It's no secret that compression is one of the most important processes in mixing. However, dynamic range compression remains somewhat mysterious to beginners, and many aspiring mixers don't use it to its fullest potential in their mixes.
In this article, I'll present you with what I believe to be the top 11 best compression tips to help you utilize this process to your fullest advantage in your mixes. Before we get started, I should clarify that some of the tip titles are a bit overgeneralized, but I go into more practical detail in the dedicated sections for these “vague” tips. Be sure to read through!
With that stated, let's get to the tips.
The top 11 best compression tips for mixing are:
- Always A/B test and set appropriate makeup gain
- Set release times to match the song's rhythm
- Set attack times to allow appropriate amounts of transient information to pass
- Know the genre and general mix aesthetic
- Automate volume post-compression
- Not everything needs compression
- Consider top-down mixing
- Use serial compression
- Use compression for colour
- Use parallel compression
- Use sidechain compression
I talk about these 11 compression tips in the following video as well. Check it out if you'd like a supplement to this article (or if you prefer the video format)!
Always A/B Test And Set Appropriate Makeup Gain
My first tip for you is to develop the habit of A/B testing your compression choices within the mix. Really, we should be A/B testing most mix processes by toggling them on and off, but we'll obviously be focusing on compression in this article.
Put simply, A/B testing tests “thing A” against “thing B” to decide which option is better.
When it comes to compressors, we can turn them on and off and listen critically to help make an objective as to whether the compression helps the mix or not.
Our ears/hearing have a natural propensity to adjust to whatever it is we're hearing. Suppose we're adjusting our compressor parameters to find the right amount of compression. In that case, we can quickly lose our objectivity or, at the very least, our original point of reference (what the audio sounded like originally).
So once we've set a compressor to our liking, it's always a good idea to A/B test by turning it off and back on again. Give yourself time to adjust your ears and listen critically to the differences. Ten seconds or so is typically enough.
By A/Bing this way, we can get an idea of what our compressor is doing without the incremental changes we heard as we were originally setting the compressor.
If the compressor indeed does help the mix, we can move on. If we find that it doesn't, we'll likely want to adjust the parameters or get rid of the compression altogether.
When A/B testing our compressors, it's critical to set an appropriate level of makeup gain for level matching. In other words, we should set the makeup gain to make up for the compressor's gain reduction so that the audio is at the same perceived level regardless of whether the compressor is on (A) or off (B) during the A/B tests.
Level matching in this way will get rid of the loudness bias, which is our natural tendency to prefer louder material. It will also help with gain staging throughout the mix, which will make balancing the mix much more straightforward.
For more information on gain staging, check out my article Mixing: What Is Gain Staging & Why Is It Important?
While many compressors offer automatic makeup gain, I'd recommend setting the makeup gain manually, if possible, to achieve proper level matching.
I'd also advise A/B testing your compression moves in the context of the entire mix rather than in solo. Remember that mixing is about mixing multiple audio tracks together and not necessarily about making any particular track perfect on its own.
I have a YouTube video discussing the most important A/B tests you can incorporate into your mixes. Check it out here if you're interested! I also have a YouTube video on the importance of mixing in the context of the mix rather than in solo, which you can check out by clicking here.
To learn more about A/B testing, check out my article A/B Testing & Its Importance In Mixing (With 5 Best Tests).
Set Release Times To Match The Song's Rhythm
Compressors don't act instantaneously to reduce audio levels. Rather, they have attack and release times that affect how they interact with the audio they're compressing.
The release time is the amount of time it takes for the compressor to disengage (to stop attenuating the signal) once the input signal drops below the threshold. Like the attack time, it's a rate of change whereby the compressor gradually disengages over time. It's not a delay of action, where the compressor will suddenly stop compressing after a set period of time.
In fact, if we were to have instantaneous compression, we'd have intrawave shaping, which would cause significant distortion as the tops of the waveforms would be squashed.
