The importance of quality at the source

So two things struck me today, until I realised they were actually one thing in two places, my work and my audio. That one thing is the concept of the quality of data at the source.

I used to think that even the grimmest of sounds could be polished into something beautiful in the studio with the various tools and whatnot. I used to scoff at the idea that you had to capture the "quality" performance or the performance with the "best feel to it". My problem was that I felt that that view was restrictive to further processing, chopping and whatever else I wanted to do with the sound later on down the line. In short, it felt like an "old" way of looking at the matter. In actual fact it makes just as much sense for me as it does for someone who intends to leave those takes perfectly alone.

Getting a quality source for a sound will make processing it easier down the line because it gives you more options and more freedom to play with it. A beautiful vocal take will sound better whether it is an opera piece sung from start to finish, and never touched apart from subtle eq, to a screamer that you are going to chop up and make sound like a robot. Better in => better out, whatever you are doing. So in this sense, it really is worth getting the best quality you can at source. And im not talking about gear here, im talking about technique, performance, timing etc. All those age old things that make good recordings from yesteryear sound great, they still make a good recording now. So maybe its worth tweaking that synth on its own at source just a little, or tuning up that guitar, or doing that take again to get it just right. Sure you could touch it up after, but wouldnt that time be much better spent knowing you have a great take and then making it sound like something more creative to you?

It is worth practicing, it is worth playing well and it is worth putting those extra takes and time into each part of what you are doing, regardless of what you intend to do with it afterwards. It doesnt make you some kind of virtuoso, its just a quicker way of getting a better end result than trying to fudge it afterwards (something you will still have to do sometimes, but generally is best kept to a minimum).

Now for the other time this popped up in my life recently, at work. I work on a system where very often large extrapolations and assumptions about data have their rooting in single tables or data points. The point being that often small errors in this original source can snowball, very quickly, into much larger errors from something that at first seemed in no way dangerous to the extent it is.

This is the same issue. It doesnt matter how much work you do with something down the line, its worth doing the leg work on where you get your source from, to begin with.

I guess the moral here is "take it and break it" as always but isn't it more fun to break something truly great? The pieces you get back tend to be so much more interesting that way.