Mixing and mastering audio is the process of balancing instruments, frequencies and tones within a song. Using an EQ in the process is simply adjusting the volume level of a specific frequency, or frequency range, for the entire song. But how do I know what frequencies to adjust for different instruments in a mix, or how to change the tone of my finished mix?
Frequency adjustment in mixing is generally done per track and bus and is either the first or second insert on a track. In mastering too, EQ is done early in the signal chain. In mixing, radical and surgical adjustments are commonplace, mainly when combatting instrument masking or individual problem frequencies. On the other hand, any applied processing during mastering aims to produce the perfect tonal balance.
Mixing and mastering are carried out towards the end of music production. Mastering always follows mixing, and each has very different roles. Mixing studios and mastering rooms serve different purposes, but both share the need for frequency adjustment.
Experienced engineers can listen to an individual or fully mixed track and instantly recognize problem areas that can be corrected with EQ. Home studio owners, on the other hand, are less experienced and often look for shortcuts, tips, and tricks, to assist with their music. While there may be some valid tips and tricks, there are no shortcuts to experience and hard work.
If you came here looking for a magic formula, or a frequency cheat sheet to short cut your demo into a radio-ready, or commercial release, you are well advised to rethink your strategy and look much more deeply into what it takes to mix or master your audio.
Figure 1: Mixing and Mastering Magic Frequencies
There are no magic formulas for mixing and mastering frequencies. Every song ever recorded or produced, in all the genres, from all the artists, all over the world are unique. You may even think that songs within a single genre would be mixed and mastered in a very similar way. But this is simply not true, even within the same styles of music, mixing and mastering require an almost unique approach each time.
In conclusion, there cannot be a set of rules that could be applied to turn recorded audio into a sonic masterpiece. To create a sonic masterpiece takes experience, technical ability, production know-how, and for novices, a great deal of time, patience, and dedication.
Audio heard by a listener is perceived as having an equal amount of sound pressure energy across the audible frequency range. The perfect tonal balance of music varies, person to person, and even song to song, and depends a lot on taste (and our individual DNA!). Our ears all hear things slightly differently, but generally, the interpretation of perfect tonal balance is as follows:
We discussed earlier that to mix and master audio, we need to consider the balance of frequencies in a recorded track, or on a final mix ready for the mastering process. Why do we need to think about balancing frequencies? Why isn’t it enough that the mixed audio is ready for commercial release?
Apart from the seemingly noticeable difference in the overall volume level of a mix and a commercially available song, the tonal balance of the mix is almost always something that can be improved. The job of the mastering engineer, without a doubt, is to improve the tonal balance of a mix.
The mastering engineer usually has the advantage of a more suitable listening room to hear the tonal balance of recorded and mixed audio, and make subtle adjustments for improvement.
A “perfect” mix theoretically doesn’t need to have equalization added in the mastering stage, but that would be a decision for the mastering engineer and in any case mastering is not just about EQ.
On the face of it, balancing the tone must surely be a simple process as all that is needed is to bring all the frequencies or frequency ranges up to the same level. Right?
If this were true, I’m sure by now that somebody would have developed a plug-in to analyze the frequencies on a stereo track and apply a calculated dynamic EQ curve to raise or lower frequencies throughout the song to the same level.
Unfortunately (or fortunately depending on your point of view) our ears behave in a very non-linear manner.
As humans, we perceive frequencies at the same sound pressure level (equal volume) as having very different volume levels. The shapes of our ears determine how we hear different frequencies. At different volumes our ears follow specific contours across the audible frequency range, and this is important when it comes to tonal balance. These contours are known as the “Fletcher Munson Equal Loudness Contours.”
If you would like to learn more about the equal loudness contours, please read the article here.
Figure 2: Fletcher Munson Equal Loudness Contours
Looking at the Fletcher Munson equal-loudness contours, we can see, broadly speaking, the frequencies below 200Hz and frequencies above 4kHz rise on the graph and are therefore perceived as louder. Understanding this point is essential, as we’ll see later because the tone will be altered using only bass and treble adjustments.
In the following parts of the article, we will discuss how the application of a well-known EQ curve, some high quality professional outboard gear, mid-side processing, and frequency balance can help us understand, implement and refine the tonal balance of our mixes.
