I've seen you carry on about USB implementations, but you never discuss the basics. It's reminiscent of a sort of fetishization - while I'm sure for some people it's fun to be covered in shaving cream and then get kicked in the testicles by a nun in high heels, it's not exactly the correct way to go about procreation.

So to go back to the basics, is the data getting to the DAC unaltered and relatively on-time? The robustness and fault tolerance of the transfer mechanism are important to designers and make for some interesting academic discussion, however after a certain point those concerns are moot.

In terms of jitter I prefer to think of that point as audibility. I go by studies on audibility thresholds in live human beings as the raw, sexy, productive arbiter of what really matters.

As Analixus just attested to, the proof is in the pudding. Does a so called "modern" DAC wipe out all incoming jitter to the point that the source is rendered irrelevant? No. That's the end of it. I don't need to see theoretical studies about the supposed audibility point of jitter. You have ears, you can hear it. Or you can believe that 1s and 0s are 1s and 0s, bit-perfect is bit-perfect, anything around 1ns or so of jitter is perfectly acceptable, and promptly ignore any real world, subjective evidence. That's your call.

I wish there were super hero DACs that you could connect something like a $50 Pioneer DVD player to and they would sound just as good as if you had connected a $50K Burmester belt driven transport. I would be happy as a pig in shit if that were the case, it would certainly save me a lot of money. It's not the case.

The answer when someone asks "why does my CD player sound better than USB through my laptop" is not to say "I don't understand why you would hear that, it's impossible because DACs don't care about incoming jitter." Is it adaptive or asynch? Is the USB receiver self or bus powered? Is the laptop running on battery or AC? All of these things make a difference. If DACs "didn't care" as long as the output of the laptop is bit-perfect, which it would be, and "relatively on-time" which it would be, depending on your definition of the word relative, then none of it should matter. The laptop would sound the same as the CD player, which would sound the same as a $50K Burmester.

The way to address it would be to examine the implementation of individual components in the specific comparison. It'll help identify the culprit, rather than throwing around nebulous assertions. The loaded nature of the initial question and the broad strokes painted is nothing but flame bait. As Dave has just pointed out, there may be specific configurations when one method may overtake the other.


@Dave - What do you know of the M51's conversion of incoming PCM to PWM? This signal is also resampled to a high rate to apparently negate jitter from the source. Wouldn't these converters, championed by so many "audiophiles", hinder the process that the M51 undertakes? Most reviews I've read (not that there's much stock in them) echo that the usb implementation is very good, surpassed only the hdmi. These findings are also backed by measurements. If there's a relatively level field across the M51's inputs, wouldn't perceived differences be predominantly due to the source component and how well it transmits that signal down a particular path (usb/coax/optical/aes/hdmi)?


Fwiw, I'm quite content with the M51's performance over usb (from my MBP).    I'm skeptical of how much improvement these converters provide. Someone on another forum recently found the AP2 degraded the sound compared to usb (surprised he wasn't lynched). I'm curious to hear one in my system, just not at the asking price.

As Analixus just attested to, the proof is in the pudding. Does a so called "modern" DAC wipe out all incoming jitter to the point that the source is rendered irrelevant? No. That's the end of it. I don't need to see theoretical studies about the supposed audibility point of jitter.

I'm not talking abou theoretical studies about the supposed audibility point of jitter (and in fact at least one such study contends that sub 100ps jitter may be audible in theoretical models of jitter at specific frequencies), there are studies that take a statistically significant sample of listeners and push jitter to the point of audibility, even to grossly offensive distortion, and then ramp it back. One of the first such studies dates back to the early 70s which placed the threshold for 95% of the population at 35ns. There's not one case, ever, of anyone identifying jitter (in a controlled test) at, say, 1ns.

You have ears, you can hear it. Or you can believe that 1s and 0s are 1s and 0s, bit-perfect is bit-perfect, anything around 1ns or so of jitter is perfectly acceptable, and promptly ignore any real world, subjective evidence. That's your call.

I have ears, I've never heard it. I don't doubt that you might have, I only disagree on the reason why. (My thinking is that this has more to do with the differences between our brains rather than our ears or our gear) And I don't ignore real world, subjective evidence, I simply have a more rigorous standard for evaluating the merit of such evidence. There's no good reason why, I'm just a hobbyist who likes audio and audio gear. However having had a few bits of experience in research and peer review I acquired a very low tolerance for sloppy research or pseudoscience by professionals, including the sort who make or market audio gear. This is a form of snobbery that I am comfortable with, though I respect that many are not.

To clarify, I have no problem with talking about subjective experiences with non-transducer gear and I am also a snob for usability and performance. Nor do I have a problem with anyone else's beliefs or fancies. Nonetheless I will tend to speak up if a third party asks for a recommendation and receive what I consider to be the wrong information drawn from an aside about barely relevant engineering.

Dave, you clearly have a lot of experience and have spent time researching aspects of digital audio. What I wonder though, is the foundation of your understanding of the field. Some years back when I delved deeper into audiophilia I decided to consolidate and solidify my understanding of digital audio by reading Watkinson's "Art of Digital Audio" and Pohlman's "Principles of Digital Audio". Have you read any similar texts? If not I would be happy to send you my copies of the aforementioned for your leisure.

Of course more important than keeping abreast of any one field of knowledge is knowing how to contrast it against others, including the methodologies. The next field I'd like to delve deeper into is electrical engineering. I have a familiarity with the basics, including signal propagation. I don't dispute that there are differences between communication protocols and I'm all for studying them and uncovering faults. There are plusses and minuses to each implementation, be it coax or optical or USB, though any of t hem will work with proper implementation. I'd rather not trudge through the intricacies of each one as I think that the locus of the disagreement is elsewhere.

Speaking of Empirical, here's what Steve says on his website regarding CD transports vs. his USB implementation:
 
Overdrive® SE USB DAC/Pre
USB digital to analog converter, Pre and Head amp
The Overdrive DAC enables you to play music from your computer delivering unparalleled sound quality to your preamp or better yet directly to your amp. All you need is the Overdrive and a PC or Mac. The Overdrive will also improve the playback from your CD transport. Once you start using the computer, however you will never go back to spinning CDs. Playback using the computer will be superior to the best CD transport you have heard, even playing low-res CD rips.