Depth in a mix draws a listener in, and adds life to a song or composition. Some people believe mixes with depth are the key to making legendary recordings, but how can they be created?
When tracking audio, high quality equipment matters. The microphone, preamp, analog eq, analog compression, and AD converter all need to be high quality. Some engineers don’t use analog EQ or compression, and perform these processes in the DAW only, although there is some disagreement about whether that is the best approach. If you’re looking for high quality, the main components are not cheap. Going to a professional tracking studio is usually the best option.
Auditory masking occurs when the perception of one sound is affected by the presence of another sound. (Gelfand, S.A. (2004) Hearing- An Introduction to Psychological and Physiological Acoustics 4th Ed. New York, Marcel Dekker) Reducing masking while adding character is one of the main things that set professional mixes apart from amateur. So let’s talk about what masking is, and how to deal with it. If you layer several sounds, bass, drums, guitar, and vocals without working to reduce masking, then your mix may sound jumbled and low fidelity. You can adjust volumes, but it doesn’t solve the masking problem. Instead, working to reduce masking can make a huge difference, and the mix becomes more clear. Our video from last year described how to use subtractive EQ to solve masking problems, but let’s dive a little deeper into other ways to minimize masking. There are three main types of masking that mixing engineers should be concerned with: Simultaneous, Temporal and Directional.
Simultaneous masking, also called frequency masking is the most commonly known type. It is sometimes called frequency cancellation, or just plain cancellation. Sometimes it is mistakenly referred to as phase cancellation, which is actually a different concept. Simultaneous masking is when two sounds that are are the same frequency or close to the same frequency occur at the same time. The result is that each sound is heard with less clarity. Tracks that have overlapping frequencies, even if they are not exactly the same frequency, will reduce clarity. Minimizing simultaneous masking is a concern that should be taken into consideration throughout the entire production process from writing, composing, arranging and instrument selection, all the way though tracking, engineering and mixing. There are many things you can do. During composition you can minimize masking in the way that the song is created. If several sounds in the same frequency range occur at the same time, you could put the sounds in different octaves, or far enough apart so that they don’t overlap. You can select instruments that occupy different areas of the frequency spectrum, or just change synth patches so that things are not too close in range. During tracking, you could use different microphones and preamps if you are recording sounds that are necessary to be in similar registers. EQ is another option for solving masking problems. Also, taking the parallel mixing approach is a powerful way to prioritize elements of the mix and avoid masking on the most important parts.
Temporal masking is the masking that occurs before and after simultaneous masking. Masking can occur for up to 200ms after the masking sound has stopped. In that situation it is called forward-masking or post-masking. A sound can even be masked up to 20ms before the masking sound occurs, which is called backward masking or pre-masking. The amount of masking changes with the difference in loudness. As the masking sound gets louder, the level of masking changes. Temporal masking is part of the scientific psychoacoustic model of auditory perception, and has applications in audio codecs and other areas. For mixing engineers, temporal masking is mostly significant in applying reverb. You may want to insert a short predelay to help reduce temporal masking of your wet reverb tracks.
Directional masking occurs when something violates the natural way that a listener senses depth. Directional masking is usually at its very greatest when stereo or multi-channel recordings are converted to mono. Directional masking is a huge concern in mixing. It can be reduced in stereo recordings by using very high quality stereo reverb and delay processors which have algorithms that help reduce directional masking, or by manually reducing it by panning the reverb and delay signals away from the dry signal. Let’s talk a little bit more about this.
Reverb and delay are key for creating the magic of stereo and multichannel recordings. With that said, you have to be really careful, because using them incorrectly can do more harm than good. If too much reverb is applied, or applied in the wrong way, it can cause simultaneous, temporal and directional masking problems. Too little reverb and delay can lead to mixes that don’t give the listener a sense of space. It’s that sense of space that can really draw a listener in, no matter what style of music. For example, if you combine a few dry tracks in a mix with no reverb or delay, the mix may sound empty and uninteresting. If you tastefully add reverb and delay, you can have a sound picture that feels full and interesting. Too much reverb can lead to overlays that end up making the entire mix feel muddy. When you listen to muddy mixes, you may feel that your equipment isn’t good enough, your ears aren’t good enough, or your skills aren’t good enough. Finding this balance doesn’t require any equipment or a new set of ears, only a little bit of knowledge. Let’s talk about how to create a balanced mix with reverb, and avoid masking.
There are times when reverb or delay should and shouldn’t be used. Let’s talk about this. There are typically two spaces for tracking. First, you may have an acoustically designed space that sounds excellent. The second is the most common, which is a dead room which has been treated to minimize room reverb. If you have a home or project studio, you should probably aim for a dead room, because well designed acoustic spaces are extremely expensive and usually require major building modifications. Dead rooms are the most common. If you have an acoustically designed space that sounds great, you probably won’t want to add much artificial reverb to tracks recorded in it, because you will be using the natural reverb of the space you are recording in, and reverb times can be accomplished with mic placement. If you’re working in a dead room, you will need to add artificial reverb and delay to give the sense of space that makes great sounding mixes. Although today’s recordings are trending to be more dry, good mixes give the listener some sense of depth. Skillfull use of stereo and multichannel is mostly about the skill of using reverb and delay.
