We've all been there at some point in our mixing journey, spending a good amount of time on a mix, only to have it play back much quieter than our references and all other commercially-released tracks we compare it to.
With the loudness war spanning multiple decades, loud mixes have been considered “necessary,” so even if we aren't interested in pushing our mixes loud, we should at least know the strategies involved in doing so. If you're wondering how to increase the loudness of your mixes, you've come to the right place.
In this complete guide to louder mixes, we'll discuss the following 11 strategies:
- Gain staging
- Parallel compression
- Serial compression
- Harmonic saturation
- Mix bus compression
- Mix bus limiting
- Multiband compression
- Multiband limiting
- Digital clipping
We'll also be covering what loudness actually is, the importance of dynamics, whether loudness is even worth striving for in the mix, and the role mastering plays in loudness maximization. Let's get into it!
This is a fairly long article. Click here to skip to the strategies if you'd like.
What Is Loudness?
Loudness, in acoustics, refers to the subjective perception of sound pressure level.
Loudness, in the realm of audio, refers to the recreation of acoustic loudness during playback. Of course, in many playback cases, we can use an amplifier with a dedicated volume control to alter the loudness.
However, we can also make the audio louder itself, which relies on increasing the peak and average levels of the audio signal.
The peak level of audio is the highest amplitude of the signal.
In the realm of digital audio, we have a 0 dBFS (decibels full scale) ceiling that our peak values cannot exceed. Any signal peaking higher than 0 dBFS will be effectively cut off at the ceiling, causing digital clipping.
As a side note, we have sample peak and True Peak in digital audio. Sample peak measures the peak amplitude of the samples within a digital audio signal, while True Peak takes into account all the inter-peak samples as well. Having True Peak values below digital clipping is important for digital-to-analog conversion (playback) and any digital codec or conversions required. True Peak values are measured in dBTP rather than dBFS.
The same idea applies to the analog realm, though it's more complicated to give a strict ceiling. Different analog equipment and storage media will have different maximum peak values and will distort well before these maximums (soft clipping).
The “average levels” of a signal are typically noted as RMS (root mean square) values since audio signals have positive and negative polarity (meaning symmetrical waveforms would technically have an average of zero).
This can be visualized in the following Fletcher-Munson curves and Equal loudness contours:
Here we can see frequencies along the x-axis, sound pressure level (SPL) along the y-axis, and phon lines (a logarithmic unit for perceived loudness) drawn on the grid. We see that we're much more sensitive to the midrange frequencies (especially in the 2-6 kHz range) than the low-end and high-end. We also see that this natural frequency response of human hearing tends to flatten out as the SPL increases.
That stated, RMS values consider all frequencies equally, yet we do not hear all frequencies equally. Therefore, we need to find a different way to understand our perceived loudness.
Enter the LUFS (loudness units full scale), otherwise known as LKFS (Loudness, K-weighted, relative to full scale).
Loudness units are units of measurement to quantify the perceived loudness of audio by analyzing the average loudness levels across the frequency spectrum over time. In other words, this measurement unit considers our natural hearing response and how we're more sensitive to some frequency bands over others.
Because LUFS is relative to full scale, the readings will be negative (0 LUFS, like 0 dBFS, is the absolute digital ceiling).
Theoretically, two tracks with identical LUFS readings will be perceived as having the same loudness, even if their peak levels and RMS levels are not the same.
LUFS can be tracked by the momentary (averaged over 400 ms), short-term (averaged over 3 s), or integrated value (averaged over the entire track).
So, to increase the loudness of our mixes, we need to increase the LUFS.
Increasing the peak level and LUFS of a mix is pretty straightforward until we reach the ceiling of maximum amplitude. To keep things simple, I'll discuss digital audio and its hard ceiling at 0 dBFS.
So let's assume we've increased the mix level to 0 dBFS or, even better, 0 dBTP. We're right at the ceiling, so it's worth bringing it down just a bit to be safe (even to -0.1 dBTP). The dynamics of our mix remain intact, and our LUFS rating will increase to a point.
However, we're likely leaving potential loudness on the table in this scenario. It's possible (and commonplace) for us to push mixes (or at least masters) up against the ceiling, reducing the transient information to effectively increase the LUFS. As we'll see in an upcoming section, this is typically done via brickwall limiting.