In the early 1950s, a British Audio Engineer named Peter J. Baxandall came up with a revolutionary, and now legendary, tone control circuit. The analog circuit provides independent adjustment and control of the bass and treble frequencies. In addition, the circuit utilizes two sets of capacitors to smooth out attenuation or boost frequencies. Understanding this point is also crucial as broad, smooth, and subtle frequency adjustments should have little or no added processing artifacts when mastering audio.
One disadvantage seen in the audio hi-fi world of the Baxandall circuit is that it responds more at the extremes of its controls. This means that as the tone control is turned up, the effect on the audio signal changes exponentially. To my mind, this is a considerable advantage for mastering, because the rotary control discourages the engineer from over processing frequencies.
The audio company “Dangerous Music” produce, amongst other things, a mastering EQ (BAX-EQ) which utilizes the legendary Baxandall circuit in its design. The BAX-EQ is hailed by producers, mixing engineers, and mastering engineers within the audio industry, as one of the number one mastering EQ units on the market today.
Figure 3: Dangerous Audio BAX-EQ
The BAX-EQ contains a 12dB per octave low cut filter, a low shelving filter, a high shelving filter, and a high cut filter. Shelving filters have a very broad Q, which, according to the manufacturers, give a very “natural, open transparent sound.” Each filter can be adjusted from -5 dB to +5 dB in 0.5 dB increments. Table 1 below shows the notch settings for the BAX-EQ hardware unit.
|Low Cut||Low Shelf||High Shelf||High Cut|
|12 Hz to 54 Hz||74 Hz to 361 Hz||1.6 kHz to 18 kHz||7.5 kHz to 70 kHz|
|12 Hz||74 Hz||-5dB to 5dB||1.6 kHz||-5dB to 5dB||7.5 kHz|
|18 Hz||84 Hz||1.8 kHz||9 kHz|
|24 Hz||98 Hz||2.1 kHz||11.1 kHz|
|30 Hz||116 Hz||2.5 kHz||12.6 kHz|
|36 Hz||131 Hz||3.4 kHz||18 kHz|
|43 Hz||166 Hz||4.8 kHz||28 kHz|
|54 Hz||230 Hz||7.1 kHz||(70 kHz)|
|361 Hz||18 kHz|
Table 1: Notch Settings for the BAX-EQ Hardware Unit
The hardware unit can be used in stereo mode or in mid-side (M/S) mode and is equally at home when recording, mixing, or mastering. Dangerous Music, in collaboration with Brainworx, has also modeled the hardware unit and made available a plug-in version (the plug-in version is available from Plugin Alliance who offer a 14 day fully functional trial - you can also hear many audio examples on their website).
What is Mid-Side mode? Generally, we think about audio in terms of left and right side channels which produce a stereo image. There is another way to think about how the stereo image is produced. On many mixes, it is usual to place the kick drum and bass guitar in the center of the stereo image and pan mid and high range instruments to the extremes.
The audio signals of the low-frequency instruments are sent to the right and left speaker, in equal amounts. And corresponding amounts of the signals produce the phantom, centralized, bass image, known as the mid-channel. Conversely, signals which are not sent to the right and left in equal amounts make up what is known as the side channel.
Now that we understand how powerful the Baxandall circuit is, and which frequencies the well respected and loved Dangerous BAX-EQ can be controlled, we can apply these settings, using our own mid-side (or stereo) EQ.
It is, of course, a bonus if our mid-side EQ also has the Baxandall curves available in it, such as the Izotope Ozone 6 Equaliser. Check if any of your own EQ is have the Baxandall curves built in.
Figure 4: Izotope Ozone 6 Equaliser
When adjusting EQ bands, it is worth noting that changes at one end of the spectrum have an opposite, relative, effect at the other.
For example, boosting a high-frequency shelving filter at 4 kHz can not only make some instruments or vocals sound brighter, but also make the kick drum and bass guitar sound weaker. We cannot change one frequency without making a relative, or perceived, one to another. To balance a seesaw, you need two equally sized children each placed at one end of the seesaw.
So, let’s re-cap and consider how we can make use of what we have learned and how we can apply it to adjusting the frequencies while mixing and mastering. Consider the following:
So now, whether you are a novice or professional, you have a unique approach to investigate, hone your skills, and practice adjusting frequencies on a finished mix. The procedure below can:
For the procedures below, please refer back to Table 1.
