Audio Toyes #6 - the Mix
January 25, 2011
ďOne commonality between all great mixes is found in the relationship between the Kick, Snare, Bass and Vocal. Get these 4 elements pounding and sitting right and its hard to mess up the rest...no matter what genreĒ - Brian Toye
Mixing can take some engineers a few hours and some a few days. It is always beneficial to have an idea how the tracks were recorded (microphones/mic pre's/any pre tape EQ or dynamics) and the recording engineer will usually send such notes along with the the audio files to the mixing engineer. It is also very important the mixing engineer be very familiar with the equipment he will be employing throughout the mix. This is why we see so many similar pieces of gear in all the professional studios. This way any mixing engineer can lease out any room all over the continent and expect to use the same pieces of gear he is used to using. With that said I have also heard absolute CRAP come out of some of the worlds best studios and some pretty AMAZING things come out of someones home Pro Tools HD rig. So the equation for a great sounding product includes an engineer with good ears + familiar with his gear + has pro gear + is willing to spend the time your mix deserves.
I create an AUX for each effect I will employ in the song, usually vocal VERB, drum VERB, vocal DELAY, guitar DELAY. Named as such. Now in PT 9 you can route tracks to auxillary tracks and through sends with a few clicks of the mouse. The old way took some clicking around and could only be done one track at a time.
I then pull up each musical element individually and sweep each for the sweet frequencies I am looking for (that low end boom on kick, the higher freq snap of the beater, the pop of the snare skin and the lower freq body of the snare sustain).
I write down all the frequencies I find by sweeping around and locating the sweet spots within each track.
I then pull up the whole mix minus the vocals. As a starting point I will attenuate the frequencies of every track where I found the sweet spots in other tracks. If this is not clear here is an example:
Say I wrote down 350hz and 1000hz for the snare. I then attenuate these two frequencies from ALL other tracks. Use your ear to determine how much. You obviously donít want to make it noticeable. Always use a high Q when attenuating so you are only pulling out the desired frequency. Then, If your sweet spot on the acoustic guitar tracks were 200hz and 1155hz you would then attenuate the BASS, DRUMS, MIDI strings etc.. all other tracks at these frequencies. Once this has been done for all the frequencies you have written on your reference sheet go ahead and BOOST those sweet spots in the appropriate tracks, use a wide Q for boosting frequencies. Iím talking about 2.5 - 5 db here donít go crazy.
Once complete, sit back and listen to the track at a low volume level. At a medium volume level. At a loud volume level. Can you hear all the musical elements? If so move on, if not tweak the EQ so everything sits in its desired frequency range.
Now I like to begin using my stereo field. I will pan everything to its desired location. Pretty simple.
Now I begin using dynamics. I find its best to keep a high threshold so that the compressor only catches the highest of peaks at first. Use a high ratio and quick attack, medium release. For example: Slap such a compressor across the bass track, then create and arm a new track and record this compressed bass to the new track. Hide and make inactive the original bass track. You can now see your bass waveform is leveled out nicely compared to the original giving you more headroom to mix and having not really changed/altered its sound. I do this for bass, guitars, and at this point also convert my midi tracks to audio tracks.
You might be thinking. Why did you EQ and then compress. Since compression would bring up in volume those frequencies we attenuated relative to the frequencies we boosted. In other words closing the gap of relative loudness between them. Your right compression will do this a tiny bit. This is why you can (if u think necessary) save the EQ settings and re apply them on your newly recorded compressed tracks again. This is where you will REALLY hear your track clean up. Apply compression to the new tracks, use to taste. Set relative levels/volumes and now we are ready to listen to those vocals.
SOLO the lead vocal along with the click and guitar track. If there are pitch or rhythmic problems I suggest using melodyne to fix both of these things. If you have questions on how I employ melodyne please PM me.
Once these edits are complete we are going to copy the whole main vocal track to another new track. This new track will act as our "vocal exciter". I slap an EQ and Limiter across this copied main vocal. I attenuate everything I can below 4.5k and boost everything I can above 5.5k (-12db below, +12db above). This results in a very harsh vocal that can cut your skull open so donít playback loud (WARNING). Use this track to taste by slowly bringing it up and mixing it with the original vocal track just to the point where if you mute the "excited" track you can hear the difference. This will help your vocal punch through the mix and be right up front.
The main vocal track usually gets a DE-esser (-10db over 6.5k) and EQ (boost around 2.5k and attenuate below 100hz) and a compressor set to levels similar to the Pro Tools pre set vocal leveler (med attack, med release low threshold, ratio of 3:1 and as much gain boost as needed to bring back to the original input level (or a bit more if you like, just watch for clipping)
At this point I bring up all the instruments and make some fine adjustments usually for the kick drum and bass guitar making sure they are both sitting in their own space. Make sure the snare is not interfering with the vocals. Listen on a few different sources (speakers) to make sure your high hats are not too bright. I mix all levels to taste and move onto FX.
Now I apply my sends to all FX. I use a different reverb on instruments and vocals. I use a different delay on guitars and vocals, but usually apply the same delay times based off the tempo of the song (PM me if you want more info on getting your delays to rhythmically follow the tempo). Be aware that sending a signal to an effect and bringing that effect into the mix will in a sense make that instrument louder, so be gentle with effects and pull down the main dry track a dB if needed.
At this point I usually bounce a MIX1 file, save it to a USB stick and go out to my car stereo. Stick in the USB and listen. Make mental notes and return to the studio. I make fine volume adjustments and play with panning my FX sends if necessary. I continue this process Listening on my Tannoy Monitors, Enclosed AKG headphones, and Car stereo (after market, no sub). After about 3 bounces my mix is pretty much ready for the mastering engineer. I make sure to leave at least 3db headroom when sending it to be mastered so the mastering engineer can work some magic. If i am not mastering the track (not in budget) then I will strap an EQ and LA 2A or something across the 2MIX to level out the overall frequency range and boost the volume to just below peaking.
One great Mastering trick to get rid of uneven bass notes is to locate the bass note that sounds overpowering relative to the rest. Determine the frequency of that note. Attenuate by 5-6 dB with a medium Q. BAM now all the bass notes sound level with each other.
Thanks for reading, again remember these are just some guidelines and things to try out at your home studio. Always trust your ears. Next week I'm going to talk about recording electric guitars.