Review – Pro C 2 from Fab Filter
After 8 years, FabFilter’s long-awaited first major update to the popular Pro-C general-purpose compressor is finally here! No kidding, this could be the only compressor you’ll ever need.
by Dave Townsend, Sept. 2015
FabFilter Pro-C 2
Since its introduction way back in 2007, FabFilter’s Pro-C compressor has become one of the most popular in-the-box mixing tools of its type. For many of us, Pro-C was the gateway drug that led us happily into the FabFilter cult.
Yeah, you can count me among the believers. Pro-C has been my go-to compressor for most tasks since 2008. After this update, “most tasks” will likely become “pretty much everything” for me, because Pro-C is now even more versatile than before, even blurring the lines between compressor, limiter and gate.
Given its long-established popularity, I won’t waste your time with a general Pro-C introduction. It’s been reviewed and discussed so many times, and expertly demonstrated by Dan Worrall’s insidiously-persuasive videos that it’s unlikely you haven’t already heard quite a lot about Pro-C. Instead, I’m going to assume you’re either already a Pro-C user or at least have been thinking about joining the club, and focus on what’s new and improved in version 2 – as well as a couple of minor features that have been lost or deprecated.
What got everybody excited back in ’07 was that Pro-C actually showed you – visually – what it was doing. That scrolling graph was a literal revelation. (Yes, I can hear the “use your ears” naysayers’ chorus of disapproval – I don’t care. Visual aids are fun! And occasionally useful.)
When rumors began circulating that Pro-C 2 was under development, there was plenty of speculation about what new features might be added (most of which we got). But nobody was seriously asking for a better graphical user interface. I mean, what more could you possibly pack in there?
Well, surprise! Those clever Dutch boys did indeed figure out how to squeeze even more helpful information into the GUI. Such as a spectral display for the sidechain filter, a picture of the transfer curve overlayed with signal levels, and – my favorite – the circular level indicator around the Threshold knob (first introduced in Pro-MB).
The basic meat of the levels graph remains the same, which is good because a) I’ve become quite comfortable with it, and b) it was already perfect.
Aside from being larger and resizable (including a full-screen mode), the levels display hasn’t changed much from version 1. You still have three graphs showing input and output levels plus the amount of compression. If you don’t want to see the graphs at all, or need to minimize CPU usage, you can still hide them as before. However, now when the levels graph goes away the whole UI gets smaller, which is great when you have a bunch of Pro-C instances onscreen at once.
Unfortunately, version 2 no longer lets you adjust the opacity of each graph. I guess that’s OK, it’s not a feature I used very often.
The UI is only continuously-resizable in the VST3 version. With the VST2 version, you have a choice of four screen sizes: small, medium, large and full-screen.
Also beware of this potential gotcha: if you choose the large size, and it’s too big to fit onto your screen, the size-selection button slides out of view so you can’t click it to get back to a useable size! If this happens, click the Full-Screen button, which is in the upper-right corner and always visible. In full-screen, the resize button in the lower-right will now be accessible and you can choose a better UI size for your monitor.
The Side-Chain Section
The biggest GUI changes are in the Sidechain section. It now sports a spectral graph over the sidechain filter, and the filter section now has a new bandpass filter in addition to the previous high-pass/low-pass filters. The high- and low-pass filters are no longer fixed at 12dB/octave, but may now be set to one of 9 different slopes, from 6 to 96 dB/octave. This is going to open up new creative possibilities and enable greater precision for responding to specific elements in a track.
The band-pass filter can be either fully-parametric or left at the default “automatic” mode, which positions itself midway between the high- and low-pass filters’ cutoff points and automatically becomes a boost when narrowly focusing in on a specific frequency. Great for quickly targeting a hi-hat that’s bleeding into a snare track.
But here’s the best new side-chain feature: the Audition button. Click this to hear just the side-chain signal, whether internal or external. What you’re hearing is what’s going to the detector circuit post-filter, so if you’re wanting to key in on something very specific such as a hi-hat, a vocal sibilance or a room resonance, it becomes very fast and easy to dial in the sidechain filters.
Mid-side channel linking has been simplified for greater ease-of-use, at the expense of making M/S compression slightly less versatile. It is no longer possible to adjust Mid and Side compression independently without using two instances of Pro-C2. However, the more common M/S scenarios are now simpler – more on that in a moment.
A welcome change is that you no longer have to load specific versions of Pro-C in order to use the external sidechain input. Whereas version 1 came as four separate DLLs (mono, mono w/sidechain, stereo and stereo w/sidechain), version 2 dispenses with the separate sidechain-enabled versions. Pro-C 2 consists of only two DLLs, one for mono and one for stereo.
One of the keys to Pro-C’s versatility was that it featured three separate compressor models (Opto, Clean and Classic). In Clean mode, it was a surgical tool, while in Opto mode it was a quick set-and-forget vocal, bass or drum bus compressor with LA-2A-type soft-knee attack and program-dependent release characteristics. Classic mode is a vintage feedback-style model. (In version 1, Classic mode had a bit of an aliasing problem under certain circumstances, but that is no longer an issue with the new oversampling feature.)
