Originally Posted by Jason Stoddard
With respect to I2S, I'll leave Mike to comment more if he'd like, but I believe it comes down to this:
1. We make (by and large) inexpensive products that are geared at the vast majority of applications, rather than specific, ultra-high-end implementations.
2. The opportunity in lower cost products is much, much greater than any high-price-tag stuff.
3. Because of this, it's more likely you'll see us working on connectivity such as WiFi or Bluetooth (yeah, I know, barf) before I2S, because this is more congruent with where we are in the market, and the engineering resources are much better used in this space. Note that this isn't a promise of future products; this is merely an example of other mass-market connectivity which we currently do not support.
4. Similarly, we won't be working on things like $2500 USB cables or reintroducing ST optical connections for our products, for the same reasons.
5. We don't ascribe any overarching advantage to I2S for connectivity, because every connector/format has its challenges in implementation--especially when working with a format that was never designed to run down a cable (I2S stands for Inter IC Sound).
6. If we're talking I2S via HDMI, HDMI has its own set of headaches (licensing, special test equipment, DRM, ongoing revision levels) that we'd prefer not to deal with. The best way to deal with HDMI is with a separate upgradable board, to ensure future compatibility.
All that said, sure, if I2S becomes a significant part of the connectivity mix, sure, we can add it in. Of course, that requires a bigger upgrade than our current modular upgrades, because it will affect the mix of connectors on the back of the product--which means a different inner chassis.
As I understand it, the big advantage of I2S over the other digital signal formats, especially USB, is that the source clock solely controls the signal transmission, so that clocking jitter becomes a non-issue, at least in theory. In more practical terms, and of course implementation is always a factor, but I’ve compared the various forms of S/PDIF against it, and the difference in musical clarity and fidelity is immediate and significant.
Of course data signal jitter elimination is one thing, cable and connector impedance is another. Listening comparisons consensus during development of a DIY I2S LVTTL/LVDS converter board and its DAC installations seem to indicate that higher quality interconnecting wiring of as short of lengths as possible is optimal. The beauty of HDMI for I2S signal formatting, in addition to supporting LVDS balanced signal protocol and growing industry standard acceptance, is that high quality cables are already readily commercially available for the discerning audiophile.
If Schiit’s 18K+ timing interpolation filter taps and multi-bit R2R ladder conversion DAC is all that you and Mike claim it to be, the Yggdrasil can potentially stand right up there with the best of them, at least 2, most noticeably. That's rather exciting to contemplate and explore, and it would be disappointing if quality of input signal proved to be its weak link. There is at least one very good USB to I2S converter available for alternative USB implementation options, so that additional I2S input capability would be a perfect segue into the high end audio market that you claim not to pander to. It also doesn’t make good business sense to deliberately eschew a significant market segment, if a little can potentially gain you a lot.
I’m not really pushing an agenda here at this point, just food for thought.