For more information on the topic of decibels, be sure to check out my article What Are Decibels? The Ultimate dB Guide For Audio & Sound.
Dynamics Vs. Loudness
Music needs dynamics. Loudness needs higher average levels. These two aspects of a mix/master are often at odds with each other.
As I stated previously, there's only so much headroom before we reach the maximum ceiling, where our peaks will get cut off.
Limiting is our primary tool for loudness maximization in mastering (and mixing). This tool acts like a compressor with a ratio of ∞:1, reducing the loudest parts of the audio so that the audio never peaks above the set threshold.
Using a limiter, we can bring up the LUFS of a mix/master without clipping at the expense of reducing the loudness of most transient information. In other words, limiting allows us to trade off dynamics for loudness.
Of course, this is a compromise we need to get right in the mastering process. Many mixes have been destroyed dynamically by over-limiting.
To learn more about headroom, check out my article What Is Headroom In Audio? (Recording, Mixing & Mastering).
Is Loudness Something To Strive For In The Mix?
In the era of streaming, most platforms will normalize to -14 LUFS to maintain consistent levels across different songs.
However, there's a certain “sound” to loud tracks. While it's often distorted with squashed transients and fatiguing on the ears, many listeners have grown accustomed to loud records over the last 30 years of the loudness war.
So while we aren't necessarily beholden to loud mixes for the sake of competing on mixtapes and CDs, we may want to push our mixes and masters louder to maintain a trend. We can often get great results at -9 dBFS integrated (often referred to as “CD level”) and above, though pushing louder than this for loudness' sake usually isn't warranted today.
Of course, it's worth knowing how to craft loud mixes with minimal negative side effects as a mixing and mastering engineer.
Related article: 6 Reasons Your Audio Mixes Sound Distorted & How To Fix Them
Loudness Maximization And Mastering
Loudness maximization is generally a concern for mastering. However, there are ways to achieve greater potential loudness and “pseudo-mastering” for maximization in the mixing stage. We'll discuss both in the following strategies.
Even before mastering, we'll likely want to maximize the loudness of our mix for referencing against other songs outside the studio.
So we'll be discussing strategies to increase the loudness of our mixes, though loudness maximization is generally a concern for mastering.
As a side note, mastering engineers usually want some amount of headroom in the mix to ensure they have room to work with and that the full dynamics of the mix are present.
With that, let's get into the strategies.
Loudness Strategy #1: Gain Staging
The first strategy I have for you is a bit indirect in regards to increasing the loudness of a mix, but it's certainly part of optimizing the mix for potential loudness. That, by the title, is gain staging.
Gain staging is the process of feeding each audio device in a signal chain with the optimal signal level. It helps avoid clipping and unwanted distortion by ensuring we stay under the clip/distortion ceiling while also giving us adequate signal-to-noise ratios by ensuring the signal is much stronger than any inherent noise being introduced along the signal chain.
While the noise issue is mostly in regard to analog equipment, the headroom part applies to both analog and digital systems.
Proper gain staging in the mix typically means keeping the individual tracks around -20 to -18 dBFS or 0 VU. Of course, there are instances when driving certain pieces of equipment or plugins a bit hotter is beneficial, but this is the general rule of thumb.
Gain staging in a mix also means doing our best to maintain signal levels across the inputs and outputs of our gear/plugins. This level matching ensures we have similar levels whether the processor is on or off, making A/B testing easy and keeping signal levels healthy across their signal chains within the mix.
So how does gain staging help with loudness?
First, it helps us to avoid unwanted/uncontrolled digital clipping and distortion within the mix. If we aren't clipping any tracks or buses and we're driving our processors with optimal signal levels, we won't be introducing clipping. We can then be more mindful of the loudness practices of compression, saturation and limiting on our own terms.
Second, it makes mixing much easier by having all tracks at similar levels at unity gain. Additionally, our levels won't jump all over the place if we need to bypass any processes. If our mixes are easier to mix, we can focus our full attention on improving the loudness potential of the mix when it comes time to.
