Get a limiter in there….fast!!!

Hi people

Lets set the scene. 8.30 am, feet up on the sofa with laptop, recorded episode of top gear on the tv. Patio door open, with the most fantastic day outside. Sun, cool breeze, the sound of bird song.

Now lets spoil that and talk tech!

Well, I’m finally going to shed some light on some of the more trendy topics. Compressors and limiters are everyone’s favourite toys, but their role in the whole sound making process can be misunderstood. I’m going to attempt to clarify some things, and hopefully help you start getting  the best from your dynamics tools.

“Why the title Tone?” . If you’ve read any of my articles or heard anything I’ve worked on, then you’ll hopefully realise that for me the sound comes first. So why am I suggesting you get a limiter in fast? . 3 major reasons.

1) You will be less inclined to make mistakes.

2) Humour the limiter

3) Protect your ears

Here’s more info on that

Most material we work on contains frequency spikes. Especially if it’s been through a compressor. There aren’t many occasions where you will be using 0 ms attack times, most models don’t even go that low, because it’s pretty pointless. So those 5ms attacks for example need dealing with. “How is the limiter going to help prevent me making mistakes?”.

Lets say you’ve done your compression, you’re now starting on the eq for the track. The material was a bit spikey, and the attack allowed through the compressor has added to that. Now what we’re looking at, are freq overshoots/imbalances hitting your ears at possibly a fair bit louder than the main body of the sound. The worst thing can be, that these spikes are so short in duration that you may not even notice, but your ears will !. At the early stage you don’t want to be hearing the limiter working at all. It’s there to stop any nasties hitting your ears before you eq the track.

What will it affect? . Virtually every decision you make!. Without something catching the spikes, Your perception will go dull in the freq areas being hit. Ever gone back to a mix the following day to find a nasty imbalance you hadn’t noticed?. If you have a fantastically acoustic treated room, then all this becomes less of an issue, but few are so fortunate, and if you have that kind of studio it’s unlikely you’ll be reading my articles .

Humouring the limiter

Even if you’re not making an entry to the loudness wars, you still have to humour the limiter a little. Doing all the eq work etc  and then expecting to be able to just put one inline wont give you the desired result. I used to leave it until the very end, so it didn’t feel like I was favouring volume/loudness over sound, but I found I always had to go back on the eq to please the limiter.

Compressor vs Limiter

So what’s the difference between the 2?. Have a look at this pic, it very roughly shows the body of the sound, where the compressor does it’s thing, and the transients or attacks that the limiter deals with.

This is a very basic example, so no super technical comments on the intricacies of the subject…PLEASE!

I’ve used a particularly bad example of attack overshoots to make it clear, but imagine those spikes hitting your ears, that far above the gain of the body of the sound.  It can force you to make bad decisions, like going back over the compression and trying to fix it with that. Attempting to remove the imbalance with eq etc etc.

Preferably, these issues should be dealt with at mixing stage. If left until mastering, the compromise in sound has already been made. Multiband compressors and limiters can help clean it up, but there is a trade off . Yes…multiband limiters have a use. Forget having one as a final brickwall in the chain, they sound unnatural to me, but in the right place they can be useful.

What model limiter?

I’m not about to start recommending various makes etc, that’s personal preference and a part of the character you put into your  mixing/mastering. I would suggest looking for models that offer options to help transparency rather than just “it stops stuff going through” .

Release times are more important than you think!

“But what about auto release Tone!” . I haven’t actually found a model where  I trust it. In fact, not many even offer the  option. I’m not in the habit of letting a computer decide what sounds good. Humans will be listening to my output, and most of those have ears .

Gain reduction defines release times. If the limiter is taking 2db off the peaks, it will need a slower release than if it’s removing 6db (god forbid). If you want the best from your limiting, automate the release times for different sections. I actually write them in real time as the track plays through. The final job has to please peoples ears, if it sounds good to me, then we’re on the right track. You will find the release will work at a few different times, and not at all at others. I never seem to use fast times, they sound fatiguing and unnatural.

So, protect your ears, avoid time eating mistakes, and Get a limiter in there…. fast!  . Trust me.

Thanks for reading

Tone