Archives for November 2010

It takes a bit of skill and the right knowledge to record audio that sounds good.  And the knowledge I refer to is really quite basic, though I was having a hard time coming up with a good metaphor to explain it without having to use overly-technical terminology.  So I asked my wife to help me out, and she likened audio to coffee.  The more we drew comparisons, the better the analogy between audio and coffee seemed to work.  So I’m trying this theory out on my unsuspecting readers.  I hope that you will let me know (via the comment form below) if this fails spectacularly.  So without further ado, I give you the coffee-recording analogy.
Just like it isn’t that easy to make a really good quality audio recording, it isn’t that easy to make a great cup of coffee.  Anyone can make a bad cup of coffee…even a so-so one.  In fact a quite large number of people do it every single day.  The same is true for audio recording (a quick visit to YouTube will confirm this).  But both tasks seem so simple.  Pour hot water over a bunch of crushed up beans and drink what comes out of the filter; or make some noise into a microphone hooked up to a recorder.  Now just to make this unusual comparison complete, I make this additional observation: Just as you can still make a bad cup of coffee even with an expensive brand of beans, you can make a crappy recording with a pricey microphone.
Is the reverse case also true?  Can someone make a good cup of coffee from lousy beans?  Or can someone make decent sounding audio with a plastic microphone and a computer?  The answer to that is only up to a certain point.  The folks who have the savvy and experience will make the coffee/audio as good as it can possibly be with the materials provided; probably surprising the average Joe (no pun intended) with the level of quality that CAN be obtained with such low-end ingredients.  These same people could make FAR better quality products with higher quality (usually more expensive) materials, to be sure.  What is most important about this last bit is that even that average Joe can learn the all-important fundamental skills of quality coffee/audio production on the cheap materials with cheap ingredients.  Once they have the proper skills and awareness of key principles, they can make  not-so-bad coffee/recordings out of the bad materials.  That is the time for these people to upgrade to better equipment and ingredients.  Before they know it, they’ll be making gourmet/professional coffee/recordings from just average gear.
Now you may think this metaphor has run its course, but actually I haven’t even gotten to the comparison that started the whole thing.  All that stuff above was just bonus analogy;).  Let’s talk about what I think (some other experts may disagree) is the single most important piece of knowledge you can use to make the best sounding audio recording with the equipment you have.  The reason I asked my wife to help me out here is that I really didn’t want to have to use the somewhat scary term gain-staging.  Like a lot of tech-y stuff, the term is not descriptive to the uninitiated.  So this whole coffee thing is an attempt to explain why gain-staging is the most crucial element to good audio.
Okay, here we go.  Let’s say you use a drip-coffee maker.  How could you make a bad cup of coffee?  Given that there is an ideal mix of beans-to-water you could screw it up in two major ways (and any number of minor ways that we’ll ignore for the now).  You could make the coffee too strong with too high a bean-to-water ratio.  Likewise, you could make it too weak with too low a ratio.  Either way, once you’ve screwed it up it really can’t be fixed once the coffee is in the pot.  You can’t just add water to too-strong coffee to get the concentration right.  Some people try to cover up the bad taste by adding more cream and/or sugar, but it’ll never really be “good.”
And I don’t even KNOW what to try to do with really weak coffee once it’s made. It can’t be good.  If someone were really determined, I imagine they could try to drink from a cup that is double or triple the normal size in an attempt to get the same amount of caffeine as usual.  But they’d have to drink double or triple the amount of vaguely coffee-flavored water to do it.  Or I suppose really desperate folks might try to boil off the excess water to get stronger coffee.  Seems like a lot of effort to go to when the result will be iffy, at best.
If, when making the coffee, the amount water is always the same (let’s say you always fill it to the bottom of the metal ring at the top), then the only way to mess up is with too much or too little dry coffee in the filter.
Now it’s time to make the mental switch over to audio recording.  Let’s say the volume of your voice (our audio example for today) is like the amount of coffee.  If the “amount” of your voice in the recorder (usually a computer sound card these days) is too high, it will end up sounding distorted once it gets through the sound card and onto disk (like coffee in the pot after passing through the filter).  You can NOT just turn the volume down to get rid of the distortion AFTER it’s been recorded, the damage has been done.  Sure, you can try to add lots of cream and sugar…oh wait, I mean reverb and echo and other audio to cover up the distortion, but you cannot fix it.
Now what if your voice is too low in volume when it hits the sound card?  Well, the card is going to pass the same amount of stuff through to the disk, just as the coffee filter is going to pass the same amount of water through the filter, regardless of how much or how little coffee is in the basket.  So if the sound card doesn’t have enough voice to work with, it will just record lots of noise along with the little bit of voice it hears.  The only way to make the voice audible after it’s recorded would be to turn the volume up.  But when you do that, you’re also cranking up all that noise (hissing, crackling, etc.) as well.  So you end up with noisy audio, just like weak and watery coffee.  At this point you can try to fix the recording by using noise reduction tools.  But this is akin to trying to reduce weak coffee by boiling off the water.  Even the best noise reduction tools can only do so much.  If there is a lot of noise mixed in with the voice, noise reduction will reduce the noise, but at the expense of the audio quality.  You’re usually left with a swirly, under-water-y sounding voice after such an exercise.
So the solution (no pun intended) to both coffee and the audio recording quality is to make sure the amount of coffee and voice loudness is not too high, but also not too low, in the process.  For the coffee, use a measure and be very precise about it.  For audio, use the meters found either in the mixing software (Windows mixer, for example), or even better, use your ears and eyes.  The advantage we have with computer audio recording is that we can both hear and see what the level is before committing to it.  The loudest part should never “go into the red,” while at the same time, there should be enough voice volume that the quietest part can still be heard.  Get it (and by “it” I mean the thing you’re recording…voice or  music, etc.) as loud as you can without any part of it distorting and you’ll have cracked the code for getting the best sounding audio possible on whatever gear you are working with.  To get some guidance and tutorials on how to do some of this stuff, click here: home recording tips.
Now go make some coffee.  For some reason I want some now.
Ken