FREQUENCY RESPONSE
THD+N VS. LEVEL
DISTORTION PLUS NOISE (THD+N)
DISTORTION ONLY (THD)
THD VS. LEVEL
THD+N VS. FREQUENCY]
THD+N VS. FREQUENCY
1 KHZ FFT
10 KHZ FFT
INTERMODULATION DISTORTION
AC LINE NOISE
THD+N is useless because it conflates two parameters which vary across the range into a single value.
THD is also more or less useless because it conflates the differences in performance across the frequency range into a single value. THD at 1kHz is insufficient.
Frequency response is the most obvious, which is why it usually comes first, but most decent DACs are flat from 20 to 20,000 Hz; this ties in with the response above Nyquist which shows the filter behaviour (Jussi Laako from HQPlaer defends that measurements should go at least as far as 1MHz)
IMD at the top of the range
FFT at different frequencies and levels show the noise spectrum, HD, linearity
Multitone is also revealing
My assertion is that: nobody has created a list of critical objective measurements required to assess the performance of a DAC, ranked the list in order of importance, given each measurement a weighting and score, and shown a definitive calculation to reach a conclusion as to the *total performance* of the DAC. And further that this list has not been submitted for peer review and agreed upon by the industry.
This is a normal method of evaluating the performance of a complex item that has many performance measures. And yet in the Audio world this has not happened as Tuga if it did you would be the first to link it to me as I have asked several times in the past for it and so I can only assume that this objective assessment work has not been done.
So what are we left with? A list of measurements that whilst objective in each individual case need to be assessed as a whole. As there is no agreed method for doing this then what people will do is apply their *opinion* as to the criticality and weighting etc of each measurement. And so we are left with a *subjective opinion* of the disparate *objective measurements*.
If you are to state that DAC A is better than DAC B then you have to sum the measurements otherwise you cannot make that statement
We agree again! That equipment performance is complex, which is why many parameters are measured.
Unless one of the measurements shows a significant level of noise/distortion/issue then which one is better becomes irrevelavant. It's time to listen.
And when I listen I don't focus on what it does well but try to identify issues and then see if I can find correlation with measurements (when available) and perhaps try to understand what is behing the audible issue.