Bel Canto and Digital Volume Control

Again, it must be pointed out that digital volume controls do not "drop bits". They use a digitally-controlled resistor ladder, which is like the "attenuator" control that some people prefer to a pot. There is a chip to do this job, and probably most manufacturers use this same chip.

Some of the digital controls do; the true digital ones that work in the digital domain exclusively. That is the type I intended to refer to. Wadia has a nice white paper with some explanation.

http://www.wadia.com/technology/technicalpapers/Digital_Volume_Control_2.pdf

Patrickp...Wadia puts the 16 bits into the MSB end of a 22 bit converter so that several shifts can be done before the LSB is lost. However, the resulting volume quantization is very coarse...(they say 18 dB, but I would have thought 6 dB). The quantization of the digital volume control of my McCormack preamp is 0.5 dB, and so is my Behringer DEQ 2496. I know for a fact that the McCormack uses the attenuator chip.

All -

I think I may have handle on the digital volume thing. Please feel free to provide additional insight and education as required.

My speaker amp:
90db/w/m Speaker rating + 61db amp gain (1200watt @ 4 ohms) = 151db SPL max
(I know that would kill both my speakers and my ears)

Assuming a low level average listening level of 60db?
(I used the average conversational level of 40db-80db as a point of reference)

Average CD recording level usually at -20 to -6 peaks
(It seems this recommended as a target for setting levels in the mixing studio)

Attenuation required for the average listening at low level is -70db?
(max possible gain less attenuation to get to the low level listening volume assuming the recoding level indicated for CD’s)

Id I understand it correctly 6 bits will provide 64 potential levels of attenuation. That would still leave be with 18 bits of resolution when using a 24bit digital volume control. That seems to be almost about the limit that may be realized with current state electronics. It’s likely all I can hear anyway.

That would seem to dictate about 1.25db of attenuation per step.

Thanks in advance all -

In regards to the attenuator IC -
Is there a pin compatible version that provides 1 - 1.5db / step instead of the .5/db commonly used? I don't mind losing a bit of 'fine tuning' in the remote in order to preserve another bit or so of resolution.

If that is not available I suppose I could to the sensible and easy solution and use a quality audio resistor or two into my XLR patch to provide the attenuation. - or heaven forbid I get a more reasonably powered amplifier. LOL