ProtegeManiac
Headphoneus Supremus
Read the info on these links for a sample of how they do native DSD.
http://astellnkern.com/eng/htm/ak240/ak240_feature01.asp
http://astellnkern.com/eng/htm/ak380/ak380_feature01.asp
I've read it and there's really nothing helpful there. What do they mean exactly by:
Since the main CPU does not process Native DSD, we needed a chip that will work with it. It took us a long time, but we made it work. Through countless hours of testing, we were able to achieve Native DSD playback by having the main CPU process the data, send it to the exclusive XMOS chip, then through the DAC to deliver the sound.
What exactly does the XMOS chip do to the DSD file and how does a DAC that can't do 2.8Mhz decode it?
-----------------------------
I found more info just looking up the CS4398 chip, which A&K did not mention specifically as the key to all that. Apparently, it can go up to 216Khz (still around 1/10th of what SACD actually uses), and has a 50khz filter. The only thing that is really identical to SACD is the claimed 120dB dynamic range. So, if in effect it can deliver such a dynamic range despite the other spec disparities, then it's delivering nearly all that you can get out of SACD.
The other practical reality that needs to be sorted out though is whether what one listens to actually has a 120dB dynamic range. AFAIK, it would likely (not necessarily the only ones) be properly recorded classical music on SACDs that would have that kind of range, so it depends on how much one will invest in terms of money, HDD space, and compatibility with devices that one uses if DSD will be worth it.