In this article Julian considers why he favours emulations of classic compressors for certain duties and if its about the limitations which give its character, why not try an alternative with it’s own set of characteristics rather than the flexibility of a modern all digital design?
Compression and compressors are one of those subjects which most people in audio spend a lot of time thinking about and discussing, probably more than is strictly necessary. A compressor is just a tool and in software it has been possible to create a ‘perfect’ compressor for a long time. Input exceeds the threshold. Audio is attenuated by the ratio at the speeds dictated by the attack and release. It’s easy to describe compression but its much more difficult to describe the subjective effect of compression in all its forms. Therein lies the itch we spend so much time scratching - the interesting stuff happens at the edges of what is ‘supposed’ to happen and just like with rigorously flat microphones or forensically clean playback systems, we often find that we miss that interesting stuff when it’s no longer there.
The Music In The Rough Edges
We need only to look at the popularity of saturation plugins and preamps which market their colour as their raison d’etre to see how many of us actively seek colour and character. The designers of classic professional audio equipment sought to make their equipment as clean as possible, the idea of deliberate colour as either a design goal or as an affectation is relatively recent. Only since we had demonstrably linear systems has the desire to deliberately regress become popular, and although it might sound counterintuitive, actually it this makes perfect sense. Colour as a choice is very different from colour as a consequence of the limitations of what is possible.
Back to compression. It’s common to hear of engineers explaining that they use X compressor with the needle barely moving, or indeed not moving at all. Look at the techniques of Michal Brauer, who in his ‘Brauerizing’ mixing method uses multiple compressors in parallel as flavours to be mixed and matched. His technique is unconventional and takes some getting used to, but central to his approach is the fact that the behaviours and tones of different compressors are distinctive and different. It’s based in technical performance but what these examples are about is ‘feel’.
An interesting experiment for me was to look at my own shifting tastes when it comes to compression. Many people find something which works and stay with it, and there is nothing wrong with that at all. If your equipment choices help make your mixes sound the way they do then great. It should however be remembered that it’s overwhelmingly the person, not the gear which makes the mix and change can be a good thing, it can inspire and keep us fresh.
I’ve always liked the fixed nature of the UAD LA2A Silver. It is what it is and it works. However I’ve found I’ve been using the Softube Tubetech CL1B in its place. The two are very similar and both are eminently suitable for the kinds of places I use this style of compression. The usual suspects of bass, vocals and piano are in evidence here. Things which are sustained rather than percussive is where I tend to use these. However just because they are similar in their approach to compression doesn’t mean they are then same. The CL1B has a manual mode for attack and release which the LA2A does not but the two sound different too.
To quantify these differences I ran both through some tests and the differences were evident. I’m not attempting to do a full breakdown of the performance of these plugins, and it should be noted that I’m using the original CL1B plugin, not the MkII version as that’s not AAX DSP. The action of the compression is very similar, with a slightly softer knee on the CL1B but the frequency response is very different, with a darker top end from the CL1B as opposed to the pronounced lift at the top and extreme bottom of the LA2A.
Further investigation might reveal further differences. For example in our recent podcast on Atmos for Music Steve Genewick shared his experience of using a (real) Fairchild in which he described not only the EQ curve the hardware imposed on audio just by virtue of being there but how the response gets darker with increasing levels of compression. There’s more to a compressor than just dynamic control!
Compressor Choice
We’ve all got more compressor plugins than we ‘need’. Exactly what motivates the specific choice we make when choosing a compressor is so often about what I see it as ‘good at’. This isn’t really fair as with care most compressors can do many jobs. The 1176 might be best known for aggression but it can be used in less shouty applications and to see something like the all buttons in technique as anything other than a party trick would be to miss the point. What it was designed for was as a flexible compressor which was faster than the competition at the time and offered adjustable attack and release - a novelty in 1968. It’s only since ultra-fast, infinitely adjustable compression has become common elsewhere that the idiosyncrasies of the 1176 have become the headline. While an 1176 will never be clean, it can be used without shouting its presence, but we all know the tasks which really suit it.
Something I’ve become particularly interested in recently is experimenting with alternatives to regular choices. Seeing what changes and whether the change is refreshing or a distraction. Some, like the LA2A/CL1B example have stuck around, others came and went but here are some substitutions I’ve tried where I identified a software emulation of hardware compressor which I use for a particular task and tried what I see as a natural, but contrasting, alternative. Using a well-featured utility plugin can emulate the style of different compressors, indeed this has been a feature of the presets in plugins like McDSP’s excellent CompressorBank for decades but it’s the unintended other characteristics of hardware emulations, like the difference in frequency response fo the LA2A/ CL1B example which makes this interesting and ‘hardware-like’.
