Has anyone ever answered this question? I notice that you have asked the same question or one very similar to it elsewhere on this site. It raises a fundamental issue: How do differences in these specific measurements relate to what one hears? What data can be presented to support the assumption that a SINAD of 115 is superior (from the standpoint of listening) to one of 100? Otherwise, I can't see how this anything more than elaborate upmanship about specs.
SINAD is not a measure of sound quality. One can have 2 DACs with the exact same SINAD yet can sound very different depending on phase or amplitude deviations.
SINAD only says something about the noise level + distortion of a 1kHz testtone (at a specified level) at a certain output voltage.
You see it says nothing about sound. It is just a number generated by a test device.
Just like S/N ratio or THD at a certain frequency, IM distortion at certain frequencies or specified within a certain frequency band are generated numbers one can easily rank.
When one ranks such numbers you can get a good sense how various devices perform in that single department.
It does not include ALL measurable aspects that matter. Just the one in the graph.
One can say the one with the best numbers is the best performing one and could conclude that (at least at 1kHz) it is objectively the best measuring DAC.
Then look at FR response, filter response, multitones and other aspects such as waveforms etc. and then you could rank the whole DAC and perhaps get a better 'feel' of the actual sound quality.
Our persistent, but hard learning, friend in this forum compares output to input (nulling) and ranks according to that. In principle this could generate a better 'single number'. Nulling, however, has the disadvantage that time and amplitude differences are both converted to amplitude. So one needs to 'ignore' time issues that are inaudible and psycho-acoustic effects of certain types of distortion should be taken into account as well to get a meaningful 'number' out of it. Requires more effort.
In the end specs matter and DO say a lot about signal 'fidelity' but one needs to look at far more than SINAD and needs to understand audibility thresholds with music (and not all music and recordings are the same). It takes effort and knowledge to say something about SQ based on a set of measurements. Those that do not (fully) understand will say: measurements are worthless.... yeas to them because they do not understand it and may have a warped sense of audibility thresholds of various measurement aspects. Doesn't make measurements worthless.
Then we have the next hurdle .. the brain ... but that's another story.