I have a video dedicated to gain staging. You can check it out here:
To learn more about gain staging, check out my article Mixing: What Is Gain Staging & Why Is It Important?
Loudness Strategy #2: Arrangement
The arrangement of the song is another factor to consider in regard to potential loudness. Some arrangements are easier to make loud than others.
As mentioned in the section What Is Loudness?, we're the most sensitive to midrange frequencies. So then, if we have an arrangement that features mostly midrange elements/energy versus one that has a lot of low-end energy, we can have an easier time getting the mix loud.
Remember that low-end frequencies require more energy and eat up more headroom than midrange frequencies. We can get a higher LUFS rating with equal peak and RMS values if our mix is focused on the midrange.
Of course, that's not to say that the low end isn't important. It definitely is. However, an acoustic singer-songwriter mix without bass and low-end percussion will be easier to drive louder than a full band with notable bass guitar/synth, kick drum, toms, etc.
Sticking with the comparison of stripped-down singer-songwriter mixes and full-band mixes, it's also the case that denser arrangements may be more difficult to get loud.
With more elements competing for space in the mix, we have to make sacrifices for separation. This often means EQing each element somewhat unnaturally to blend things together. Without proper separation, we can have poor definition of the distinct tracks in the mix, which can be exacerbated when we try to push the levels up.
Furthermore, dense mixes tend to have build-ups in certain frequency bands, particularly in the low-end and low-midrange. If these build-ups aren't taken care of in the mix, we'll have a harder time getting a great deal of loudness.
If elements are clashing in the mix, we may need to do some work to solve the frequency masking issues to get the best mixes results and the greater loudness potential.
Beyond the arrangement itself, we should take into account how we're balancing each element in the mix. If we want to maximize the potential for loudness, we'll want to reduce the levels of low-end elements and/or boost the levels of midrange elements.
I have a YouTube video dedicated to frequency masking. Check it out here:
Loudness Strategy #3: EQ
Once again, the potential for perceived loudness in a mix partly depends on that mix's frequency content. Since EQ is the primary tool for frequency-dependent gain control, it makes perfect sense that EQ can help us to make our mixes louder.
In general, turning down the low end and turning up the midrange will give us more perceived loudness at the same peak and RMS levels since the low end requires more power and takes up more headroom.
This can be done on the mix bus, effectively altering the entire balance of the mix. However, we'll have much more control doing this work on the individual track or subgroup/bus level. That way, we can hone in on the balance of the tracks at each frequency band.
Related article: Audio: Buses Vs. Subgroups, Aux Sends/Returns, VCAs & Groups
Getting the loudest possible balance with EQ can be difficult on the first try, as we're typically not concerned with loudness maximization throughout the bulk of the mixing process (it's generally a process done after mixing). EQ rebalancing, once the mix is done, is, therefore, a strategy worth considering.
EQ rebalancing can give your mix greater perceived loudness at the obvious expense of altering the frequency balance you've likely worked hard on in the mixing process.
Equalization is a critical tool for addressing issues of frequency masking. This is important for separation, particularly in the low end, between common elements like the kick and the bass guitar, synth or other bass elements.
Clashing low-end elements can cause poor mix results, including reduced potential for loudness.
The frequency masking can make the elements ill-defined in the mix. Fortunately, EQ can help us reduce this clashing and get us on track for louder mixes.
Mirrored EQ can help with this frequency masking. Taking the kick and bass guitar as an example, we can boost the fundamental frequency of the kick drum while simultaneously cutting that same frequency from the bass. We can then boost the important first harmonic range of the bass and cut that same frequency range from the kick. This will help make room for both elements in the mix and clarify the low end.
We can also use EQ to high-pass filter everything that doesn't have important musical information in the low end. By eliminating the low-end noise from the majority of the tracks in the mix, we can make more room for the important low-end elements and ultimately give ourselves more headroom for potential loudness.
Additionally, phase issues can lead to inconsistency between different low-end transients and low-end notes.
For more info on the importance of phase, especially in the low end, check out my article Mixing: What Is Phase & Why Is It So Important To Get Right?
Keeping with the low-end and low-mid-rich elements of the mix, we can easily eat up headroom and reduce the potential for loudness if we simply increase their levels in the mix. Remember the importance of the midrange frequencies? Giving low-end elements like kick and bass a bit of presence boost in the midrange can help them be heard in the mix without overdoing the low-end.
