On a slightly down note, it seems the race for the top place in our ranking is resulting in some design decisions such as removal of output buffer and increase of output impedance. I am not too happy about this but haven't decided what to do about it.
I just wanted to follow up a bit on the buffer aspect. I specifically bought this DAC
because it uses a passive filter output - I don't consider the lack of a buffer after this passive filter a defect or 'chasing specs'. Personally, I'm happy that someone did this, because if I were designing a DAC, I'd do the same. The best solution to get the least distortion and lowest noise from an active electronic stage is to
not have an active electronic stage. Passive components (if not taken to silly extremes of course) do not affect anything other than the function they're designed to affect, whether its a filter, attenuator or whatever.
Like I posted previously, the voicing filters I use in my active crossover system (all the components of which I designed) are entirely passive LCR (inductor, capacitor, resistor) types. There's just a buildout resistor and then a couple of parallel band-attenuation stages which are tuned to the specific frequencies I need to tune to get flattest speaker response in my large room. In the 'old days' this was extremely common - it was the rule rather than the exception. Up to the 50s / 60s, motion picture sound mixing consoles were largely passive with gain stages only as needed to provide output levels compatible with the recording dubbers. Recording studio mixing consoles were largely the same, being mostly passive. I built the 16 track mixing console I used in my first recording studio to be passive except for the microphone input gain stage and the summing amps.
In these older systems, 600 ohm termination was the standard, and most audio lines provided this source/load impedance (there were systems with 'bridging' inputs but that's beyond this discussion). Today of course the 1:10 rule is the norm with the source <= one tenth of the impedance of the destination. This applies to this M400 DAC if no affect whatsoever is required on the DAC's output. With input impedances less than this, the result is some level attenuation, but with normal consumer gear this would be quite small (my system just violates the 1:10 rule since my input is 10k Ohm).
With gear with very high output impedance such as some vacuum tube preamps without cathode followers, cable capacitance can cause high frequency rolloffs, but that is not the case here.
I haven't had this DAC long enough to comment on reliability aspects, but in all other areas I'm happy with it. As I mentioned in another post, I wish the manufacturer would be more communicative, especially since they have a presence in this forum, but perhaps they will jump in.