The MC2000 plug-in has the power and flexibility to emulate a wide variety of compressors in two, three, and four band configurations. But when the honeymoon of multi-band vintage emulations ends (if that were possible), the numerous uses of the MC2000 become apparent. This edition of Colin’s Corner focuses on the wide range of applications that have made the MC2000 one of the most popular multi-band compressor plug-ins ever created.
Why Multi-Band Compression?
The idea behind multi-band processing is that by breaking down the audio into ranges of frequencies, referred to as ‘bands’, each band can be processed separately and therefore more accurately. Multi-band compression first starts with a set of filters, commonly referred to as a ‘crossover’, to split the signal into selected frequency ranges. Once separated, each band can be fed into a compressor. Then the signals are recombined into a full range signal.
The wide range of frequencies a single instrument or voice can make the task of ‘good’ dynamic range compression very challenging, even for the seasoned engineer. The low end of a vocal plosive could falsely trigger a compressor right in the middle of some important dialog or lyric. The thump of a kick drum might have a good bottom end, and really only need compression in its mid range. And what about the powerful singer who is all over the map when it comes to dynamics and pitch? Throw in a touch off ‘essing’ and you’ll wish your day job was cleaning out chicken coups (but let’s hope not).
Thankfully, the clever folks at McDSP have created multi-band compression algorithms in the MC2000 plug-in to deal with all the above, and then some.
Many multi-band compression applications start from a simple way of setting up the controls. Below is a typical multi-band compression approach I have found works well in many situations.
Choose a 4-band configuration to start – you get the most flexibility and can always experiment with fewer bands after getting the sound you want. The default Knee, Bite, and Ratio controls are good starting points.
Leave the cross over point frequencies at their defaults (100 Hz, 1000 Hz, and 10 kHz). For vocals and dialog, you might find better use from the highest band (band 4) by changing the 10 kHz default to 6 or 8 kHz.
Setup these attack and release settings:
- Band 1: 3 msec attack, 300 msec release
- Band 2: 2 msec attack, 200 msec release
- Band 3: 1 msec attack, 100 msec release
- Band 4: 0.5 msec attack, 50 msec release
Lower each threshold equally until gain reduction is shown in three of the four bands (all four bands is ok, but if you are going for minimal compression try to only trigger compression in 3 of the 4 bands).
Adjust input pad level if needed (like on the Venue where the input to the plug-in may be artificially low). Be sure to watch for clipping.
Adjust output level as needed. Again watch for clipping.
Some variations to the above include increasing ratio settings for each band (ex: Band 1 Ratio = 2:1, Band 2 Ratio = 3:1, Band 3 Ratio = 4:1, Band 4 Ratio = 8:1) and/or different threshold settings for each band.
Once you start to experiment with different input thresholds your into the ‘heavy adjustment area’. This can lead to a better sound, or get you in trouble. Be careful out there.
Great singers have one primary problem. They are great singers. Their impassioned dynamic performance is nearly perfect – you just have to mix it with the rest of the band. Breaking the ‘problems’ into three categories, they are – plosives, ess-es, and the actual vocal itself.
Fortunately, the MC2000 multi-band compressor takes care of all three issues. Following the basic operational tips in the previous section, setup the MC2000 to trigger on 3 of the 4 bands. Most likely, the highest band is the one that is triggering the least. For starters, lower the highest crossover point to 6 or 8 kHz as suggested above to get more of the vocal into the highest band.
In order to de-ess the vocal, reduce the highest band (band 4) threshold until it triggers during the ‘sss’ events. The faster attack and release times will insure each ‘sss’ is reduced sufficiently. Try a higher ratio (8:1 or higher) in band 4, while keeping the rest of the ratios at 4:1.
The vocal plosives are usually low frequency content, and can be handled in nearly the same way as the ‘sss’ events. Using the lowest band (band 1) reduce the threshold to be about 6 to 9 dB lower then the threshold in bands 2 and 3, where the primary vocal content is assumed to reside. Keep the attack and release times ‘slow’ as listed above – attack at 3 msec, release at 300 msec. You may find faster attack and release times are more effective, but try to keep the attack time longer than 1 msec, and the release time longer than 100 msec, to avoid distortion artifacts.
