I thought that I’d share my findings with the seemingly cryptic parametric EQ values that A&H uses. The user manual declares the bandwidth value to be in octaves, but it’s not the same standard that many other systems use. The following table converts the A&H parametric EQ bandwidth (BW) value to the more common Q and BW values according to https://www.w3.org/TR/audio-eq-cookbook/. (approximate conversion)
Parametric EQ Translation:
A&H BW
Std Q
Std BW (octaves)
1.5
0.367
3.22
1.4
0.393
3.06
1.3
0.423
2.90
1.2
0.458
2.72
1.1
0.5
2.54
1
0.55
2.35
0.95
0.579
2.26
0.9
0.611
2.15
0.85
0.647
2.05
0.8
0.688
1.95
0.75
0.733
1.84
0.7
0.786
1.73
2/3
0.825
1.66
0.6
0.917
1.50
0.55
1
1.39
0.5
1.1
1.27
0.45
1.22
1.15
0.4
1.38
1.03
1/3
1.65
0.86
0.3
1.83
0.78
1/4
2.2
0.65
0.2
2.75
0.52
1/6
3.3
0.44
0.13
4.23
0.34
1/9
4.95
0.29
Shelving EQ Translation:
Shelving Type
A&H Freq
Center Freq
Std Q
Std BW
Low
f
f * 2.5
0.471
2.67
High
f
f / 2.5
0.471
2.67
Note that the shelf slope Q is a fixed value and not variable. Many of us wish that A&H will someday add that missing option.
UPDATE: Unfortunately, I can’t edit my original post above after realizing that the dLive PEQ uses a dynamic Q which depends on the magnitude of the boost or cut. That table assumed a boost/cut of about 10.6 dB.
My revised estimation of the relationship is as follows:
Peaking:
Q = (0.0008 * dB * dB + 0.0004 * |dB| + 0.45) / BW
Low shelf:
Center frequency = f * (|dB| * -0.07 + 3.34)
Q = |dB| * -0.0045 + 0.5254
High shelf:
Center frequency = f / (|dB| * -0.07 + 3.34)
Q = |dB| * -0.0045 + 0.5254
where:
'BW’ is the A&H bandwidth value on the console EQ band.
‘f’ is the A&H frequency selected on the console EQ band.
‘dB’ is the magnitude of the boost or cut (in dB).
Therefore, choosing a value that is independent of the gain and thus constant is actually more correct than the often-used Q.
Mixing Station, by the way, refers to approximately half of the possible gain, namely 8dB, at which the Q values are correct.
From a user perspective, yes, it is easier to grasp. However, knowing the actual Q used by the console can be useful. For example, I’ve written an app that can calculate the sum of multiple cascaded PEQ’s and predict and compare the results offline. This is helpful for when I’m using a channel PEQ which then routes to downstream PEQ’s found in groups and buses. With this, I can confidentially modify the collection of PEQ’s offline to be more efficient after inevitably playing whack-a-mole in live situations with feedback sensitive speaking mics.