By increasing the presence of bass elements, we can rebalance the low-end while still hearing the important low-end elements. This will then allow us to push the loudness a bit more.
To learn more about EQ, check out my article Complete Guide To Audio Equalization & EQ Hardware/Software.
Loudness Strategy #4: Parallel Compression
Parallel compression (also known as New York or Manhattan compression) is a technique where one audio track (or several) is sent to a send/return (aux bus), and that return is heavily compressed. Both versions of the audio are then mixed together to achieve a punchy sound without losing the dynamics of the dry signal(s).
Rather than having a downward effect like normal compression, where peaks are reduced in level to reduce dynamic range, parallel compression has an upward compression effect, where quiet parts are effectively brought up in level to reduce dynamic range. In other words, upward compression increases the levels below a set threshold rather than decreasing the levels above a set threshold.
This reduction in dynamic range allows us to increase the “average level” at a much greater rate than the peak level, thereby increasing the loudness.
Although there are other processors capable of upward compression, the most typical method of achieving upward compression is via parallel compression.
Parallel compression is typically achieved via an auxiliary track (effects send and return) routing setup. However, we can also opt to duplicate a track and heavily compress the duplicate for the same effect. Additionally, many compressors include a dry/wet mix control, which allows for parallel compression within the compressor itself (dry is the original signal, and wet is the compressed signal).
For more info on auxiliary tracks, check out my article Mixing/Recording: What Are Auxiliary Tracks, Sends & Returns?
This compression technique can be used with individual tracks, multiple individual tracks, subgroups/buses, and even the mix bus.
Send a track, bus or entire mix to a parallel bus, apply a good amount of compression, bring the fader down, and push it up to mix it appropriately with the full mix.
Listen for unnatural pumping and audible artifacts when pushing up the fader of the parallel bus in the context of the mix. We can get away with pretty aggressive compression here, which may sound awful in solo. Remember that the only thing that matters is how it sounds in the entire mix.
Be sure to A/B the results by muting and unmuting the parallel compression bus.
I have a YouTube video on the importance of mixing in the context of the mix rather than in solo. Check it out by clicking here.
Expect slightly less punch on transients (though not as much as “regular” mix bus compression), less control over the peak levels (which can lead to clipping if we're not careful), and greater relative level increases in sparser sections of the song.
To learn more about buses, check out my article What Are Audio Buses? (Mixing, Recording, Live Sound).
I have a video dedicated to parallel compression that you check out below for more detail:
I also have a video where I discuss parallel processing in broader detail that you check out here.
For more information on compression, check out my article The Complete Guide To Audio Compression & Compressors.
Loudness Strategy #5: Serial Compression
The second loudness strategy involving compression is the use of serial compression. As the name would suggest, serial compression is the practice of using multiple compressors in series (one after the other) to combine desirable attributes from different compressors, lighten the workload of any single compressor, and help increase the potential for loudness.
Downward-style or “regular” compression reduces the dynamic range of an audio signal by reducing the gain above a set threshold. This process helps bring down the peak level relative to the RMS level, thereby increasing the potential for loudness.
When using compressors in series, we reduce the workload on any particular compressor while achieving the same amount of gain reduction. Hitting any compressor too hard can lead to undesirable pumping, distortion and artifacts. Spreading the work between multiple compressors can help mitigate these negative side effects, as each compressor can reduce the gain by a fraction of the overall amount.
Compressing with multiple compressors in series will typically yield a more natural-sounding compression, allowing us to increase the perceived levels (with appropriate makeup gain) without the typical negative side effects of overcompression.
Finally, as we mix audio, we can have compressors on individual tracks, buses and groups, final mixes and masters. This means several compressors can process any single track before the final mix.
I talk about using serial compression for loudness in more detail in one of my YouTube videos that you can check out here:
Loudness Strategy #6: Harmonic Saturation
Distortion often gets a bad rap in audio. I've already discussed the dangers of clipping distortion in the article. However, distortion is also an invaluable tool in mixing when used properly.