When dealing with very short release times, the compressor will quickly return to zero gain reduction. This can give some grit and aggression to the sound but can also cause pumping artifacts.
With very long release times, the compressor may never return to zero gain reduction if the signal regularly exceeds the set threshold. This can be good for smoothing signals out and making them sound distant in the mix but all too often sucks the life out of a dynamic performance.
Perhaps a better strategy is to adjust the release time to fit with the rhythm/tempo of the song.
Set the release time so that the compressor has just enough time to disengage before it's triggered once again by the following transient. This is most easily done with percussive audio tracks, though the initial attack of many vocals and instruments will have defined transient information to time the compressor to.
Having the compressor release just as the next transient peak feeds into it will reduce pumping and also allow the track to breathe.
Once you hit that point, consider if the mix needs to be a bit smoother or more aggressive. If it needs smoothing, consider lengthening the release time further. If it needs more aggression, consider shortening the release time.
Consider the tempo of the song (in beats per minute), the note values (1/2th, 1/4th, 1/8th notes, etc.) and the release time (in milliseconds). Take the bpm as the number of 1/4 notes in a minute (60,000 milliseconds). Divide the 60,000 ms by the tempo to find the release time to match the 1/4 and move one from there.
Note that the audio material itself will take up some amount of time, so reducing the release time slightly from the time calculated is generally necessary.
That said, here is a table listing out the relationships between a few different tempos (in BPM), note lengths and time (in ms):
|Tempo||1/2 note||1/4 note||1/8 note||1/8D note||1/8T note||1/16 note|
Set Attack Times To Allow Appropriate Amounts Of Transient Information To Pass
Moving on from release times, the attack time is the amount of time it takes for a compressor to engage/react once the input signal amplitude surpasses the threshold.
Once again, it's a rate of change whereby the compressor gradually reaches its full ratio over time. It's not a delay of action, where the compressor will clamp down fully after a set period of time.
Short attack times will cause the compressor to clamp down quickly on the signal as the threshold is exceeded. This can help tame transients (especially when paired with a slower release) and improve the perceived sustain and weight. However, it can also have the effect of reducing the punchiness and sucking the life out of an otherwise dynamic performance.
Medium attack times, often defined as single-digit or very low double-digit millisecond values, are great for dynamic control without overly affecting the signal's transients. These attack times are well-rounded and work on a variety of tracks. If we want articulation in the transient attack with good dynamic control, we can opt for a medium attack time.
Long attack times sound transparent as the initial transients are allowed to pass through uncompressed (assuming the compressor release time is set shorter than each successive transient). With greater levels of gain reduction, we can actually enhance the perceived dynamics of a signal by allowing the transient to pass before clamping down aggressively on the signal. Of course, this transparency isn't always the best. If we want more control to tame the overall dynamics, we'll likely want shorter attack times.
I talk about using compression to enhance transients and even increase the dynamic range in this video.
To learn more about compressor attack and release times, check out my article Dynamic Range Compression: Attack & Release Controls.
Know The Genre And General Mix Aesthetic
A more generalized tip for compression is to consider the genre or style of music you're mixing. This tip spans far beyond compression and can be asked in regard to many other mixing processes, but we'll focus our discussion on compression here.
There are certain mix aesthetics that listeners have come to expect from certain genres of music. For example, an EDM hit will be mixed differently from a Country song, which will be mixed differently from a Metal song, and so forth.
Some genres are, generally speaking, more compressed than others. How much compression is typical of the genre and the specific element in question? Is it jazz or classical (little to no compression) or EDM and pop (much more compression)?
So keeping these “standards” in mind, we can err on the side of more or less compression, or even no compression at all. This applies to the individual tracks, buses, and mix buses.
Have a few good listens to reference tracks in the same style or genre and get a feel for commonalities in terms of production and mix aesthetics. Carry this research with you as you go about mixing with compression and more!
I have a video in which I discuss reference mixes in more detail. You can check it out here:
For more information on reference tracks, check out my article Mixing: What Are Reference Mixes & Why Are They Important?
