For the first time, Weiss Engineering brings an exciting new feature to its EQ palette: dynamic control in the frequency domain.
You adjust the EQ to a nominal setting the input signal is then analyzed to determine exactly how much equalization need actually be applied. If the signal is already loud in a certain band, it is not boosted even more. Or, alternatively, if you want to attenuate certain frequencies, these are only attenuated when they exceed a certain level.
Multi-band Dynamic? Or dynamic multi-band?
The EQ1-DYN features four freely adjustable dynamic bands, and additionally three linear bands per channel. Setting up the dynamic bands is as easy as setting a linear band. There is just one additional parameter to be adjusted by the user, the threshold control. This makes multi-band dynamic control as easy and as versatile as adjusting a parametric equalizer.
Bands 1, 2, 5 and 6 are the dynamic bands (each individually switchable to linear) and bands 3, 4 and 7 are ordinary linear bands. The EQ1-DYN operates at 88.2 or 96 kHz, depending on input sampling frequency (88.2kHz processing with 44.1kHz and 88.2kHz inputs, 96kHz processing with 48kHz and 96kHz inputs).
A dynamic band operates similar to a compressor, with two additional features:
1. The band is sensitive to signal level in its frequency range only. So when applying dynamic bass EQ, only the bass-band of the programme affects the dynamic behaviour. There is no cross-modulation from other bands. Some compressors (like the Weiss 102 Compressor) also have this feature, called “sidechain EQ”. The exact centre frequency and bandwidth of the sidechain is dependent on the centre frequency and Q setting of the dynamic band.
2. The gain is only applied to certain frequencies, and not across the whole audio band.
3. Other modes (low-cut, high-cut, low-shelving, high-shelving) are linear (no automatic boost / cut adjustment), and behave exactly the same as in the EQ1-MK2.
In addition to the dynamic operation of the bands, the overall dynamic frequency response is displayed in real-time in the LC display. This is necessary for the proper setup of the parameters of the dynamic band, and it gives direct visual feedback on the operation of the EQ1-DYN, similar to gain reduction meters on compressors.
The reason for choosing exactly bands 1, 2, 5 and 6 is the following:
As the name implies, a dynamic band is sensitive to the level of the input signal. Sometimes it is desirable to first add some EQ, and afterwards dynamically add some more. Or vice versa, first do some dynamic corrections and aftwards add overall EQ, including in the corrected band. Because the EQ bands are connected in series, this is only possible if there are linear bands before and after the dynamic bands.
Why dynamic EQ?
We give two different applications for dynamic EQ, but the creative engineer is sure to find more!
Correcting musical instrument volume imbalances
EQing is the process of changing the spectral balance of any audio programme. By EQing, we wish to, for example, to attenuate or boost a certain instrument within a complex piece of music. Fig. 1 shows the spectral components of some instruments.
Fig. 1: Spectral components of different musical instruments – thin lines are harmonics only
It is clear from the graphic (and also from our hearing experiences) that most instruments overlap in their frequency ranges, so trying to affect one certain instrument by EQ without affecting others is usually impossible. This especially becomes a problem if two instruments in the same band are not in volume balance, i.e. one is too loud and masks the other.
One remedy for this is to apply either some boost or cut EQ in that band (either boost the soft instrument or attenuate the loud one). Except that this process affects both instruments. However, when using dynamic EQ, the boost is only applied if the loud instrument pauses. On the other hand, one can apply some dynamic attenuation whenever the loud instrument plays. The dynamic filter will adjust itself to correct volume imbalances between the two instruments.
“Ess”-sounds usually occur on, but are not limited to, recordings with human vocals. They are the product of over-compression (common on TV or radio recordings), or similar effects, like saturation of magnetic mediums.
These sounds are characterised by an exaggerated hiss or “s”-pronunciation (hence the name). They are mostly concentrated in the band between 1kHz – 10kHz.. Just applying attenuation by EQ in this band will render the recording dull and muffled. High frequency content is very important for hearing enjoyment, so simply attenuating will reduce programme quality. By applying dynamic cut EQ, the “ess”-sounds can be reduced, without affecting low-level high-frequency content, improving the quality.
The photgraph above shows a typical frequency response display with outlined nominal frequency response (thin line, as set by the user) and actual frequency response (solid black) applied by the DSP according to the program level and dynamics parameters. Note the threshold parameter on the right hand side.