DACs

Benchmark DAC2 HGC

ARTICLE INDEX

Design and Set up of the Benchmark DAC2 HGC

It has a plethora of both digital and analog inputs as well as one balanced and two unbalanced outputs that can be programmed to be amplified or line level. The 'HGC' in the name stands for 'Hybrid Gain Control' and refers to the different types of gain employed depending on the type of input signal. The types of gain are active analog (this is the standard sort of gain for a preamp), passive low-impedance attenuators, and a 32-bit digital gain control. The analog control is used for analog sources, digital for digital, while the passive attenuation refers to the ability to adjust the output level on the balanced (XLR) outputs. The manual talks about the importance of using this feature to match the DAC2 to your particular amplifier when the DAC2 is being employed as a preamp. But in discussions with Benchmark's Chief Engineer, John Siau, it came out that the adjustment is recommended even if you are connecting the DAC2 to another preamp. In that case you would try the different settings on the internal jumpers - 0, -10 or -20db (-10 is the default setting), selecting the one that gives the system the loudest volume setting you think will employ with the external preamp's volume control set to near its max. If you are not using an external preamp the process is similar but calibrated to the DAC2's own volume control.

That front panel volume control is used for both the digital and analog volume control. In the case of analog it's a servo-driven gain circuit control built around a custom-made Alps potentiometer. That in itself is already music to an audiophile's ears. The digital volume control is used for digital sources and while this might sound like 'not-music' to an audiophile's ears we should note that digital volume control has come a long way. While it's true that it still means a loss of resolution as compared to analog volume control, new techniques mean that the amount lost is 'down in the noise' and has to be compared against the noise in the analog option. Basically, the method is to pad (fill out) the 32 bit data word with 0's and use mathematics – i.e., multiplication, to adjust the volume rather than simply throwing away bits from the original word. If you're into such things it's explained for real in this paper by ESS:

http://www.esstech.com/PDF/digital-vs-analog-volume-control.pdf

Benchmark isn't the first or only company using this technique of course, but to my knowledge they are the only company building headroom into their digital processing. If you understood all of the above about digital vs. analog volume control then you know that a premise of digitally sampled music is that the loudest possible signal uses all of the available bits and there is no more, if the actual volume of the input signal is higher than the digital system has been calibrated for, you have noise of the worst kind. Commercially released CD's should not have this issue (though some probably do). Benchmark has discovered though that even if a CD is recorded properly, the interpolation process used as part of the processing can still produce 'samples' that are above what was meant to be the 0db level. By 'leaving some room' – headroom that is, in the digital processing chain the DAC2 avoids the artifacts that come from processing these samples.

There are a multitude of internal adjustments possible with the DAC2, in addition to the aforementioned output level, the gain of the headphone amp can be adjusted and the muting behavior of the outputs when headphones are plugged in. I stayed with the default setting in all cases.

Like the DAC1, the DAC2 has a utilitarian front panel array of LED's to indicate which input is active and the bit rate detected on the input. The remote is a delight, slim yet substantial all-metal construction.

Since I received my review sample of the DAC2 Benchmark has come out with two reduced-feature models. Both have the same internals and performance as the HGC but remove an item or two to get the cost down (a little). The DAC2 L is all the same as the DAC2 HGC but without the headphone amplifier. The DAC2 D keeps the headphone amp but loses the analog inputs and the 12V trigger. Each of these new issues is $1795 or two hundred less than the HGC.

Also in the design realm I think – the user's manual. Benchmark is in the habit of putting out some of the best manuals in the business. Besides exceedingly clear instructions (with one notable exception in this case), they contain a nice smattering of design theory explaining what you've paid for – engineering expertise. The exception in this case is an incomplete instruction in regards to the driver installation. The manual says: "run the setup.exe from the zip file. Do not decompress the zip file and run the setup.exe". I'm not sure how you run anything without decompressing the zip but the difference seems to be in whether you unzip to some location on your computer other than where the zip file landed when it was downloaded.