Automate Volume Post-Compression
Mixing is about balance. Balancing the levels of our tracks within a mix session is vital if we want great results.
Faders are the primary tools for achieving this balance. We can imagine compressors as being “automatic faders”.
A compressor is designed to apply gain reduction (similar to moving a fader downward) as the signal surpasses a set threshold. The compressor's gain reduction will be applied according to the ratio, attack and release times. As the signal drop below the threshold, the compressor will disengage and stop its gain reduction (similar to moving the fader back to its original position).
So, in theory, we could carefully and meticulously automate the fader (or some other volume control) to achieve the same effect as using a compressor. Of course, compressors are much more efficient, but this is a good way to understand the similarities.
Because of the efficiency of compressors, it can be tempting to try to control the dynamics of a vocal with absolute perfection using only compressors.
While serial compression (multiple compressors in series) can help share the load of dynamics control to produce more natural results than overworking a single compressor, we still may have work to do post-compression to really hone in on controlling the dynamics of the audio—more on serial compression in Tip 8.
If we can't quite get the results we need from a compressor (or multiple compressors), we can often easily make up the difference with fader automation. This can lead to more natural results than forcing a compressor to do all the work independently.
This can be useful in any overly dynamic audio, such as vocals.
For example, if the vocal is still too loud in some parts and compression causes unwanted results in the rest of the vocal, we can utilize a simple automated fader move to bring the vocal down only in that part.
Conversely, if the vocal gets lost only in a few select lines during the mix, we can bring the fader up only in these sections rather than relying on more compression during the rest of the vocal.
We can also choose to automate the volume of the signal pre-compression by inserting a gain/trim plugin before the compressor on a track. Alternatively, we could utilize a bus setup where we can control the amount the track is sent to the bus with a knob or fader before compressing the bus.
For more information on inserts, check out my article Audio: What Are Inserts? (Mixing, Recording & More).
Automating volume levels before compression can help to feed the compressor with a more consistent signal, helping it to work less hard and yield potentially better results.
So don't be afraid to experiment with volume automation before or after compression.
I have a video discussing my top 11 automation tips for mixing. Check it out here if you're interested.
Not Everything Needs Compression
Although compression is an excellent tool in mixing, it's also true that dynamics are incredibly important.
Over-compression is obviously bad. But even small amounts of compression can be detrimental to a mix if they're not needed. Of course, there's a fine line here. What I mean to say is that not everything needs compression.
Compression is great for glue (especially on buses), taming or accentuating transients, controlling dynamics, adding movement, and much more. However, many tracks will sound better in the mix without it.
In Tip 4, we discussed the importance of knowing the genre and general mix aesthetic. Beyond that, specific tracks within a mix may or may not benefit from compression.
For example, compression may not be warranted if:
- We need the dynamics of a track to poke through the mix.
- We're dealing with a distorted or “square-like” source, which is already compressed.
- We have significant bus compression down the line.
- Applying compression to the track makes the mix worse (be sure to A/B test – check out Tip 1).
Use your discretion and critical listening when applying compression, and know that it's perfectly fine not to use it at all if it's not benefitting the mix.
Consider Top-Down Mixing
Top-down mixing, as the name would suggest, is a style of mixing where we begin processing at the “top” (the mix bus) before moving “down” to the instrument busses/groups and eventually to the individual tracks.
It's common practice, as we begin mixing, to insert our mix bus compressor as soon as possible and mix through it. That way, we make all our mixing decisions with the final mix bus compression in line.
This will generally lead to better results than waiting until the end of the mixing process to compress the mix bus. We can spend the entirety of the mix mixing through the compression rather than hoping for the best when it comes time to insert the compressor at the end (or in the middle) of the mixing process.
Once the initial mix is set with appropriate headroom, we may want to start the mix with a mix bus compressor.
One step down, it may be much more efficient to compress groups of instruments together at the bus/subgroup level instead of compressing each individual track (though we may also want compression at the track level of some sources, too).
