Implications of Esoteric G-0Rb atomic clock

The latest TAS (March 2008) has an excellent piece by Robert Harley: a review of the Esoteric G-0Rb Master Clock Generator, with sidebars on the history and significance of jitter. This Esoteric unit employs an atomic clock (using rubidium) to take timing precision to a new level, at least for consumer gear. It's a good read, I recommend it.

If I am reading all of this correctly, I reach the following conclusions:

(1) Jitter is more important sonically than we might have thought

(2) Better jitter reduction at the A-D side of things will yield significant benefits, which means we can look forward to another of round remasters (of analog tapes) once atomic clock solutions make it into mastering labs

(3) All of the Superclocks, claims of vanishingly low jitter, reclocking DACs -- all of this stuff that's out there now, while probably heading in the right direction, still falls fall short of what's possible and needed if we are to get the best out of digital and fully realize its promise.

(4) We can expect to see atomic clocks in our future DACs and CDPs. Really?

Am I drawing the right conclusions?
Ag insider logo xs@2xdrubin
So, now we are finally approaching the pure "bits-are-bits" theory with the USB based digital music reproduction (as was intended from the beginning) and we're leaving the vagueness of the bits-are-bits + something magical concept. This could be a manifesto. To all audiophiles worldwide: Get rid of those mediocre CD transports!

AFAIK any digital signal that requires conversion to analog (which is the format of the sound that we hear) requires a clock. So even USB based DACs will require a clock for the data (asynchronous or otherwise) to be converted to analog. Though not exactly the same, it's like tuning into your FM radio station.. If you're not tuned into the right frequency, you won't get the full sound. If your recording runs at 44.1khz, you need a clock that runs at 44.1khz.. Can't get away from that clock.

Bottom line: bits can be bits until the point where it's converted to analog, at which point you would need the most accurately constant (i.e. lowest jitter) clock possible to ensure the sound's integrity with the lowest jitter. One solution?...the atomic clock.
Kamil is correct about the need of some sort of "clock" . The point I've been making with my posts is that the synchronization of TWO clocking signals, between a "traditional" type of transport and a DAC, is completely eliminated with USB connectivity (provided it is designed correctly). With USB, we are down to just ONE clock at the DAC. With a hardrive having the ability to stream perfect music data and NOT having to carry an overlying clocking signal on the USB cable, the improvement is substantial. The sonic benefits are HUGE. Most importantly, the need for a "bank account breaking" UBER clock like the Esoteric is eliminated altogether. The sonic potential of the USB approach is way cheaper (no need for an UBER clock) and yields a superior sonic result to boot. This is the future of the best digital sound.
Ehider ... you don't need USB, you just need a buffering and reclocking DAC, like the Lavry 924. A relatively small RAM buffer will allow for any drift between the transport clock and the DAC master clock.

Of course, the more stable the DAC master clock the better. This may be where a rubidium type clock is useful .... I don't know.

I'm not sure where current technology is with respect to the relative influence of clock imperfections versus DAC non-linearity (both are probably an order of magnitude better than current speaker technology).

One last thought ... it's true that ADC clocks are equally important, and therefore a better ADC clock will lead to a better remaster. However, the implication is that all early digital recordings (ADAT) are forever unrecoverably distorted by the relatively poor clocks used at the time.
Sean, is it possible once jitter is introduced into a specific recording to remaster this recording in such a way that the jitter could be eliminated? If this is not possible, why not? Are those early digital recordings sounding bad because of the high jitter content? There are also oldskool digital recordings (from 1980/1981/1982) that sound great, how could you explain that? Pure luck?