Start with a 12 dB per octave cut at 36 Hz and sweep the cut to the lowest point, back to 36 Hz and then sweep up to 54 Hz. If you’re wearing a good pair of headphones, you will be able to hear the bottom end cleans up. Adjust the low cut filter just below the frequency where it starts to make a small difference.
If you are not in Studio B at Bob Katz’s digido will probably need a good set of headphones to hear what’s going on at the low end.
Start at 12.6 KHz and sweep down the frequencies to 7.5 kHz and back up to the maximum. If you have a decent pair of studio monitors and a good listening environment, listen to the high frequencies being cut out, as you dial in the filter. Leave the cut just above where it doesn’t seem to do anything.
Set the low shelf at 116 kHz and boost 3 dB. A/B the filter on/off to hear what is happening to the mix. With the filter on, pull the gain down to -3 dB. Again A/B to listen to what is happening to the mix. Next, repeat the process for all the “set” frequencies in Table 1. Use A/B comparisons to listen to the difference with the filter on and off.
Repeat the procedure above (for the low shelf) using the values in Table 1. Start at 2.5 kHz with a 3 dB boost and work your way through the 1.6 kHz to 18 kHz range, then repeat with the gain set at -3 dB. Use A/B comparisons to hear the difference with the filter on and off.
In the next steps set the EQ up in M/S mode and repeat the four steps above for each channel - that’s eight steps in total. You should notice a few different things in this exercise. For example, check out how your mix sounds:
While you are dialing the frequencies in and out, the sound should be smooth and free from artifacts. If you have a good pair of headphones should be using those.
Making boosts or cuts and A/B-ing remember that your ears will naturally be attracted to the louder version. Do not get misled by higher SPLs!
What is a good frequency range for speakers? For mixing and mastering, audio monitors need to match the audible range of human hearing. Humans can hear frequencies between 20Hz and 20kHz. All speakers designed for mixing and mastering will reproduce these frequencies and are known as full range monitors. Most full range designs contain a two-way speaker system. One speaker, the woofer, reproduces low and mid frequencies, while another speaker, the tweeter, reproduces high frequencies. Manufacturers endeavor to design speakers that have a very flat response. This means that the individual frequencies within the full range have the same magnitude. This is beneficial for mixing and mastering engineers as the audio signal, which enters the speaker, is faithfully reproduced.
What is the best frequency response in audio? Frequency responses in audio cover bass, mid, and treble tones. Bass frequencies are generally between 20 Hz and 300 Hz. Mid frequencies between 300 Hz and 4 kHz, and treble frequencies between 4 kHz and 20 kHz. As humans, we cannot interpret the three bands of tone equally. Our hearing emphasizes bass and treble audio reproduction. Therefore, the most pleasant frequency response reproduction in audio typically boosts low and high end. This is one reason why bass and treble tone controls on home hi-fi systems are so popular.
What frequency is best for bass? Bass frequencies are typically referred to as the frequencies below 250 Hz. The “normal” bass frequency is between 60 Hz and 250 Hz, and sub-bass below 60 Hz. When mixing music, the bass guitar and kick drum sit together at the low end and require adequate processing to allow them both to be heard clearly. Typical home studios cannot faithfully reproduce recorded audio at the low end, and the advice here, as always is to use a good set of open back audiophile headphones when mixing your music.
What does a piano EQ chart look like? Below is a chart that provides the notes of a full range, eight octave piano, and the relative frequencies. Use the chart as a reference for finding frequencies of musical notes. Notice how each separate note doubles in frequency as the octaves increase.
|C# / Db||17.324||34.648||69.296||138.591||277.183||554.365||1108.731||2217.461||4434.922||8869.844|
|D# / Eb||19.445||38.891||77.782||155.563||311.127||622.254||1244.508||2489.016||4978.032||9956.064|
|F# / Gb||23.124||46.249||92.499||184.997||369.994||739.989||1479.978||2959.955||5919.91||11839.82|
|G# / Ab||25.956||51.913||103.826||207.652||415.305||830.609||1661.219||3322.438||6644.876||13289.752|
|A# / Bb||29.135||58.27||116.541||233.082||466.164||932.328||1864.655||3729.31||7458.62||14917.24|
Table 2: Notes and Their Corresponding Frequencies in Hz