Professional Analog Mastering.
Professional Analog Mastering.
220K+ subscribers
Compressors are best known for attenuating peaks - by doing so, we increase headroom. Subsequently, we can amplify the compressed signal without causing unwanted distortion from clipping. This post-compression amplification increases the level of quieter details that were previously masked by louder signals.
So, that’s a simplified version of why we compress and what a compressor does.
However, compressors also affect the frequency response through the introduction of harmonics and intermodulation distortion; they affect the ADSR, which can completely change a signal’s timbre. They wave shape, meaning they can change a sine wave or any other wave type into a completely different shape (which is a big part of why they introduce harmonics).
What’s amazing is that each compressor type you’ve come across does this differently; even emulations of the same gear but by different developers do this differently - they introduce different harmonic orders or formations; their attack and release behave differently, all of which have a huge impact on the ADSR and timbre.
Additionally, modern compressors are introducing functions that drastically change the sound, like lookahead and hold.
Furthermore, the attenuation itself may have unique non-linearities or variations in the input-to-output ratio, contingent upon the frequency of the incoming signal and a slew of other variables.
For example, how an STA tube-leveler compressor attenuates 250Hz will vary from how, say, an 1176 compressor attenuates 250Hz - and no amount of tweaking or adjustment will cause their behavior to match.
The same could be said regarding wave shaping and the subsequent harmonics - even with similar amounts of attenuation a 60Hz. Wave compressed by a Tube Leveler has a distinct shape, while a 60Hz. Wave compressed by an 1176 has a differing and distinct shape.
And this waveshaping isn’t static - in other words, the waveshaping isn’t occurring simultaneously to all peaks and throughs. As the signal is clamped down by the compressor, the wave is affected differently at different points in time, even if the amount of attenuation is the same.
This means that the harmonics occurring at 20ms will be different than the harmonics occurring at 40ms, which means the Hz. Response alterations caused by harmonics will be different at 20ms than at 40ms, and so on.
And to be clear, the plugins I’ve shown so far are very thoroughly designed to have this level of depth. Even plugins that I recommend to you all as free alternatives, as awesome as they are, don’t match their analog counterparts with the same attention to detail.
For example, this 1176 emulation by Analog Obsession, an incredibly talented independent developer, introduces minimal waveshaping. The frequency of the incoming signal, the amount of incoming signal, even the attack and release don’t cause significant alterations to the actual shape of the peaks and troughs.
I want to show you some practical uses for differing compressors so that we can explore how these seemingly esoteric behavioral differences all add up to something we can use.
Watch the video to learn more >
An 1176 compressor on a snare is kind of a classic combination, but why?
The super short attack and release times of the compressor cause significant harmonic distortion, almost if not as much as a saturation plugin on moderate settings.
The most distortion seems to occur with a faster attack, set around 4.5, and the fastest release setting - since these settings are faster than 1 complete oscillation to any frequency below 150Hz give or take, this distortion becomes more pronounced as the compressed frequency becomes lower.
We can also observe that the lower the frequency, the more prevalent the waveshaping.
Practically speaking, if the snare’s fundamental is within this range, then harmonics will form in the low-mids, resulting in what’s often referred to as a warm or full sound.
Additionally, this compressor has an expansion effect on the transient by reducing the sustain and release. At least for this emulation, if the attack is around 4.5, and the release is around 6.5, the initial 10ms off the transient will be significantly higher in amplitude than what comes after. This is particularly true right around the frequencies where we get the crisp and bright sound of a snare drum.
Furthermore, the phase rotation of the transient, in which the positive value is amplified more than the negative, will cause subsequent processors to read the incoming signal as higher in level, meaning their effects will also be more aggressive.
Lastly, we have the simplified reason why we use compressors - since I’ve attenuated peaks, I can now use post-compression makeup gain to increase quieter details.
Let’s listen to a snare processed with these settings, and then I’ll do my best to match the behavior of this compressor with another. Even after tweaking and looking behind the scenes to find all the right values, note that the timbres do not match, as they behave uniquely when compressing.