Audio Recording

In Waves linear phase multiband; you will notice that it’s not using compression ratio but instead it’s using “range”. According to their manual, range is related to compression ratio and gain reduction but the difference is not thoroughly explained. For those that are confused by the manual and looking for more details, you can find answer in this post. This is how the linear phase multiband looks like:

To better understand how to use this tool particularly when adjusting range and threshold, you need to know some multiband compression fundamentals. Audio compressor reduces or controls the dynamic range of the signal. It will work by compressing loud peaks above a certain threshold. In a multiband compressor, you can assign different compression settings for 5 different frequency bands. This is very useful because in most audio mastering work, you cannot apply the same compression settings for the entire frequency band. The primary reason is that each frequency band behaves very differently from other bands. For example, in some track the bass is too powerful than the mid and the high. If you are applying only one compression setting regardless of frequency bands; it will also affect the mid and the high. It might solve the bass compression issue but it will introduce problems to mid and high frequencies. This is where you should be using a multiband compressor like Waves Linear phase multiband because each band has its own compression setting. When compressing you need to set three things:

a.) Threshold
b.) Compression ratio
c.) Frequency band where dynamic range manipulation will be applied

a) and c.) are easy to adjust in Waves Linear phase multiband because it’s clearly indicated. The only problem is the compression ratio which is not used but instead it is using “range”. Supposing you have a compression ratio of 3:1; this means that if the input level is 3 dB over the threshold value, the output level will be 1 dB over the threshold setting. Let’s have an example. Supposing you need to apply this compression ratio and set at -45dB threshold; then it will start compressing signals louder than -45 dB (for example, -40dB, -36dB, -11dB, etc).

Problem1: Compute the output of the compressor when the input level is -42dB.
Solution: Compute the difference of the input level with respect to the threshold:

Difference= Input signal – Threshold
Difference = -42dB – (-45dB)
Difference = +3dB

Since the compressor is using 3:1 ratio and the input is 3dB over the threshold, then the output would be:

Output= 1dB + Threshold
Output= 1dB + (-45dB)
Output = -44dB

The rest of the input-output relationship is depicted here graphically:

The x-axis is the input while the y-axis is the output of the compressor. How does this example relate to “Range”? According to linear phase multiband manual, they provide this other example:

If the compression ratio is 1.5:1 and the threshold is -35dB. Then the equivalent LINMB setting would have Range set to “about” -9dB

The question is why the range is set to “about” -9dB? First, it’s negative because you are compressing peaks. Loud peaks above threshold will be attenuated or compressed. The question of why the value is 9dB is confusing. Bear in mind that the manual says “about”; so this implies that it is probably an “estimate” rather than an “exact” value. To better understand this, let’s plot the response of a compressor at 1.5:1 compression ratio at -35dB threshold. See below:

So it shows that -11.7dB is the maximum gain reduction of the compressor. But according to LinMB compressor manual, this example will have a range set at -9dB. It is because “range” is defined as:

LinMB Range <= Compressor Maximum Gain reduction.

The range should be less than or equal to the maximum gain reduction of the compressor given a specific compression ratio. Therefore range and compression ratio are related by the compressor maximum gain reduction. If you know this value, you can set the range easily. So how can you compute the maximum gain reduction without needing these graphs? Below is the formula:

Maximum gain reduction of the compressor = Threshold – (Threshold/CompressionRatio)

Problem2: How much range would you apply to a LinMB compressor if you like to apply a threshold of -25dB and a compression ratio of 2:1?

Solution: Substituting the values:

Maximum gain reduction of the compressor = (-25dB) – (-25dB/2) = -12.5dB

Thus the maximum gain reduction for a 2:1 compression ratio is -12.5dB, your range should be less than or equal to this value. In most instances, LinMB compressor defaults to -6dB range why? It is because the input signal to the compressor during mastering stage is still LOW, so you won’t be applying much gain reduction during multiband compression stage.