly to begin with can be quite a challenge. This is certainly the case when the guitar part has been recorded with more overdrive or distortion than the track needs as you can't really undo these effects.
In such cases, it can be useful to create your own guitar sound that you think works, record a short section, then use a fingerprint EQ type of plug-in to make the spectrum of the problem guitar part match that of what you are playing. This won't always produce the desired result right off, but there are often pleasant and unexpected surprises along the way, and in any event, looking at the audio spectrum and the shape of the EQ curve being applied can teach you a lot. I've used this tool on several occasions by setting up my own EQ correction curve using a conventional multi-band, parametric equaliser based on the shape of the correction curve the fingerprint EQ has come up with. While some engineers know exactly how to treat any sound, some repair jobs require very complex EQ curves that often aren't easy to arrive at by intuition alone.
The same dilemma also presents itself when you come to work with synth sounds, because again there is no 'real' sound to compare it with. Usually it is better to craft the right sound at source, but if you're presented with a finished recording, sometimes you just have to reshape what you have using EQ and other processes. Often it's simply a matter of rolling off excessive highs and/or lows to make the sound sit in a narrower part of the audio spectrum to prevent it conflicting with other sounds, but at other times you have to get really creative to make things work.
The proliferation of broadband Internet has enabled us to put more audio examples of mix processing on the Sound On Sound web site, making it much easier for you to get an idea of what we've achieved, and in some cases it also shows the limits of what can be done without recording new or replacement parts. I hope that you're finding the Mix Rescue series useful, but I should really thank you, because since embarking upon this venture, I've also learned a lot of things that I might never otherwise have come across.
Paul White Editor In Chief