Amir’s “review” does offer some interesting data, and reflects some of the shortcoming of the DSS, some that I think do need improvement in a future iteration of the hardware.
But it is important to note that Audio Science Review is a measurement fundamentalist site where the apparent majority of the gear they review and judge is never even listened to. When they do listen to gear it tends to be an afterthought, through inadequate systems, and heavily biased by their opinions of the measurements. There are many examples of good sounding gear getting bad reviews based on the numbers, but in none of the reviews will you ever see a negative listening critique of gear that measures well.
It seems apparent that the unit wasn’t burned in by the original owner or himself.
To the important data their measurements show:
-
The DSS noise floor is way too high. I think everybody knows this, if you need to use a 20db attenuator in order to use the DAC as a preamp, you have 20db too much noise.
-
The output voltage is low
-
The transformers add a bit too much coloration and distortion. This would be way less of an issue if Jensens were used in the original design as they are in the TSS.
I think the dynamic range thing is relevant as a lot of 24 bit files I play sound a bit dynamically compressed, even if the texture is excellent.
However, one point Amir seems ignorant of: People like the sound of transformers. Just about every pop vocal, rock album, you hear etc. etc. was recorded with transformed microphone preamps, transformered (and often tube-based) compressors, often transformer based mixing consoles, and tube microphones. There are even mix engineers who will run the entire mix through a pair of transformers. This is done specifically for the sound.
Because Amir (when he does listen) uses budget equipment he does not understand that what works in a real audiophile setup may be counterintuitive, the reality that DACs with high noise floor like the DSS will reveal no flaws and sound exquisite in a tube pre/amplifier setup. What makes a resolving and satisfying system is more than high dynamic range and low harmonic distortion. I think to properly evaluate audiophile equipment, you have to listen within an audiophile ecosystem. Hooking up a DAC to a cheap headphone amp and headphones with cheap cables won’t tell you the whole story of how it will sound in a particular high end system.
I hope he never gets his hands on a Lampizator!
The way Amir qualifies “detail” is interesting too. I have other DACs that easily best the DSS in terms of dynamic contrast, transients, and noise, but except for the Chord stuff I haven’t heard anything else that had better low level detail, and I’ve heard nothing with better texture. In this case he is being a sloppy listener and mistaking conventional PCM sound signature for detail. This is par for the course in that there is nothing here to demonstrate that he listened for more than a few minutes on a budget headphone system. I think for many of us, Amir’s reference for “detail” would probably sound pretty clinical and ugly.
Just to add, although we are talking about a digital product, from the perspective of professional studio recording and audiophile analogue equipment, this kind of design and measurement is pretty normal. That sweet Taylor Swift vocal was recorded on a transformered Neve preamp with far less dynamic range, and way more distortion than the DSS.