Pro-C 2 takes the modeling idea further, adding five new models to the list. For example, you now have a “Vocal” mode optimized for vocals that uses automatic knee and ratio settings so that dialing it in is basically a matter of tweaking the threshold.
If you’re into EDM or other modern styles that incorporate compressor pumping, Pro-C2 has a special model just for you, called “Pumping”. As a classic rock guy I wondered if this mode would have any appeal for me. Certainly not across the master bus, where I usually go out of my way to avoid pumping. But I’m actually liking it on individual tracks and sub-mix busses, especially the drum bus.
Rounding out the new compressor models are “Mastering” (ultra-clean), “Bus” (for glue, especially drum busses), and “Punch” (quick-acting transient enhancement). Of course, all of these names are just suggestions, so don’t feel that any of them are limited to one application. The Vocal model, for example, works great on bass tracks.
The Basic Parameters: Attack, Release, Threshold and Knee
If you’re looking for the Input knob on Pro-C2, which used to sit to the left of the Threshold control. It’s not there anymore. An odd thing to take away, but no big deal. You can simply use your DAW’s track gain slider instead.
Aside from the missing Input knob, the other controls look the same as before, and you still have the helpful auto-release and auto-gain options for quick ‘n dirty setup. But even these standard controls have acquired a couple tricks and changes.
The range for Release times has been shifted downward. Whereas Pro-C 1 gave us 50 to 5000 milliseconds, with Pro-C 2 it’s now 10 to 2500 milliseconds. That faster release time allows for very fast peak control on percussive material. I haven’t decided yet whether not being able to set release to 5 seconds anymore is a significant limitation – just because I’ve never done it doesn’t mean I might want to someday.
The threshold parameter has been extended, down to -60 dB from the previous limit of -36 dB. What this means is you can have continuous compression happening on even very quiet tracks, useful for leveling purposes.
What I like best about the Threshold knob is the ring-shaped level meter that wraps around it. I find this to be easier for dialing in the threshold setting than the traditional knee graph.
The transfer curve is now quite variable. Whereas version 1 just gave you a choice of soft or hard knee, you can now do anything in between, from an extremely gentle knee for unobtrusive leveling or tape-like saturation to a brickwall backstop for taming peaks. The new Range control further expands what can be done with a custom transfer curve – more on that later.
Attack times can now be much shorter – down to 5 microseconds (previous minimum was 50). That gets Pro-C down into ultra-fast FET territory. Beware, though, that such ultra-fast attacks will cause audible distortion on low frequencies, and increase the likelihood of aliasing. When using very short attack times, make sure to enable oversampling and possibly lookahead.
Speaking of lookahead, this is another advanced feature that Pro-C didn’t have before. To be fair, it’s never been a common feature on general-purpose compressors, being something you’d normally find only on limiters.
If you’re not familiar with lookahead, it’s an added buffer that lets the processor “look ahead” to see what’s coming down the line, allowing more time to react to abrupt changes in signal level. It’s one example of something you can only accomplish with a digital compressor, since it’s very difficult to implement in conventional analog gear. But Pro-C has never sought to emulate any piece of vintage hardware.
You’ll want to use lookahead mainly when using Pro-C2 to control peaks, in particular for very dynamic percussive material. Without lookahead, the plugin may not be able to react fast enough to sudden spikes and might therefore let some excessive levels slip through. Lookahead lets it see that spike coming and figure out how to handle it transparently.
Be aware that when lookahead is enabled, there will be a fixed latency of 20 milliseconds. Normally, Pro-C has no latency, but lookahead necessitates the delay in order to work. There has been some criticism of FabFilter’s decision to fix the latency at 20 ms – some users reflexively hate any latency at all – but having a fixed delay means you can experiment with various settings and even automate lookahead without the glitching that usually results from resizing the buffer during playback. In practice, lookahead isn’t something you’d enable during tracking anyway, and the added latency has zero impact while mixing.
“Hold” is a new feature in version 2, something you’d more commonly see on gates but rarely on compressors. What it does is force the compression amount to remain at maximum for N milliseconds after the sidechain signal drops below the threshold, where N is between 0 and 500 milliseconds.
Normally, when the sidechain signal drops below the threshold, the Release phase is initiated which begins to lessen compression at a rate determined by the Release parameter. In order to prevent pumping, we often lengthen the release time. The Hold feature can be used that way, too, by delaying the start of the release phase and extending the amount of time that peaks are clamped down.
However, it gets interesting when you couple a long Hold parameter with a short Release. Now, the trailing edge of the peak does not immediately trigger the Release phase but instead starts a countdown timer. After N milliseconds’ Hold time, the release phase begins, but because we’ve set a fast release the amount of compression rapidly switches from a high value to a low value, almost like a gate.
Try this: set the hold time to a long value based on the project tempo, e.g. 125ms for 120bpm (quarter notes are 125 milliseconds apart at 120 beats per minute). Abrupt releases will now occur in time with the song’s beat.
The Range Control
This is a new feature for Pro-C, one I’ve long wished for after being introduced to the concept in other dynamics processors. What it does is set an upper limit to the amount of compression that will be applied.