Harmonic saturation is one style of distortion that proves helpful in mixing by effectively adding and enhancing the harmonic content of the audio it affects.
Harmonics are integer multiples above a pitched instrument's fundamental frequency. Each instrument will have its own harmonic profile, with each harmonic having its own relative amplitude and amplitude envelope.
The harmonics of sound explain why an acoustic guitar, oboe and piano will all sound different when playing the same note. For example, these three instruments could all play A4 (fundamental frequency or “first harmonic” of 440 Hz), and each would have some harmonic content at 880 Hz (second harmonic), 1,320 Hz (third harmonic), 1,760 Hz (fourth harmonic), and so one. However, each would have a completely different harmonic profile that would give them their tone and timbre.
As an aside, non-pitched instruments and noise are considered “inharmonic” and will be made of collections of frequencies that do not follow the harmonic series.
Getting back to our discussion on loudness and revisiting the idea of increasing the midrange presence of instruments (particularly low-end elements), saturation is an excellent tool as it increases the harmonic content in the midrange.
While EQ can help increase the presence of tracks to help increase the potential for loudness, it can only work on what information is already contained in the audio. Saturation, on the other hand, creates new harmonic content based on the frequency content of the signal (not only on the fundamental but also on the original harmonics and any noise within the audio).
So saturation can help us increase the perceived loudness of bass elements and other tracks without needing to push faders or increase the low-end, eating up headroom. We can add midrange content and increase the LUFS without having a massive impact on the peak level.
In fact, the soft-knee compression inherent in saturation can actually reduce the peak level while simultaneously increasing the LUFS level.
It also works well on the mix bus in some instances. By adding saturation (tape, tube, transformer or transistor) to the mix bus, we are able to add harmonic content, thereby increasing the perceived loudness without having a huge impact on the peak values of the mix.
Saturation can be applied directly to the mix bus or a parallel bus to add levels to the mix. Parallel processing gives us more control and affords us more colour (more distortion) on the parallel track that we can mix in rather than colouring the original audio.
Of course, any distortion processing can be easily taken too far. Watch out for the all-too-common side effects, including harshness in the mid-range and high-end, excessive sibilance, increased frequency masking between tracks and unwanted changes to the overall tonality of the mix.
Loudness Strategy #7: Mix Bus Compression
Mix bus compression is super-common and will help us tame the peak transients of the mix while bringing up the average level with makeup gain. By controlling the dynamic peaks of the entire mix, we can increase the LUFS relative to the peaks.
Great care must be taken with mix bus compression not to overprocess the mix as a whole. It's advisable to keep the ratio low (1.5:1 to 4:1 at the most) and to set the threshold so that gain reduction is kept to a maximum of about 3 dB.
Because this processing affects the entire mix, we must be careful to avoid pumping, distortion, and other artifacts that would ruin a great mix otherwise. Adjust the attack and release times to make the compressor work better with the rhythm and feel of the music and to avoid the notorious pumping effect.
Mix bus compression will tame transients at the expense of transient definition, so we should be listening to how the compressor affects the overall punchiness of the mix's drums and percussion (and other transient elements).
As always, A/B the process by matching perceived levels and bypassing/engaging the compressor.
For more information on A/B testing, check out my article A/B Testing & Its Importance In Mixing (With 5 Best Tests).
Another side effect to consider is that the background and side information will likely be brought up in level, which can draw attention away from the main, centre-panned elements of the mix. This can be great for width and detail, but only to a point. Delay, reverb, and panned tracks can become too loud in this case.
Additionally, we should be aware that mix bus compression will cause sparser, quieter parts of the song to sound relatively louder. It may be the case that we have to revisit the balance of these quiet parts to ensure the long-term dynamics of the song remain intact. This is one of the reasons why many mixing engineers opt to mix into a mix bus compressor from the initial balance rather than waiting until the end to incorporate it into the mix.
As a side note, mix bus compression naturally belongs to serial compression chains (so long as we're using compression on any other track or bus feeding into the mix bus, which is almost a certainty).
I have a video discussing top-down mixing versus bottom-up in greater detail. Check it out here:
Loudness Strategy #8: Mix Bus Limiting
Limiting is a common tool for increasing loudness and is often the first process to reach for when trying to make things louder.