By leaving the compression bands 2 and 3 to handle the rest of the vocal, you have a good chance of compressing just enough of the performance at a variety of pitches to please the client, the performer, and hopefully yourself!
Using a multi-band compressor on a drum kit mix can be an effective way to get some balance (or fix some problems) in the sound. While individual EQ and compression on the kick, snare, etc. may have already been setup on the separate tracks, it is very convenient to look at the four compression band meters (inputs, outputs, gain reduction metering in each band you know!) to see if the overall kit mix has the right blend of lows, mids and highs.
If any one band looks to be too loud in comparison to the others, you can either 1) reduce the makeup gain in that band (and not apply any compression), or 2) reduce the threshold of the offending band (with the basic settings described previously in the ‘Getting Started’ section) until enough gain reduction is occurring to balance out the sound. You can even add some low end or high end to the drum mix to get a different kit sound altogether.
For drums and other percussive instruments, a faster set of attack and release times might be better. An example set is listed below:
Drum/percussion faster attack and release settings:
- Band 1: 2 msec attack, 200 msec release
- Band 2: 1 msec attack, 100 msec release
- Band 3: 0.5 msec attack, 75 msec release
- Band 4: 0.25 msec attack, 50 msec release
Note this faster set still follows the basic pattern – for each higher band the attack time is 2x faster, and the release time is 1.5 to 2x faster. Be careful to not get too fast, or else you might introduce some distortion into the mix.
Trying some other variations in the attack/release times is useful for getting that ‘thwack sound’. For example, when you have a drum kit sound that is ok, lower the threshold in band 3 until about 12 to 18 dB of gain reduction is occurring. Then increase the attack to about 3 msec, with a short release time of 30-50 msec. Increase the makeup gain in band 3 to roughly match the amount of gain reduction. The pop of the snare will fly right through this compressor, giving the sound a more exciting effect. You can also get a similar effect by instead increasing the BITE control to 3.0 to 5.0 – this will allow transients to pass the compressor relatively unaffected.
For that matter – once you have a drum kit mix you like, just alter the Knee control of the MC2000. The range of the MC2000 (and CompressorBank) Knee control covers several families of classic compressors, as listed below:
- dbx 165 (soft knee, over-easy sound): Knee values between –10.0 and –3.0
- 1176 (modest ‘analog’ knee): Knee values between –3.0 and 0.0, Hard knee: Knee value at 0.0
- Neve 33609 / 2254E (slight overshoot): Knee values between 0.0 and 4.0
- LA2A, LA3A, Opto comps and limiters: Knee values between 4.0 and 15.0
Indeed, the mind-boggling knee shape combinations of the MC2000 multi-band compressor are pretty darn cool, in my completely biased opinion. So be sure to experiment with them and come up with your own opinions too!
Like the vocal performance, the fret play (noise?), aggressive strumming, combination of chords and single notes, and wide dynamic and pitch range make the acoustic guitar hard to deal with at times.
Use the same basic ‘getting started’ settings as described earlier, along with the adjustment of the highest cross over point (between bands 3 and 4) to somewhere between 6 and 8 kHz. The low band (band 1) can deal with the ‘thumping’ from aggressive strumming and other playing styles, the mid bands (bands 2 and 3) can deal with the chords and solos, and the high band (band 4) can be used to deal with any fret squeaks.
As an alternative, the high band may come in handy to add some ‘air’ – some extra of gain from the makeup gain of the band 4 compressor – say 3 to 6 dB, to brighten up an otherwise dull guitar.
All the above sounds great in the studio, and the MC2000 is great for live sound applications too!
The ultra low latency (4 samples) makes the MC2000 perfect for the any part of the mix. Even in ear monitoring sends can benefit from the multi-band compression capabilities of the MC2000.
The MC2000 input pad (up to 12 dB) comes in handy on the Venue systems that are setup to reduce all audio buss levels to avoid digital clipping. The plug-ins get signal at a reduced level – so you need a make up gain at the front end of the plug-in, and then can accordingly turn levels down with the output control.
While we’re on this topic – other McDSP plug-ins that have input and output level controls include Analog Channel, Channel G, Channel G Compact (custom made for the Venue by the way!), Chrome Tone FilterBank, FutzBox, the afore mentioned MC2000, and NF575 (noise filter). If you have not tried out the McDSP Live Pack bundle maybe it’s time to do so!