Think of it like this...
DSP works with samples. It deals with samples only. It doesn't really care much what the sampling rate is because it is just looking at the bits that are available.
In 24 bit, you have 24 bits of info per sample.
So, when you go to apply DSP, which is based upon algorythms that work with induvitual bits of information, the algorythm is set, it is not going to do any more then what was coded. Right? Right.
So, since you have X amount of bits per sample, if you have more samples, you have more bits of infomation for the algorythm to play around with. What this usually means is that the processing will be smoother.
When you down sample, or resample, another algorythm is used. With higher sampling rates, of course this algorythm is going to work better because it is some sort of DSP. Not only will a higher sampling rate mean better DSP "percieved accuracy" when it is applied to the higher sampling rate file, but DSP "percieved accuracy" when you use the DSP to down sample. So, you start with a good file, then apply DSP for whatever, and it is good DSP because more processes took place, then the process of "dummying it down" is also better for two reasons: 1- because there is a better audio source for it to work with 2- more processes are going to take place during the process of "dummying it down".
Now you know.
Rubert Neve is an electrical engineer, not a recording engineer. I am sure he has picked a few things up over the years in the studio, but I have never seen his name on a CD credit.
About the 192KHz sampling rate deal, this is not really true or false. It has merits in some circles, and none is others. Hell, there are people who still believe that 20Hz-20KHz is adequate for "real sounding" audio....
Since digital can effectively record half it's sampling rate with a great degree of accuracy, then it would stand to figure that being able to record sound that is around 98KHz would provide significant sonic accuracy. In theory, this is a good idea. Many brass instruments produce tones that push 100KHz. But, one small little problem with this....
It would not be that hard to create 24/192 converters with a full dynamic range and signal to noise ratio available. But, a sound has to be delivered to it. That is the problem. EVERY microphone has a self noise. Even Rupert Neve's preamps generate noise, and only have around (at best) 100db of sound to noise ratio. So, while sound that we hear live if effected by frequencies that are well beyond what we can hear, you also need to be able to catch that sound on tape to have "true" accuracy. Not a single microphone I know of, even test microphone that are used for research, are capable of "accurately" catching tones that are pushing 90KHz. The mic itself will change those tones through it's mechanical conversion. Throw in the preamps limitations, and the wires that are used to connect this stuff into to mix, and well, you just can't record stuff that high with true accuracy. Period. Presently, it cannot be done, and if it can (who knows) it cannot be done affordably (even to the big time studios). So, 192KHz sampling rate is bogus presently with even the best mics and preamps and wire available.
Analog recorders only can record up to about around 40KHz, and above I believe around 30KHz, it starts to really suffer from some weird stuff that will effect the audio.
I don't think that 192KHz sampling rate is what is needed to produce "analog sounding" results in the digital realm. I think that digital will always have a more "cleaner" sound to it, and that converter quality is a big issue. 96KHz sampling rate is relatively new, and I believe that as the manufactures look to "reinvent" the stuff, that 96KHz will prove to be quite adequate to produce "analog type" quality. But, analog, even at it's best, is not very sonically accurate, so even if higher sampling rates become the rage, it still will not sound like analog, because it is not analog, and it will never be analog. I think the only way you will get digital to really "sound" like analog is through emulation. Of course, now we are back to having higher sampling rates available so that the DSP can replicate the "modeled" tape deck much closer.
You wanna know what will be next is digital? Higher internal bit processing. Most stuff is at 24 bit (the stuff most of us use) and most high end pro gear starts at 48 bit, with a few high quality digital boxes providing up to 72 bit.
Once more actual processes can be applied to the audio that is in the digital form, the better "percieved" accuracy DSP will have. Until internal bit resolution is up'ed significantly, digital is going to suffer from it's present day lack of "warmth" and "subtlety" that high end analog enjoys.
Hell, I could go on and on, but the point is that more bits are available for DSP at higher sampling rates. Plus, aliasing and what not are significantly minimized with 96KHz. While you ears may not be able to tell much difference between 96KHz and 48KHz, you DSP will!!!
Ed
. I bet his monitors don't suck. And his ears are probably not too bad. 
