Production Expert

View Original

Are We Over-Engineering Our Audio Productions?

In this article Julian dares to consider using less gear with fewer features to get the job done. Considering the extraordinary capabilities of even the humblest studio, why wouldn’t you squeeze every last bit of potential out of your hardware and software?

I recently wrote an article listing some essential plug-ins that I use and something that struck me was the mix of plug-ins in that list. Some were at the cutting edge of what is currently possible, but others were 20 years old with limited features compared to current offers, but I still favour them. This made me wonder, do we over-engineer our solutions? In short, do we place too much value on features we don't actually use that much, and having these possibilities. Do we do too much because we can?

A 20 Year Old Plug-in

The example, from the article I'm thinking of is my use of Avid’s EQ III over more capable equalisers. I have FabFilter’s Pro-Q 3 and while I use it regularly, when I need a utility EQ I probably only use it 10% of the time. The rest of the time I'm using EQ III - Why?

The main reason is that 90% of the time I don't need the extra features that Pro Q3 offers. Of course Pro-Q 3 does everything that EQ III does, and if I hadn't been using EQ III happily for many years before I tried the FabFilter EQ I would probably be using Pro Q3 all of the time, I totally get that. But we shouldn't assume that more necessarily equals better, especially if we have features we don’t use.

Keeping things simple is rarely a bad idea but I know many of us, myself included, infer a status onto solutions which are complex, powerful or ‘advanced’. 20 mics on a drum kit, 4 mics on a guitar cab, very tightly comped vocals, manipulating timing and pitch. All of these things feed a narrative that more control is better and that people who use more involved solutions, using more resources and more feature-rich equipment are more ‘professional’  - a word which itself is used to mean variety of things. But however interesting individual examples might be, does this attitude cost us more than it benefits us?

Garageband Recordings

My very first recordings were made of bands I was in my very early teens. We got a cassette recorder with a built-in microphone and recorded our practices. These recordings were appalling and we were too loud for the fixed gain mic, resulting in very distorted recordings. We figured out that if we put the tape recorder outside the room, it was quiet enough not to distort the tape, but the sound was reverberant and distant. The next step was to get hold of a cassette deck with external mics and variable gain. With more control, we got better recordings.

A cassette recorder very like this has a lot to answer for…

The next step was a four track cassette machine. The ability to have different things on different tracks and to overdub was revolutionary. Next came an 8 track reel to reel with a mixer and outboard effects. Even more control and results that starting to sound like records. Next was a 16 track hard disk recorder. More tracks meant yet more control and no longer having to submix drums to a handful of available tracks, and the ability to bounce without degradation meant pretty complex recordings were possible. Finally, a DAW with plug-ins brought a level of control which that teenager putting a tape recorder on the floor halfway down the passageway wouldn't have thought possible.

It was at this stage that many of the last remaining restrictions fell away. If I wanted to insert a compressor on every channel in my mix I could. I always wanted more compressors in the hardware days, but I found the temptation to over-compress too much to resist. Listening back to mixes I made in those days illustrates this clearly…

It's only when you gain the ability to record everything to its own track that you really become aware of spill. If you are recording using multiple mics but premixed to limited tracks, as we used to in the days of tape machines, you don't have the opportunity to hear spill exposed in quite the same way. What I wanted in those days was to be able to raise a fader and only hear the thing that was supposed to be on that track - Chasing this minimum spill ideal that I thought I wanted. I finally achieved it on a band project over 20 years ago, I was extremely pleased with the results after the recording session, with almost no spill anywhere apart from the overheads. 

This is very much a case of be careful what you wish for because everyone, myself included, preferred the earlier sessions with far more spill because they just felt better. They had more vibe. As someone once said “the music is in the rough edges”, and I think I'd managed to squeeze all of the music out of the music.

Good Enough?

Control isn't the goal, it's only the means by which we fix problems. And while somebody describing something as ‘good enough’ might sound like the kind of thing a slapdash engineer doing a half-assed job would say, ‘good enough’ also means that something is sufficient to achieve the goal. There is a balance to be struck here.

There are lots of examples where we might be over-engineering our solutions. We’d all agree that a slow computer is a pain to live with but it doesn’t follow that there is a straight line relationship between how powerful our computers are and how much benefit we get from them. I bought an M2 Mac to replace my ageing Mac Mini but I still do a surprising amount of work on the Mini because its fast enough for these jobs. I don’t need a sports car to go to the shops. A fast computer is great to have but if your computer is fast enough for the work you do, then it’s fast enough. As long as its still supported there’s no problem until you say there is. The availability of faster computers isn’t relevant until that point.

Similarly, so many of us don’t consider the lower tiers of software products. I’ve had several conversations with people who are still of the opinion that they need Pro Tools Ultimate because it's the top tier, and admittedly in the old days of only two tiers they probably did need Ultimate. But these days the majority of people working in music I speak to, unless they are using HDX hardware and need Digilink, only need Pro Tools Studio. We've commented before about how more people than probably realise it could do the work they do using Pro Tools Artist.

Just because we can do something in a way which is closer to what we understand to be a professional workflow doesn't mean that we necessarily should. I don’t need solutions to problems I personally don’t have! Much of the reason why Final Cut Pro was so popular and became such an attractive alternative to Media Composer for certain parts of that industry is that it was appropriate to the kind of work a large part of the industry were doing.

The Right Tool For The Job

Russ has moved to doing his video work entirely within Da Vinci Resolve, including audio, rather than doing his audio work in Pro Tools precisely because for him doing the work he does it's a more appropriate solution regardless of how larger projects are completed. At the scale he is working is the right solution for him.

Taking this video example even further, I do very little video work beyond screen capture but when I do edit camera footage I've pretty much abandoned Final Cut Pro in favour of cutting video in Screenflow just because I know it so much better than Final Cut Pro. And while if I need the sophisticated features of Final Cut Pro I have them available to me, for the kind of work I'm talking about cutting in Screenfow is simple and, to resurrect that phrase once again, it's ‘good enough’.

Often when I write, I write into TextEdit. I do this for the same reasons - it's good enough. What I'm writing in, doesn't affect the content of what I've written and, in so many cases, the tool we choose to use is there to represent our ideas, not to influence them. They are not the point of the exercise, they are the means by which the exercise is realised.

What do you think? Do you choose to use simpler solutions when more feature-rich alternatives exist? Do you see unused additional features as potential, or clutter? Share your thoughts in the comments.

See this gallery in the original post