As mentioned previously, limiting will help us make use of the potential a mix has for loudness. It can bring the signal level to maximum levels without peaking.
In mastering, we'll start with a mix that has some headroom below clipping. Limiting is used to maximize the potential for loudness by increasing the mix level to the maximum ceiling and beyond without suffering digital clipping.
Limiting is effectively compression with a ratio of ∞:1. It stops peaks from surpassing a defined threshold while keeping the rest of the levels untouched. We bring up the input level or gain of the limiter so that the peaks get squashed while the average level is increased.
Of course, this is ideal, though there will be some effect on the sub-threshold signal and even some signal that surpasses the hard limit/threshold.
Side effects of pushing limiting too far are similar to compression, including pumping, distortion and audible artifacts.
Because limiters are designed to stop peaks at a set threshold, they must act fast. Having a fast attack, however, can actually cause waveshaping in longer waveforms (low frequencies). That means that pushing limiters too hard can have a disproportionate effect on bass distortion. This is something else to look out for when limiting the mix bus.
With modern limiters, we can often get away with 3 dB of gain reduction or more on the peaks before pumping and distortion become noticeable.
Like with serial compression, we can opt to set up serial limiters to bring up levels in two stages rather than relying on a single limiter to do all the heavy lifting in terms of loudness maximization.
In increasing loudness, limiting has the negative side effects of softening transients and reducing the perceived levels of drums and percussive elements while increasing the relative levels of quieter parts of the song.
It's generally advisable to set the threshold of the master limiter to -1.1 or -1.2 dB to help ensure the song peaks at -1 dBTP (Decibels True Peak), which may mean reducing the limiter's threshold slightly if it's working hard. This allows some room before 0 dBFS clipping and is necessary to avoid digital distortion when the audio is converted and encoded to most lossy formats.
For more info on using bussing in your mixes, check out my YouTube video here.
Loudness Strategy #9: Multiband Compression
Multiband compression splits the audio into multiple bands (often 3 or 4) with crossovers. Each band then has its own compression controls.
We know how useful compression is for controlling dynamics and increasing the potential for loudness. Multiband compression gives us independent control over specified frequency bands, making it possible to control the dynamics with greater intricacy.
While multiband compression can be used on individual tracks and subgroups, it is often used on the mix bus.
For more information on subgroups, check out my article What Are Subgroups? (Audio Mixing, Recording, Live Sound).
Multiband compression on the mix bus gives us great power. It also gives us the responsibility of ensuring it doesn't throw off the balance of the entire mix. Fortunately, when applying mix bus compression, we're often only after a few dB of gain reduction, which works in our favour. However, we must be vigilant to ensure the frequency balance is maintained.
Additionally, the crossover frequencies are liable to experience distortion and phase shift due to the filters used. This is especially true if adjacent bands experience dramatically different gain reductions due to their signal levels and compression settings.
Furthermore, as the song's arrangement changes, so will the balance across the set bands of the multiband compressor.
Loudness Strategy #10: Multiband Limiting
As the name suggests, multiband limiting splits the audio into separate bands, giving us unique control over each band.
Take the pros and cons of multiband compression and consider the implications of hard limiting in each band (from the information above).
Loudness Strategy #11: Digital Clipping
While digital clipping is often frowned upon, and you might not want to hear it (in practice or in this article), it does have the potential to increase the perceived loudness of a mix.
Digital clipping has been used (and continues to be used) in an abundance of commercially successful tracks.
Clipping can have detrimental effects on tone, causing notable digital distortion. Yet, it can effectively get extra loudness and maintain transient energy better than limiting.
The distorted sound of clipping is often best used in harder styles of music where there's already significant distortion, though it may be used sparingly in softer styles.
With digital clipping, we push a signal past the maximum ceiling and cut off the tops/bottoms of the waveform. While this process stops the peak level at the maximum, it continues to increase the “average level” until we effectively distort the signal into a square wave.
This sounds much more distorted than limiting but has a similar effect when loudness is considered.
So I'm not suggesting you digitally clip all your mixes from here on out or that you ever clip your mixes and masters. I'm simply here to state the fact that digital clipping can get us increased loudness in our audio. Don't shoot the messenger!