To complement the example of contrasting optical designs in the CL1B/LA2A example above, here are four tasks and four first choice compressors, followed by alternatives. I’m referencing hardware here as a brand agnostic reference but I’m using plugin versions of all of thes. While particular plugins will vary, they all seek to emulate the original hardware. Importantly thought I’m not suggesting different emulations of the same hardware.
Bus Compressor
SSL G Bus/API 2500
For years I’ve been a fan of the SSL Bus compressor. And it is probably the best example of a design which earns its place because it does a particular thing really well. It took me a long time to figure out but on anything which is drum-heavy (which covers a lot of things) there’s something it does to the relationship between the snare and the rest of the track which sounds right. Hard to describe (especially if you’re deliberately avoiding use of the ‘G-word’) but just a little SSL does something to the back of snare hits and how the track sits which really works. I felt like I was missing something because I always used the same settings but talking to some very experienced engineers it seems I’m not the only one who spends the majority of their time on the same setting!
Curiosity and the availability of the UAD version natively encouraged me to try the API 2500, which is a far more complex beast, it sounds different but seems well suited to the same kind of material. There’s more here (so there are more ways to get it wrong!) but I particularly like the Thrust feature. It’s not as exciting as it sounds, its a carefully chosen EQ curve in the side chain which balances the sensitivity of the detector across the spectrum. An alternative to high pass filters in the side chain.
Drum Bus
Neve 33609/Chandler Zener Limiter
I never really ‘got’ diode bridge compressors until I tried a real 33609. They are flexible yet distinctive but the thing I really like them for is for use on drum busses where you can get some really pleasing, and very audible compression effects. They can spank if you want them to but in a different, rather more authoritative way than an 1176. This meaty compression put me in mind of a very worthy alternative which has a hardware heritage which is just as pleasing as the Neve.
The Chandler Zener Limiter is a very different beast, and is another piece of which I have hands on experience of the hardware. When I first tried the Softube emulation I really liked the weight of the compression effect. It can shout its presence with the best of them and is a great example of big bold compression.
Channel Compression
1176/Distressor
It still surprises me just how flexible the 1176 is. I’ve only ever come across the black LN version in hardware and it was the UAD Classic Limiter Collection which introduced me to the AE version. This most modern and flexible version has a 2:1 ratio option. We all know what the 1176 is famous for but as alluded to earlier, it was designed to be a flexible compressor and it’s surprising to what extent that still holds true, although you can still get delinquent on it it you like.
The natural alternative for me is the Distressor. I’ve had a little hands on time with hardware Distressors but I know their software counterparts well enough to know that, in spite of its name the Distressor is really flexible. There isn’t much you can throw at one it wouldn’t do an excellent job on. You can say that about many utility compressors but they wouldn’t do the full-on stuff which gave the Distressor its name nearly as well. Really worth getting to know properly.
Bus Compressor
Fairchild 670/Capitol Mastering Compressor
Such is the reputation of the Fairchild that, although I’ve never got on with it, I keep coming back to it. I find it illogical and inflexible (the same has been said of me…). But I do like the tonal and subtle dynamic ‘thing’ it can do on mixes. If only I could get on better with it. I used to use the UAD Manley Vari Mu for the kind of tasks a Fairchild might have seen use on with someone else driving it. I like this particularly on sparser music, particularly acoustic material.
The substitution I’ve made for both to these is the Capitol Mastering Compressor. I’ve spoken about this already on the blog and I’ll repeat here that I think it’s wonderful. Giving a kind of ‘everything sounds better effect even though there’s basically no gain reduction going on’ experience I associate with a well deployed Fairchild or Manley.
There are of course other examples I could give but the point of this article is to talk about the value of the ‘known entities’ that are classic hardware compressors and although plugin processing removes the constraints hardware places on a design, maintaining these and working within them in software can be really useful. Elements of the ‘feel’ of using a particular compressor survives the transition to software and choosing and using processors for what they are best at is a natural choice when you don’t have to drop thousands of dollars on the hardware.
Do you use hardware emulations over more flexible plugin alternatives and if you do, what guides your choices? Share you thoughts in the comments below.