The main thing to think about is simultaneous masking. Here’s an example of avoiding masking using reverb or delay. Lets say you have two sources in a similar frequency range, you will want to carefully consider the technique and amount of reverb you use on those tracks. You might pan the dry tracks into opposite speakers, and pan their wet reverb or delay track opposite from those
panned tracks. This is an old trick that is still useful. Some modern reverb processors have ways to achieve minimized masking with even more realism than is possible with this technique.
Next, if you have several tracks of the same instrument, you might bus them out to one channel and apply reverb to that entire channel so you don’t have several reverbs tails layered and masking each other. (on the little icon on the left, make four of them forming a little box, and have them separate with a real greyed out reverb tail picture, and then combine them and show it not
greyed out). Next, you should consider temporal masking, and separate the reverb from the direct sound (do a black shadow around the instrument to separate it from the smear, have each picture on a row, sort of like tracks). This can be accomplished by inserting a pre-delay, which is available on most reverb processors. There are some engineers who will not use a reverb that doesn’t have a pre-delay feature.
The more reverb that is on a signal, the farther away it seems to be in the mix. Turning down a track’s levels and adding a comparitively higher amount of reverb will make it seem more in the background to the listener. This is an effective way of creating a perception of depth.
Also, delay using the Haas effect is effective in creating depth and fullness with less masking than reverb. With the Haas effect, you are emulating the first reflections of an acoustic environment, and giving the listener a sense of depth without the sound of reverb or the echo normally associated with delay. Its best if only used on one or maybe two tracks in an
entire mix. It will cause problems if overused. To apply the Haas effect, you will want to use short delay times between 10 and 40 milliseconds. In addition to that, the dry and wet sources are panned into opposite speakers and the delayed track is set at a lower volume level. This will help reduce masking and set the track on the sound stage. The closer the delay time is
to the 40ms limit, the further back the track will sit. To achieve depth, Haas delays can be used instead of reverb, or alongside tracks that use reverb.
Another effect that can be used is phase inversion. The phase can be inverted on the left or right side of a stereo track inside a mix. This effect can make it sound as if the individual track being played is totally outside of the mix and the listening space. This effect is commonly viewed as a novelty and shouldn’t be overused. There are several problems that this effect can cause, and some engineers never use it. Those that do usually only use it on very selected occasions and on very selected tracks. Tracks using this effect will disappear if your stereo mix is converted to mono. Reverb most often should not be used with bass instruments, or on instruments with bass or lower midrange frequencies without very careful settings. Reverb can really destroy the clarity and punch of bass frequenices, so you’ve got to watch out.
Let’s think about sound in the natural world. If a sound occurs close to a listener, more high frequencies will be heard. The further away the sound, the less high frequencies we hear. Some engineers work to achieve depth by adding or reducing high frequencies. While this is a commonly used method for achieving depth, there are some engineers who feel that adding or reduce high frequencies to try to achieve depth sounds unnatural.
If you’re trying to create depth in a mix that is based on groove, such as a dance, electronic or hip hop track, make sure that the elements that create the groove are prominent and are tastefully forward in the mix. These elements are most often the bass, snare and bass drum. If you’re using an 808 kick, then you might try adding a little 60hz and 1.5khz EQ bump for extra punchyness. If you are having a problem getting your bass drum to sit where you want, then you probably are experiencing some masking problems, as we discussed before.
CAREFUL WITH THE COMPRESSION
Some of the world’s top audio engineers may suggest that you use no compression at all. There are others who compress everything. Normally, less compression gives more depth.
Do you want to get the tone out of your tracks that compression can sometimes add, but maintain the depth? There’s a way to do it and it’s called parallel compression or New York compression, and it’s highly effective for achieving elusive modern drum tones. To accomplish this, you need to either bounce your drum tracks to two stereo tracks that line up perfectly, or create two identical drum buses. Then, put compression on one of the tracks or buses, with a relatively high ratio and threshold, with the release time set perhaps between 250 and 400 and the attack set for maximum punch. Mix the compressed track in with the original track very lightly, perhaps all the way down to -20dBFS. This technique provides a very natural, transparent compression that can work for drums, vocals, or most any source.
When recording a source, making a stereo recording with a stereo microphone configuration, instead of recording in mono can add very natural depth. Vocals or main harmony instruments may still need to be recorded in mono. This is something to try out and might not work for everyone, it may not be best for bad sounding rooms or very dead rooms. There is more about this approach in our video about mixing techniques.
Do you want to retain depth, even with low-budget equipment or less than professional mixing? Well, the answer might not be what you think. At Stonebridge Mastering, our clients usually want both the highest possible sound quality and the highest possible volume levels. The complication is that the volume level to which something can be pushed usually depends on how well it was tracked and mixed, and the quality of the equipment used. Of course, you can push an album to the level of the loudest pop recording, but it will not retain its depth and sound quality. Million dollar studio recordings can often be pushed to extreme levels and maintain relatively higher depth. A home studio or project studio recording can sometimes stand up against those million dollar recordings, but almost always cannot be pushed to the same volume levels and still sound good. When a loud recording, and a lower volume recording are played over an FM station, they usually go through “Orban” processing, or similar processing. That processing will make both the loud and soft recordings sound as if they’re at the same volume level. Sometimes, the softer recording can even seem louder. Therefore, the idea of a louder recording making a bigger impact on the radio is only a myth. Recordings that are not created by skilled professionals in professional facilities usually need to be at a lower volume level to be able to compete. Pushing them to the same levels as professional recordings can seriously damage their chances of competing. The only difference is that the listener has to turn their volume control a little higher, which will probably never even be noticed.
You must be logged in to post a comment.