The question “What Do You Mean When You Say Limiter?” presented itself during a team chat about an upcoming tutorial using a fast, high threshold, high ratio compressor to catch the peaks of a vocal and a slower, low threshold, low ratio compressor to even out the levels through the performance. This is a common technique and has been addressed by Dan in his double-bagging compression tutorial in which a common combination of an LA2a and an 1176 are used in these roles.
During these conversations, which initially were about whether there is a consensus about the correct order in which to place these two processes, the word “limiter” started to be used to describe the fast compressor with the high ratio. This is correct as a fast compressor behaving in this way would typically be used and would often be described as a limiter. Indeed the 1176 which is so often used in this role is a “limiting amplifier” it doesn’t say compressor on it anywhere.
There is clearly a degree of overlap between limiters and compressors and when I asked the team what they thought of as a limiter I got revealingly different answers. A caveat here is that everyone understood the formal definition of a limiter but there clearly seems to have been some divergence on what people mean when they say “limiter”…
I maybe old school, but back in the day when it was all analog, we considered a compressor with a compression ratio of at least 20:1 to be a limiter. Limiters tended to be protecting analog devices like cutting lathes or broadcast transmitters. For example, the transmission limiter in the MCR (Master Control Room) at Piccadilly Radio here in Manchester, England, back in the mid 80s in the days before Optimod, was a Neve 2254, set up as a limiter. Concepts like feed-forward limiters and true peak limiters were yet to come and even when they did, we would have never used a feed-forward limiter because the delay required to make them work in an analog world would have meant that presenters couldn’t monitor off-air which, was the way we chose to work, so we always knew that we were on air, and what we were listening to, was what the listener was hearing too.
For me a limiter has been a protection device rather than a creative device, but that I acknowledge is down to my working life largely being in a broadcast environment working to a maximum peak level and PPMs.
I’m not going to say the textbook definition is wrong, it isn’t. However when I set up a limiter I typically mean something rather different.
When I instantiate a limiter plug-in it will invariably have a ratio of infinity to 1. If I was using an outboard limiter the old definition would apply but I haven’t used an outboard limiter in years. Now when I use a limiter it will be a brick wall limiter, it will probably be being used in look ahead mode, have an attack time of zero and will effectively be a two control process: Threshold and release.
I find I don’t use high ratio compression much any more, gentle compression tends to be between 1.5:1 and 2:1. Stronger compression tends to be between 4:1 and 6:1, after that I’m looking for infinity to one as I just want to catch the peaks.
I wouldn’t suggest I’m right and anyone who does differently is wrong. I’m probably missing out of something working that way and importantly this isn’t the result of a strategy, it’s just honestly reflecting on what I actually do most of the time. So does it make sense and is there anything which can account for the difference between my thinking about limiters and Mike’s?
By “Limiter” Do We Mean The Fastest Available Device?
The thing which is common to both an 1176 at 20:1 and a look ahead brick wall limiter is that they both have the same intention - They are different iterations of the fastest, hardest compression available, they were both, in their time, the “best tool for the job”.
When it was designed in 1967, compared to its contemporaries the 1176 was very, very fast. With a minimum attack time of 20 microseconds much faster than the previous generation of hardware, for example Fairchilds offered a minimum of 0.2 milliseconds, an order of magnitude slower. When people reach for a limiter, I suggest that they want the fastest tool available. I think our perception of a limiter has changed with the technology. The distinction, which remains, is between lookahead and genuinely real time limiting because look ahead is of course impossible to do genuinely live.
Do you use analogue limiting? Do you use non-brickwall limiters? Share your thoughts in the comments below?