I have read quite a bit into that thread and nothing conclusive is there. Mostly its just people arguing about satire and post content.
From a technical perspective R2R should be better as D/S is approximating, it also makes sense why every DAC company has a "signature" sound if the DAC's run algorithms vs direct processing audio.
But yes its a more expensive design that also isn't very portable due to its enormous (by comparison) power usage. Something that they did find out in that thread though is that the current ways to measure audio output aren't really capable of actually measuring performance in music or at different frequencies.
But there is no other standard which was produced, nor any other testing methods which would allow us to effectively compare different products side by side.
to this day I haven't seen clear evidence of this so you must know something I don't.
R2R isn't approximating? you think each resistor used for the conversion is exactly at the desired value? ^_^ you think that analog music expects the signal to hold a discrete value all the way to the next sample? ^_^
don't fall for marketing. in both designs it's a mess where filters(and noise shaping for DS) come to save the day.
also let's not forget how you can find the same DS chip in 30$ DACs and in 2000$ DACs and they won't measure the same. it would be really misleading to dumb down the quality of a DAC to 1 chipset type.
I also don't get why the multibit name came out to define R2R? as dukefx mentioned, for many years now DS dacs have used more than one bit. the chip in the Odac isn't one bit as far as I know. in fact even DSD DACs dealing with 1bit signal, tend to use more than one bit nowadays.I guess R2R feels old(for good reason) so it needed some little rebranding to appeal to people again.
the audio industry has already given up on R2R, I don't know if there are still R2R chips designed for the purpose of audio anymore? but then again it's exactly the kind of reason that would make some audiophiles desire R2R. "can't have it? I want one!". ^_^ not a very technical reason.
as for Shiit, I remember Baldr saying(I'm paraphrasing) that making DS was boring because all you had to do was follow the manufacturer's advice. also not too sure how technical that reason is
.
now about measurements, there are logical limitations, when a DAC has better linearity but a little more noise(like it could happen with DS vs R2R), it's apples and oranges. how do we design a measurement technique to tell which has superior fidelity?
the other problem is that we're ultimately interfacing with music subjectively. so if we try 2 devices and prefer the one that measures worst, then we want to believe that the wrong thing was measured. it could be so, but there is also the very likely possibility that my personal taste isn't a fidelity measurement tool.
I'm not off topic, I said Odac.. twice now ^_^. sorry