Did your mastering engineer tell you?

Once mixed, your songs will sound best after the mastering process if:

They are well balanced dynamically:
Make sure it sounds like nothing in the mix is compressed more than anything else to the extent that it sounds unnatural. For example, take the relationship between vocals and the rest of the instruments - vocals should not be so dynamic in contrast to the instruments so as to make it sound like you are listening to a live vocalist singing over a compressed backing track of instruments. Another example is the relationship between vocal 'ess' sounds and the hihats or cymbals - make sure that the relationship between them is such that the mastering engineer will not be tempted to turn one of them down at the expense of the other.

They are well balanced sonically:
When soloing instuments on the mixer each individual instruments EQ should not be radically different to how the instrument sounds in real life. This is especially true of vocals. Be careful to fill the mixes with a broad spectrum of frequncies as contributed naturally by the instruments in the mix.

They are well balanced from an ambient or spacial perspective:
Avoid too much reverb and use it to create a sense of depth to the music by applying more reverb to those instruments or suonds you want to appear further back in the sound field or 'sound-stage'. Avoid putting a lot of reverb on vocals while keeping drums dry for example - unless of course you are specifically trying progressive production ideas.

Individual tracks (like a tom-mic channel for example) are muted when the instrument it is intended to capture is not being played at the time:
The cumulative noise from numerous open channels adds unecessary noise and ambience (reverb) to a mix. Rather than simply gating individual channels, go through every track and delete sections of the waveform where there is no signal from the intended source.

They are well-managed in terms of phase-relationship between the left and right channels:
Badly managed recordings or mixes can result in your left and right speakers cancelling out each others respective frequencies making the music sound thin and oily. It will also have adverse effects on FM transmission of the song causing it to break up on the edges of the transmitter's footprint.

They are not limited or compressed at the final master bus:
Leave this to the mastering engineer, making your music louder while preserving important musical detail is his speciality and he should have invested unreasonable amounts of money on equipment to do it really well.

They do not peak over 0dBfs anywhere in the track:
There is nothing worse than digital distortion to make mixes fatigueing and irritating. Avoid it like the plague.

They are tracked, mixed and bounced at high word length and sample rates:
Long word lengths (24 bit as opposed to 16 bit) leave more detail for digital processors to do mathematical calculations on. There are less mathematical 'loose-ends' so-to-speak. Higher sample rates move audible distrotion from digital equipment filters higher up the spectrum where it becomes less and less audible (see the section below on 96kHz sample rates)

They are burned on good optical media at optimal write speeds:
This is where you should spend the money on a good brand of CD-R for the purpose of delivering your master to the mastering engineer. Taiyo Yuden is an excellent and highly recommended brand of CD-R. Make sure you buy A-grade CD-Rs and burn the pre-master onto the disc at nothing higher than 16 speed to minimise write errors. Don't burn too low either since modern CD-Rs are optimised for writing at higher speeds and might not benefit from very low speed writing.

----------------------------------------------------------------------------------------------------------------------------------------------------------------

Monitor levels: Interpreting the Fletcher-Munson Curve

Fletcher and Munson did research into the response of the human ear to a wide range of frequencies over a wide range of listening volumes. They came up with a set of graphs based on feedback from people who were asked to report when two different frequencies were perceived to be at the same volume. They found that the ear does not hear all frequancies equally well at different listening levels. The graphs show that the ear is less sensetive to bass and high frequencies at lower listening levels and is always most sensetive to frequencies between 1kHz and 6kHz (vocal frequencies). What this means for audio professionals is that consistent results can be achieved (particularly when adjusting EQ) by monitoring at a consistent listening levels and at a level where your judgement is least affected by the ear's frequency response. 83dB is reported to be the ideal listening volume. This is quite convenient since you can listen for fairly long periods without damaging your ears.

You can check your listening levels using a standard SPL meter. A convenient SPL meter is available from Radioshack.com

Once you have found a level close to 83dB, make a note of your monitor level settings on your sound card or monitor control and always use the same settings for every session. Making good EQ judgements will be easier than ever!


----------------------------------------------------------------------------------------------------------------------------------------------------------------