When we boost or cut a frequency band with EQ, the phase of that frequency band is shifted as it is delayed by a fraction of a millisecond. This means that the phase relationship of the source audio is altered when using EQ. Sometimes these phase shifts are not noticeable, or may even be desirable in your mix. Other times, they may be hurting it.
You may wonder then, is there a way to minimize the effects of phase shift when using EQ? Well, in today’s digital age, we have the option to choose between regular (or minimum phase EQ) and linear phase EQ. Some equalizers, such as Fabfilter Pro Q, allow you to choose between minimum or linear phase mode for enhanced flexibility.
What Is The Difference Between Minimum Phase and Linear Phase EQ?
Minimum phase EQ shifts the phase of individual frequency bands, while linear phase EQ shifts the phase of the entire signal, keeping its phase relationship intact. While this may sound like the superior option, it can also have some downsides. Ultimately, it is important to use your ears and decide which EQ is best for each track.
How Does EQ Create Phase Shift?
Now you might be wondering, how exactly does phase shift occur when using EQ? This concept comes from the way that analog equalizers were modeled back in the day. An incoming signal was split into individual frequency bands, and those signals had to physically travel through various capacitors and inductors in order to create a frequency boost or cut.
This would delay the affected frequencies in time, resulting in a phase shift. The phase-shifted signal would then be combined with the original signal at the output of the EQ. The addition and cancellation of the frequencies in both signals cause the change in frequency content that we hear.
This type of EQ is referred to as minimum phase EQ and most EQ plugins would be considered minimum phase. This means that the equalizer is designed to create the least amount of phase shift possible.
As mentioned previously, this phase interaction is not always a bad thing. Engineers like the “colorization” that can come from using traditional EQ. Certain EQs may be chosen for their specific character (or the way that they shift the phase).
In some instances, however, it may result in a less desirable sound overall. This effect is known as phase smearing. Linear phase EQ was created to be a potential solution to this issue.
So What Is Linear Phase EQ?
Linear phase EQ also shifts phase, but instead of only shifting the frequencies being boosted or attenuated, it shifts the phase of the entire signal.
This gives you the ability to make the frequency adjustments you want, without affecting the phase relationship of other frequencies in your audio.
When Should I Use Linear Phase EQs?
Linear phase EQ can be a useful tool in situations where an audio source is heard on more than one track. Some examples of this are:
1. An audio source with multiple microphones on it
Using linear phase EQ can be useful when you are processing an audio source with multiple microphones on it. An obvious example of this would be drums. Even with gating and other plugins in place to isolate each drum, there will still be some bleed from other drums coming into each microphone.
The kick drum will get into the snare mic, the snare into the tom mics, etc. Adjusting the phase relationships between all the microphones with EQ could affect the sound of the drum kit quite drastically. This where using linear phase EQ could come in handy. You can make the adjustments you want without causing any phase smearing among the other drum mics.
Adjusting the phase relationship between the two mics with EQ could affect the tone of the guitars in undesirable ways. Again, linear phase EQ would allow you to make the necessary adjustments without affecting the relationship between the two microphones.
3. Parallel Processing
Linear phase EQ could be useful in parallel processing chains, such as parallel compression or distortion. Adjusting the phase relationship between the source audio and the parallel processed audio could alter the tone of the track in unintended ways. Using linear phase EQ here would ensure that this doesn’t happen.
Linear phase EQ can also be useful when making very narrow boosts or cuts. With minimum phase EQ, the degree of phase shifting is greater as your EQ curve becomes steeper in slope.
Using linear phase EQ for these adjustments can prevent more phase smearing from happening due to a narrow boost or cut. This is why some mastering engineers like using them in their workflow. The mastering process often involves making minute EQ adjustments to a mix.
Linear phase EQs give the mastering engineer the ability to make those adjustments without worrying about how the phase shift will affect the rest of the mix.
Linear phase EQ may sound like a great tool so far, but there are some downsides you should consider before throwing it all over your mixes.
CPU and Latency Issues
Linear phase EQs use a higher amount of CPU than minimum phase EQs do. If you use too many of them, your DAW will run very slowly.
Latency may also become an issue when using too many linear phase EQs , causing delays in your tracks when they are played back. Any plugins used on your tracks will delay them by a certain amount of samples. Most DAWs have a delay compensation feature to avoid latency issues. This delays all tracks by the same amount of samples so that they will all play back at the same time.
Even with this feature in place, linear phase EQs can push the limits of your DAW’s delay compensation. Linear phase EQs can delay a track anywhere from 3,000 to 20,000 samples. If you are working in Pro Tools, for example, there is a limit of 16,000 samples at a sample rate of 44.1 kHz. You do get a higher limit if you increase the sample rate of your session, but linear phase EQs can still take up a hefty amount of it.
If you want to use linear phase EQ in your mixes, it is best to use them when you have a specific need for them, such as the ones discussed in this article. That way you can avoid taxing your CPU and eliminate any undesirable latency.
One of the most notable artifacts created by linear phase EQs is pre-ringing. Pre-ringing is a resonance similar to a reverse reverb that you might hear before your audio begins. It is actually the sound of the EQ shifting the phase of the entire signal.
It takes a lot of processing power to do this as we discussed, so you will audibly hear the shift as pre-ring because it does not happen instantaneously. It is more noticeable if there are a lot of transients in a track. For example, you might notice a low-end ringing right before a kick drum transient.
Can You Get Rid of Pre-Ringing?
While there is no way to completely get rid of pre-ringing, there are ways to make it less audible in your mix if you need to use a linear phase EQ. Here are a few tips you can try:
Avoid Extreme Boosts
The more you boost a frequency, the greater the amplitude of the pre-ring.
Cutting frequencies makes pre-ringing less audible than boosts. If you are going to boost frequencies with linear EQ, try to only make minor boosts of a few dB.
Focus on Adjusting High Frequencies (If Possible)
While the amplitude of the pre-ring is constant across the frequency spectrum, the duration of it shortens as you go higher in frequency. If you can, try to use linear phase EQs to adjust higher frequencies to avoid excessively long pre-ringing. If you cannot do this, go for more subtle boosts in the low end.
Pay Attention to Your EQ Curve
The relationship between pre-ringing and EQ curves, or Q value, is complex. A narrow Q value (i.e. 5 or above) will increase the duration of pre-ringing, but reduce the amplitude of it, and vice versa when the Q value is widened (i.e. 5 or below).
There is not really an ideal scenario here, so it is best to experiment with the Q value to see what gives you the best sound.
Lack of “Color”
As we discussed earlier, sometimes the phase shifts that come from using minimum phase EQs create a desirable sound that we want in our mixes. This “coloration” is something you cannot get with linear phase EQ.
As we can see, there are pros and cons to both types of equalizers. Minimum phase EQs can add a nice character to your tracks, but the phase shifting that occurs can damage the relationship between tracks if you are working with stereo or multiple mic sources.
Linear phase EQs can help keep the phase relationship intact in these situations but can take up a lot of CPU, cause latency issues and introduce unwanted resonances from pre-ringing.
As with anything in music production, there is no one size fits all solution. Both types of equalizers are good tools to have for different reasons.
Ultimately, if you have both, you will have to experiment with them in your mixes and use your ears to decide which type serves your tracks best. If you do not have a linear phase EQ, remember that plenty of great records have been made with good ol’ fashioned equalization, so don’t fret. Ultimately, linear phase EQ is not needed for good mixes, but it certainly makes mixing easier when you need it.