I decided to pick up the T100 yesterday... and I love it. I brought it home, set it up. popped in a few favorite discs and enjoyed... no analysis required. My basic thoughts at this stage are that I will optimize my 2 channel setup for redbook and my HT setup for SACD & DVD-A. My 2 channel system is based on CJ tube gear and the shanling, my multi channel system is based on rotel, and tweaked to death halfer amps. I've got tubes to work there magic on redbook and SS for the extra punch of SACD (with less need for smoothing)
So here is the big shocker...
I really like upsampling... I've always thought this was pure marketing hype... just a new word for oversampling. I did not really pay much attention to upsampling when I was at the store... just so happened that the players I like the most: cary musical fidelity, and shanling all up sample. I also really liked the sim audio players which do not.. The T100 lets you switch it on/off on the fly so it is a real easy A/B. The shocking thing to me was that I notice more of a difference with the upsampling on/off than between SACD & CD. ...Don't go nuts I'm not saying upsampling is better than SACD.. As discussed here before upsampling is just oversampling with a different noise shaping..no more information.. To me the soundstage seems bigger and deeper..but why?
This got me thinking a bit...
Here a complete guess so don't take this too serious... The thing I do for a living is build Voice /IP gear for large telephone carriers. One of the things we do is generate something called comfort noise. VoIP can be heavily compressed depending on the codec used. When it is received on the other side, the background noise has been removed via something called silence suppression. The dead silence (between words) is not desirable to someone listening on the other end since they are use to hearing background noise. Voice quality is judged with something called a MOS (mean opinion score). If we inject artificial noise (comfort noise) into the receiving end the MOS jumps by a huge amount. There is obviously no new information in the audio stream but it is pleasing to the listener and get a higher subjective score. If we measured the signal it would be worst since it contains more noise... I'm wondering if there is a similar situation in audio with tube gear, vinyl, and upsampling, etc... vs transistors, CD, and SACD.