When considering this question, I think you have to break the system down to at least three basic components.
1. The A/Ds
2. The internal DSP handling
3. The D/As
Any of those factors could be responsible for a percieved sonic improvement, except in the case of importing a session recorded with Mix using other A/Ds. In that case, it could only be #2 or #3.
1. If the A/Ds are better, it's good news all along the food chain. That one's easy to see. The benefit is captured everywhere.
2. If the internal DSP handling is better, it's good news everywhere as well, unless you move the files to another platform for mixing, in which case, the result would be reduced to any benefit imparted to the recording due to more clarity while monitoring. The benefit is captured primarily when mixing internally.
3. If the D/As are better, the only benefit would be in monitoring, unless using an analog mixing or summing device. Mixing ITB would capture none of the improvement, except due to potenially better sonic choices because of improved monitoring.
The first, obvious quantifier would be to play HD recorded files in Mix. You could also route a stereo bus AES output from Mix into HD, and A/B the D/As. Or do a BTD on both, of the same session.
All interesting IMO. I will say this. Based upon what I've heard so far, which involves lots of songs from various producers and engineers that were recorded in Pro Tools being mixed on my non Pro Tools DAW, I believe much of the improvement being enjoyed by HD users is in the D/As. Some probably in the internal DSP handling. Not as much in the A/Ds as seems to be assumed. At least IMO.
Seeing as I pull PT files out of PT midstream, I immediately can rule out all of the D/A and most of the internal DSP. So what I will mostly hear on my system when comparing projects I get from Mix vs HD is the A/D converters.
I don't hear much difference between well clocked 888/24s vs 192s. In fact, in some cases I have preferred the sound of Mix based sessions as compared to HD based sessions from the same person. Weird, huh? I was pretty surprised, as that was not my expectation.
Please note that in no way am I saying HD can't, won't or shouldn't produce better end to end results than Mix. What I am saying is that I don't think people are breaking down the possible sources of improvement, then considering which ones are fully retained in the end, and how to maximize how much of the improvement actually reaches the end listener/consumer.
A question. If the D/As were to prove to be much of the reason that a Mix session sounds better when played on HD, is that always a good thing? Is that causing the engineer to get satisfied with his sounds sooner than he used to, not because they're actually recorded better, but because the DA's make them sound better than they did before?
Just a question. And it's not really just a question about PT. Do your monitors sound too good? Are they making you get sonically happy, sooner than you should?