Digital Hardware Sonic Differences

I wanted to respond to a post in the Squeezebox thread, but on a tangential issue. Rather than hijacking that thread, I am starting a new one.

Quote:

---

Originally Posted by doodlebug

BTW, in comparison with the M-audio Audiophile USB my vote, sound-wise, is still with the M-Audio. Much better bass and a wider soundstage than the SB2. The SB2, however, is far more flexible in use around the house. I suspect it will go into the system downstairs for occasional use unless I can get the soundstage wider and with more bass authority.

---

The issue I wanted to comment on was the above noted difference in bass authority. I have no issue with the stated preference since I have never heard the M-Audio device to have any opinion or preference of my own. I'm just looking for advice on a mechanism by which these two digital devices could have such different bass response. We're obviously narrowing it down to the internal DAC's since everything upstream and downstream of the DAC's would be identical. In the digital to analog coversion, I have been led to understand that bass frequencies should be the most immune to sonic differences of the entire frequency spectrum. The wavelengths are so long that the constant sampling rate results in much more data to store the LF information as compared to the HF information. This is why MP3's are generelly held to have decent bass fidelity but are lacking in mid and HF fidelity. A loss of data (due to compression) is much more critical in the HF spectrum since there is less data to work with.

I was under the impression that the known mechanisms for loss of accuracy in the digital to analog conversion are pretty much isolated to timing errors (jitter) and limitations in certain filtering technologies. However, the resulting impact of both on LF reproduction is supposed to be next to nothing. Not to mention that current DAC technology has pretty much eliminated jitter and filter distortion to vanishingly low levels.

Getting back to the subject at hand, I would think that differences in bass authority would have to involve frequency response issues (certain frequncies being accentuated in one device versus the other). I just don't know of any mechanism that can account for this. Even the cheapest digital devices can pretty much reproduce a LF analog signal that is a virtual copy of the original, with differences that are demonstrably well below the threshold of human hearing. Just what am I missing?

Don, I am sure Ian and others will chime in here, but the differences could possibly be simply due to inadequately sized coupling capacitors resulting in a cut-off frequencies (RC time constants) that are too high. This may be a design defeciency or an attempt to limit LF noise passing through post I/V conversion.

Or to delve into the realm of subjective audio, it could possibly be traced to:

-The power supply may be poorly regulated and not a low impedance design.
-The amplification stages of the analogue side may not be high performance in terms of slew rate, etc. Yes, I know it doesn't make sense that you'd need a "fast" amplifier to reproduce 20 Hz, but when capturing the total acoustic picture of the event, it can and does in my experience.

My 2 cents worth.