By default, the Range parameter is set to 60 dB. That means Pro-C2 is allowed to apply up to 60 decibels of attenuation. This setting pretty much turns the Range feature off, since 60 dB is far more attenuation than a compressor would ever apply under normal circumstances.
But if you turn the Range down to 10 dB, that instructs the compressor to never apply more than ten decibels of gain reduction, regardless of the signal level or ratio. The transfer curve for such a limited range looks like that pictured right.
What we’re saying is “apply the given ratio up until the gain reduction is 10dB; after that, let it go”.
Now, this might seem kinda wacky at first glance. After all, a compressor’s gotta compress, right? Why put it on a leash? Isn’t that dangerous?
Well, yes. On the wrong material, telling the compressor to ignore the highest peaks could indeed lead to problems.
But think about what this action would have on source material that’s already fairly level: it’s going to make the highest peaks higher in relation to the average signal level. In other words, upwards-expansion. The louder parts get louder. It can breathe life into an otherwise dynamically-challenged track.
Just be careful where you set the threshold. Such expansion can cause audible artifacts such as reverb and cymbal tails dropping off unnaturally, or allowing a higher-than-average peak to jump into the red.
One of Pro-C’s neatest features is the ability to compress just the Mid or just the Side channel. One application is as a widening technique, by compressing only the Mid channel and thus drawing attention to the sides and accentuating L/R differences for a greater sense of width.
Both Pro-C versions can do this. It bears mentioning here mainly because it looks different now. The intention was to make it simpler for users; how successful they were in that depends on how accustomed you were to the old way of doing things. I found the new way disorienting at first, but only for minute, before realizing that the new layout was indeed cleaner and more sensible.
New in version 2 is a Stereo Link Mode switch, which lets you choose between 4 ways to trigger and apply compression: Mid, Side, M>S and S>M. These modes determine whether the detector listens to the Mid (“Mid and “M>S”) or the Side channel (“Side” and “S>M”), and whether attenuation is applied to the Mid (“Mid” and “S>M”) or the Side (“Side” or “M>S”) channel.
Mode Listens To Compresses
Mid Mid Mid
Side Side Side
M>S Mid Side
S>M Side Mid
You can get some interesting widening effects by compressing the Mid channel while detecting the Side level (“S>M” mode). Or help clear the center for the lead vocal by placing Pro-C on a rhythm guitar bus, compressing the Mid channel while detecting the lead vocal via the external sidechain.
Here’s what you lose from version 1: before, you could adjust the Mid/Side linkage anywhere between 0 and 100%. Now, it’s either 0 or 100%, nothing in-between. In my opinion, it’s a trivial loss of functionality that’s justified by the new, easier-to-grasp layout.
In addition to conventional internal and external sidechain triggering, Pro-C 2 introduces a MIDI-triggering option that lets you initiate compression via MIDI notes. It’s a pretty rudimentary feature – any note-on event causes maximum compression to occur. What can you do with it? Mainly, it’s for EDM-style rhythmic pumping effects.
Metering remains a very strong feature, as it is in all FabFilter products, with a lot of information packed into a compact space.
You’ve got several useful pieces of information here:
- Momentary loudness
- Momentary peak
- Maximum gain reduction
- Momentary gain reduction
- Maximum peak
- Maximum peak and gain reduction in decibels
The graph can be scaled from 9 to 90 dB full-scale, and may be displayed vertically or horizontally. All of these features remain the same as in version 1. What’s changed is that Pro-C 1’s meters displayed average RMS (50ms window), while Pro-C 2 now displays perceived-loudness levels based on the Momentary mode of the EBU R128 / ITU-R 1770 standards for loudness measurement.
Upgrading from Pro-C 1
Pro-C 2 is a completely separate product from Pro-C 1. The two versions are not interchangeable, nor directly replaceable, and any existing automation from version 1 will not be recognized by version 2.
However, because they are separate products they can happily co-exist on disk and within your projects.
One thing the two versions do share are presets, so if you really want to replace an instance of Pro-C1 with Pro-C 2 you can save the original’s settings as a preset and Pro-C 2 will be able to read it.
Buying or Demoing Pro-C 2
Doubts about Pro-C? Watch this video by Dan Worrall, perhaps the world’s best plugin salesman:
Download the product from the vendor site here:
FabFilter does not have separate demo versions. The product you download will be a fully-functioning demo for two weeks. When you buy Pro-C 2 you’ll be emailed a license file that activates the DLLs.
This is my preferred form of copy protection: no dongle and no internet connection is required to activate FabFilter products, the plugin won’t stop working if you disable your network card or replace your motherboard, and when you next upgrade your computer you won’t have to re-authorize the plugins.
List price for Pro-C 2 is $179 (160 EUR), but you won’t have to pay that. FabFilter offers discounts based on what other products you’ve previously purchased. Even if this is your first FabFilter purchase, you can still get 10% off by letting another FabFilter customer refer you. Don’t know anyone who’s already a FabFilter user? Post to your favorite online DAW- or music-production forum and somebody will surely volunteer to refer you for the discount, because they get a small store credit for doing so.