It is one of the most common platitudes that you get to hear in a recording studio, “we’ll fix it in the mix“. But is it really a good idea to record your tracks and then to worry about fixing them, adding EQ, compression, noise reduction etc. later on? Of course not!
The saying should go, “record it right from the start and you save yourself a heap of trouble.” Yes, you should record your tracks as accurately as possible to avoid any fixing later on. To start our examination on how to avoid having to fix things in the mix, let’s examine the one big exception to this rule: Auto-Tune Plugin developed by Antares.
Let’s face it, when you are dealing with a vocalist who just can’t hit the right notes, there is not much you can do about it. You can try endless takes until you happen to get it right and indeed, that’s how it was done before Autotune or you can try to mask a bad vocal with a few backup singers, that can hit the notes correctly (also an old trick from the times of way back).
But now there is Auto-Tune. Please, turn that thing on when you have a vocalist that has problems with pitch. Better still, don’t tell them you used Autotune and compliment them on how wonderful they sang the song. That makes for a happy client. Of course, you will occasionally come across a totally tone-deaf person who thinks he or she can sing. Turn Autotune up to the max, they sound like a robot, the kids like it and they may sell a million copies of the song.
Now that we got the exception out of the way, let’s look at the three basic areas where you may be leaning toward “fixing it in the mix”
Number One – Background Noise
That is a common problem that can have causes ranging from a noisy air-conditioner during recording to using a noisy sample that was not recorded properly. You may be tempted to think that, if the noise level is kind of low it’s going to be masked by all the other tracks anyway. This is based on a big misconception, the concept of a noise floor.
Let’s say that the noise floor is -55dB, meaning that that rest of the sounds is 55 dB louder than the background noise. The reality is that we are not really faced with a “noise floor” but a “noise blanket”. Yes, the noise is 55dB lower than the main sound you are trying to capture but it really covers the sound like a blanket.
Try it yourself. Take a recording with a good amount of background noise and then use a noise gate on that recording. Yes, with the proper setting, the noise disappears from the silent sections of the recording and of course, that helps, but when you listen to the parts that the noise gate let through, there is the same background noise, sitting on top of the sound you actually wanted to hear and it is definitely altering the sound quality of the recording.
Big misconception! You cannot get rid of noise with a noise gate, just clean up the quiet passages. And cutting the high end with an EQ, the result gets even worse as you now really have killed all the sonic qualities of your recording. Let’s move on to the next subject…
Number Two – EQ
Well, you always have to EQ your tracks when you are mixing, don’t you? No, not necessarily. I always prefer to record a track in such a way that it needs little, or even better no EQ at all in the final mix.
I have done mixes with 32 tracks and not one track had any EQ added in the final mix. It’s a matter of recording things the way you want them to sound in the first place. Or choosing the sounds, loops or samples that sound right for the song. Don’t choose a drum sample that needs a bunch of low EQ to sound right, choose a drum sample that already has all the bass that you need.
The reason? Any EQ that you add to a track distorts or twists the sound to some degree. There are some very nice sounding EQs found in recording mixers and/or software plugins but no matter how pleasing the quality of the EQ is, that effect is achieved by twisting the waveform to make it sound the way you want to. Multiply that by 20 tracks and there is a whole lot of twisting going on.
I will have an article about EQs in the future and they are some of the most essential things in your recording setup but, if possible, I would always prefer to work with a track that needs no EQ at all in the final mix.
Number Three – Compression
There will always be an endless debate amongst engineers about compression. Everybody has his own “secret” method of compressing. Some will only compress vocals, some will only compress certain instruments, some will compress almost everything. But the discussion here is about using compression to “fix it in the mix”.
So, should you compress instruments or vocals during recording? You may be tempted to say that with so much dynamic range in today’s digital recording setups and such wonderful sounding compressor plugins, why would you?
First of all, that precious dynamic range that you think you have, may not quite stretch as far as you thought. This is where the issue of noise floor comes in again. Secondly, compressing vocals while you are recording makes for much more pleasing tracks that can give the vocalists more confidence as they don’t have to be too concerned about controlling their voice levels while sining.
And if you have one of those wonderful vintage hardware compressors, that analog magic may add an edge to the sound that you’d never expected. So, my advice is to record your vocal tracks as close as possible to the final result that you want to achieve for that vocal track. And a little compression during recording can help with that.
Another area where compression may better be applied while recording is bass guitar or guitar. Instant compression gives the sound a very different characteristic and a bass or guitar player will actually adjust their playing style to that unique character of a compressed bass or guitar.
For the rest, there are many instances where you want or need to compress a track afterward but please consider compressing some tracks right when you record them so you don’t have to “fix it in the mix”.
Also, you can read: Diplomacy, Torture and Self-Torture in the Studio