For more information on subgroups, check out my article What Are Subgroups? (Audio Mixing, Recording, Live Sound).
Doing so allows better “glue” compression from the get-go, making our bussed elements sound more cohesive. It also produces bigger results with fewer compressors. We may find that a bit of bus compression is all we need rather than a bunch of individual compressors on individual tracks.
Using fewer compressors frees up our time and energy along with the computer's processing power (if we're working with one) or the usage of analog compressor units (if we're working with analog compressors).
So consider a top-down approach when using compression, especially when it comes to the utilization of a mix bus compressor.
To learn more about buses, check out my article What Are Audio Buses? (Mixing, Recording, Live Sound).
I have a video where I discuss top-down mixing and its benefits in more detail. Check it out here:
I also discuss “starting big and ending small” as a mix philosophy (including top-down mixing) in this video.
Use Serial Compression
In the last tip, we effectively discussed processing against the signal path, starting at the end of the mix session and making our way to the beginning. In this tip, we'll look at signal flow in its proper direction and how compression can be set up in series.
Serial compression, as the name would suggest, is having multiple compressors in series within a defined signal path. In other words, having a compressor's output feeding another compressor's input (and so on) somewhere down the line.
Serial compression could be as simple as inserting two compressors on a single track or having a single compressor on a track that's ultimately being fed through a mix bus compressor. It could involve many more compressors in any given signal chain (a track into a bus into a mix bus, etc.).
Serial compression, in this way, is great for not overloading any single compressor. It can sound more natural and shave off more dynamics, leading to more consistent levels without as many unnatural compression artifacts (pumping, distortion, etc.). We can also utilize different compressors to incorporate the different sonic colours/characters of different compressors (see Tip 9).
So rather than pushing 12 dB of gain reduction with a single compressor, which may cause pumping, distortion and an overall sense of over-compression, we can split this gain reduction to two or more compressors.
Where over-compression can negatively affect the transient information and the sustain, we can potentially achieve a more natural result with a bit of serial compression.
For example, 2 compressors in series each with 6 dB of gain reduction, 3 compressors in series each with 4 dB of gain reduction, 4 compressors in series each with 3 dB of gain reduction, or some other combination that adds up to the 12 dB (not every compressor needs to be compressing the same amount).
When done correctly, it would lead to even less pumping, giving us a more natural sounding compression and a more consistent sound.
Try this out for yourself the next time you need a significant amount of compression to fit a track in the mix appropriately.
Use Compression For Colour
Try using compression for colour (even if only running the signal through the circuitry/plugin).
Compressors have inherent sonic characteristics, especially analog units or their emulations. Simply running audio through these compressors can give us nice sonic colouration, even if we aren't applying any gain reduction whatsoever.
Beyond the inherent characteristics (harmonic saturation, distortion, noise, etc.), each compressor will have its own sort of sonic character as it compresses audio. We can assume a few generalities depending on the compressor gain reduction circuit type.
Diode bridge compressors typically have the following characteristics:
- Very fast attack and release times.
- Non-linear compression that adds sonic character via harmonic distortion.
- Requires more output gain, which often raises the noise floor and worsens the signal-to-noise ratio.
FET compressors typically have the following characteristics:
- Very fast attack and release times.
- Non-linear compression that adds character via harmonic distortion.
- Additional saturation due to the inclusion of transformers.
- Requires more output gain, which often raises the noise floor, worsening the signal-to-noise ratio.
Optical compressors typically have the following characteristics:
- Generally low distortion due to the light-dependent resistor (LDR).
- Relatively slow attack and release times.
- Non-linear attack and release controls.
- Frequency-dependent attack.
- Transparent sound.
PWM compressors typically have the following characteristics:
- Very fast attack and release times.
- Very transparent.
Variable-mu (tube) compressors typically have the following characteristics:
- Relatively slow attack and release times.
- Slight harmonic distortion.
