Now, with the loudness war in full effect and all, we get why. But the danger lies in pushing your limiter too far. As a creative tool, a limiter can boost the perceived volume of a certain sound. It can also be used to restrain unpredictable effects or automated sounds that have too much dynamic range.
In mastering, the use of a Limiter is more straightforward. If you set the threshold just right, the limiter will cut those peaks without any audible side effects, which is called Transparent Limiting. As you know, compressing — or limiting — turns down the loudest parts of an audio signal. This effectively leads to a smaller dynamic range, the difference in volume within the audio signal. The peak volume started out at 20 dB and has now been decreased to 18 dB.
This is where the Make-up Gain or Output Level setting comes in. The most basic explanation of a limiter would be that it's a device that stops all audio whether digital or an electrical signal from exceeding a certain volume. Though it can operate at any volume level, it's mainly used at the maximum allowed amplitude to avoid the phenomenon of peaking.
Before we dive in much further, let me lay out the game plan here. We're going to define what an audio limiter is, talk about it's brother the compressor, and then explain when you would want to apply limiting and how to go about doing it. When we're done, you'll know everything you need to know plus a little bit more, all without too much technical jargon and complicated language. Let's dive right in. A limiter is a tool for signal processing like mixing music that applies a type of dynamic range compression.
That means that it can take an input signal, evaluate its amplitude volume , and attenuate lower the peaks of the waveform if those peaks reach and exceed a threshold value. In other words, if we set the limiter's threshold to -5 dB, it will not allow any audio to be louder than -5 dB. The audio may seem to grow louder as it slams into the threshold and the average volume of the audio increases, but the maximum amplitude is blocked from climbing any higher. Typically, we set the threshold to 0 dBFS decibels Full Scale, the digital scale when mixing in the box.
Both of these points are the equivalent maximum allowed amplitude and anything above it is considered 'peaking. Unintentional peaking is referred to as audio clipping , which at best leads to distortion and ruining recordings or broadcasts and at worst leads to heat damage and blown speakers. We'll explain clipping more below, because a limiter actually applies clipping purposefully but tastefully.
You find these tools in two forms: hardware and software. The hardware units deal with actual electricity. You'll see these in rack mount units in music studios or radio stations. The other type are VST plugins for digital audio workstations that act upon digital signals. A compressor is the main tool for applying dynamic range compression.
It has a threshold value that is set at a certain decibel level and any audio that exceeds that level is then reduced in volume by a ratio you are allowed to control. So if the threshold is set to dB and the ratio is set to three-to-one , then for every 3 dB that go over dB, only 1 dB actually pops out of the other side. An example would be the amplitude reaching -1 dB, which is 9 dB over dB. The compressor would reduce the peak at -1 dB down to -7 dB 9 dB went in and only 3 dB came out, due to the ratio.
That can get complicated. You can read our article What is a Compressor? Scientists have produced several loudness models based on the fact that our hearing system processes each frequency in a different way. Another problem arises because our hearing system is just so very flexible and adjustable. If you are in a silent cave, you can literally hear your heart beating, but if you are next to the stage at a rock concert, your ears quickly lower the resolution, to protect themselves from being damaged at least for a short time , so you will not hear your heart anymore.
This allows us to focus on what is important, and is directly linked to our primeval survival mechanisms which were much needed when we were living in the wild. The basic way to measure loudness is called the 'root mean square' or RMS , which averages energy over a given time interval. It has several weaknesses, but the most important factor is that random peaks don't affect the loudness as much as peak meters do.
When combined with a filter approximating the spectral properties of human hearing, RMS becomes a fairly reliable solution. Actually, we don't really need loudness meters, as our ears can do the job. It is however important to be able to control loudness, and interestingly we run into several issues here. Samples are represented as binary numbers of some predefined size. In most cases we use bit floating point numbers. These consist of 24 bits containing actual digits, and 8 bits of so-called exponent, which only multiplies the number by some exponential.
So it essentially defines the resolution in a similar way to our own hearing system. An interesting consequence is the range that these numbers can represent. However the accuracy is questionable. These numbers only correspond to about 7 decimal digits, just like 24 bit numbers. Therefore it is okay to take your mix and apply dB gain without any loss of quality except for some rounding errors , but it also doesn't mean that if you generate a mix at 0dB it would be of any higher quality than a mix rendered at dB, although it wouldn't be very convenient.
As for most reproduction hardware, dB is almost complete silence. Because of this, most high-level audio processors use bit precision internally and then convert the output back to your DAW's bit floating point numbers with only the round-off error.
The real problems occur with the final results themselves. For example, samples used on CDs have only bits resolution, about dB. In the analogue world radio, vinyl etc.
For vinyl it is actually much worse than dB. Since none of these media provide a floating point range, you as a mastering engineer have to carefully use all of the bit-space provided. On CDs the range is from dB to about Therefore your masters should always reach this point of So you use compressors to decrease the dynamic range, and then you increase the level so that your song reaches 0dB.
However, at this stage the loudness will still be much lower than the loudness of songs on professionally mastered albums which is where limiters come in. Though compression is used on nearly every instance of recorded live audio, there still seems too often be some confusion as to what exactly compression is, and when and how you should use it. Limiters are typically even more misunderstood, so today we wanted to take a look at exactly what a limiter is, and when they come in handy for your recordings.
It could be said that a limiter is an extreme version of a compressor. This is done for a number of reasons, and is primarily responsible for evening out the volume of a track.
0コメント