Introduction to Music Production 14 – Mastering
Mastering is a daunting subject for most musicians, especially home producers, because it seems to be a complex, esoteric audio process. Why would you want to do even more processing after a very clean, dynamic mixdown has been achieved? The answer: to have a commercially viable product.
First, the difference between mixing and mastering is that mastering usually is performed on a single audio file: the mixdown. Effect processing is performed on thee entire mixdown, usually including effects such as EQ, multiband and parallel compression, (sometimes) parallel reverb, saturation, and finally limiting (a form of heavy compression).
Mastering EQs usually contain a few different modes:
L/R: Left/Right mode allows a user to process the left side of the stereo field differently than the right.
M/S: Mid/Side mode allows the centermost (0 panned) items to be processed differently than everything outside the center of the stereo field. This is useful for balancing frequencies over the stereo spectrum.
Stereo: process the entire input signal as one cohesive unit.
Multiband compressors allow the user to compress different frequency bands different amounts. Most multiband compressors also allow the user to work in M/S, L/R, and Stereo modes. The goal of the multiband compressor is to differently process the dynamics of each band separately to help balance the overall dynamics of the mix. This is useful, especially in M/S mode, to create more power across the stereo field.
Parallel processing, though used also in mixing, is a commonly used item to help generate more power, brightness, and sometimes width in audio. The concept is to take an effect, set the Dry/Wet to 50% (or if there is no dry/wet, a duplicate channel with the effect on it) of an effect, and to comfortably add a processed signal back into the original. Parallel Compression is a commonly used technique on drums to give the drums a bit more ‘pop.’ Parallel saturation can give a bit more brightness and life to a mix. Parallel reverb can help ‘glue’ a mix together.
Some mastering engineers will finalize their effect chain with a limiter. A limiter is a compressor with a ratio of 10:1 or higher. The concept of a limiter is to stop any audio going over the threshold. For instance, a limiter may be set at -0.3 dB to ‘limit’ any sounds from going any higher than that.
Once the effect processing has been completed, a mastering engineer will export the master. The master peak should hit around -0.3 dB, and will usually be exported as 44.1 kHz, 16 bit WAV file for CDs.
A process called ‘dithering’ can take place when mastering and exporting masters (depends on the software). Dithering adds calculated noise to help create a ‘noise floor’ that assists when downsampling to a lower sample rate (i.e. from wav to mp3). Dither isn’t audible noise (usually at -60 dB) , but it does help prevent clicks and pops (artifacts) when taking a very dynamic song at a higher sample rate, and reducing the sample rate in file compression.
If you want to learn more about dithering, go to Ear Level’s website.
Mastering doesn’t always stop at effect processing, though. Mastering engineers often also assist with preparing music for digital and physical distribution. This can include assigning metadata (unique track information) and ISRC codes. ISRC stands for International Standard Recording Code, a unique alphanumeric code for every song ever released. You will likely see most of the metadata a track contains in software like iTunes, but ISRC codes are usually not accessible, as they help track the usage, sales, and plays of songs.