- Soft knee.
- Program-dependent ratio.
VCA compressors typically have the following characteristics:
- Very low distortion.
- Flexible threshold, ratio, attack, release and (sometimes) knee parameter controls.
- Transparent sound.
In addition to the actual gain reduction circuit, other design aspects (such as transformers and tubes) will also effectively colour the sound.
Use 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.
The easiest way to set up parallel compression is to send a track (or a group of tracks) to an auxiliary track or bus in a mixer (whether hardware or software). Insert a compressor or limiter on the return channel and hit the signal hard (low threshold, high ratio).
For more info on auxiliary tracks, check out my article Mixing/Recording: What Are Auxiliary Tracks, Sends & Returns?
Once the parallel bus is sufficiently squashed, mix it appropriately with the dry signal from the original track(s). The dry signal should be well-represented with the additional punch of the over-compressed bus signal.
Parallel compression is a particularly effective process for thickening the sound of drums. Try sending the entire drum bus to a parallel compression effects return.
It also works great on vocals to give them more weight or any other instrument that might need a bit of upward compression.
Note that parallel compression is often practiced as I've described above. However, parallel compression is technically happening whenever we duplicate an audio track (or bus tracks together); compress the duplicate/bus, and mix the two.
We can also achieve parallel compression if our compressor happens to have a wet/dry control. Mixing the wet (compressed version) with the dry (original version) of the signal effectively gives us parallel compression.
Experiment with different amounts of compression when practicing parallel compression to find the best results for your mix.
I have a video going into more detail on parallel processing that you can check out here.
Use Sidechain Compression
The sidechain of a compressor is the signal that effectively controls the gain reduction element.
Though the sidechain is most often the input “program” signal (or a slightly processed program signal—often via a high-pass filter), it can also be an entirely different signal altogether (external sidechain).
The selection of the sidechain signal can be visualized with the following signal flow diagrams (the first is a feedback design, and the second is a feed-forward design):
To learn more about feedback and feed-forward compression, check out my article Feedback Vs. Feed-Forward Dynamic Range Compressors In Audio.
“Sidechain compression” generally refers to a compression style where an external sidechain signal controls the gain reduction circuitry. As the controlling signal surpasses the set threshold, the compressor attenuates the program/input signal.
We can add everything from subtle movement to noticeable rhythm to exaggerated pumping to a signal by triggering its compression with another audio sound with sidechain compression.
For example, we can sidechain nearly everything to kick drum and dial in pretty aggressive pumping for the EDM effect. Serial sidechain compression can often get us more aggressive pumping than trying to rely on a single compressor.
As another example, we could sidechain the hats to the snare and add a subtle amount of groove and movement to the hats, and the snare triggers a bit of gain reduction. Adjust the parameters to taste.
We can also use sidechain compression to allow certain elements of a dense mix to pop out, especially if there's significant frequency masking going on. Proper use of sidechain compression can help to improve the separation between elements when they're both present.
For example, we can sidechain the guitar and keyboards to the lead vocal so that they get attenuated gently when the vocal is present and come back up to their original level when the vocal is absent.
We can do a lot with sidechain compression, and I encourage you to get creative with it!
I have a YouTube video dedicated to frequency masking. Check it out here:
Other Compression Tips For Common Elements
- Top 11 Best Tips For Compressing Kick Drums
- Top 10 Best Tips For Compressing Snare Drums
- Top 11 Best Tips For Compressing Bass Guitar
- Top 9 Best Tips For Compressing Electric Guitars
- Top 9 Best Tips For Compressing Acoustic Guitars
- Top 9 Best Tips For Compressing Pianos & Keyboards
- Top 11 Best Tips For Compressing Vocals
- Top 11 Best Tips For Compressing The Mix Bus
Determining the best compressor for your audio needs takes time, knowledge and effort. For this reason, I've created My New Microphone's Comprehensive Compressor Buyer's Guide. Check it out for help in determining your next dynamic range compressor purchases.