I also note that recently Neil Young has been actively involved in this debate by pioneering the Pono Player. Whilst I have some reservations about this 'new' technology; if it does indeed bring the original high-fidelity, well captured, professionally mixed and mastered records to the masses without the need for dithering etc then I'm all for it!
The videos of Neil promoting Pono are a bit disturbing. Renown musician after renown musician proclaim it the best they've ever heard—extremely unlikely, and the apparent fact that they just auditioned it in car makes it even more unlikely.
Yet all these guys showed up. Because Neil is such a good friend that they want to lend their names to his effort? Or because it's CD-vue—people get to re-purchase their catalogs in another, higher cost format?
One thing is for sure with Pono: Any improvement will likely be lost on 99% of the potential listeners, the iPod (etc.) crowd, who will be listening with their marginal headphones and in their cars. Now, you may counter that this is fine, since it gives audiophiles a better source material, sponsored by the unwashed—OK.
If I have an instrument that I wish to capture with a microphone accurately then 44,100 samples a second of this source (with a 16-bit bit depth integer) does NOT accurately capture all of the details, nuances and harmonic content inherent within the original wave disturbance of the air molecules.
Of course, nothing does. Not analog, first, nor any part of the analog path that we'll be using for digital, especially loudspeakers. But digital is inherently an approximation. With 96kHz, you're giving yourself one measly octave more frequency headroom. If "nuance" were the issue, you'd probably want something more like 5MHz.
If this mix then gets mastered in the SAME audio format and distributed as a lossless file of the same parameters (in a 96/24 FLAC for example) then this will sound better.
I would think that the "lossless" part is the most important. (How much depends on the lossless format, of course.)
Admittedly, it will take a while for your ears to become trained but I absolutely guarantee you there is a marked difference and one that you should all be able to hear.
The trouble with binary comparisons is that you have a 50% of picking the right one. People that I respect have asserted that they can hear whether a 24-bit audio track is dithered or not. Allow me to understate it, out of that respect, and say that this is unlikely. Poking about on the web, I read a thread where one of these mastering engineers said he could hear the difference when a certain switch was enabled—it came out later that the switch was not to enable or disable dither, as he had thought, and didn't affect the audio—he admitted that perhaps he was letting his imagination get the better of him. And elsewhere he was also honest enough to say that he could hear 24-bit dither or not "on a good day". Well, if something's 50-50, on a good day you'll be lucky enough to guess right several times in a row.
That's why I'll be skeptical right up till the point where double blind tests show that the higher sample rate is preferred consistently. I haven't seen that yet, so if you have, please point me to the study.
Anyway, i'm just commenting on your comments, not implying that you can't hear the difference, not really aiming the comments at you. One problem is that there are differences in playing back at 44.1k and 96k in the converter alone, so when a difference can be heard, it's hard to be certain it's the source material.
