That depends on what you mean by flawless. In an absolute, theoretical/mathematical sense it’s not flawless but in practice, as a medium used to record and reproduce sound, then it is. The noise floor/artefacts of digital audio are way below the noise floor/artefacts of the transducers and analogue equipment in either the recording or reproduction chain.
Again, good in what sense? I’ve got no issue with most of it, except for arguably the two most important points(!):
1. The measurements are taken with the DAC’s PLL switched off. A PLL is a necessary component of a DAC, which exists to remove/reduce jitter. So, I presume he’s switched it off because otherwise there probably wouldn’t have been any measurable differences. However, as a PPL is a necessary component, I don’t know of any other DACs (consumer or pro-audio) which allow you to switch it off, because why would you want to? Therefore, these measurements are not applicable to any other DAC, or even to that same DAC unless you switch it’s PLL off, which I wouldn’t recommend!
2. The audible effect it has on the sound, is contradictory and in places false! Early on, the presenter states that none of it is audible and that’s not what the video is about but then later he states he can hear the difference. How can he hear what he (correctly) stated is not audible? The levels of noise/artefacts he measures are so low, they can’t even be converted (transduced) into sound. So obviously, if that noise/artefacts isn’t even in the sound being reproduced, there can’t possibly be any question of audibility. Not to mention that the levels he measured are about 1,000 times below the noise floor of even the most dynamic commercial music recordings and is even well below the noise produced just by the random collisions of air molecules! The part that is false regards human sensitivity to jitter and the testing of it. It is true that humans are very sensitive to jitter/timing issues. With music, sensitivity is around 200-500 billionths of a second (200-500ns). Compare that with the blink of an eye, which takes about 200-500 thousandths of a second (200-500ms), a million times more! And in an exceptional case, someone reliably detected jitter of just ~30ns. What’s false is the assertion that this hasn’t been well tested. It started being tested in the 1960’s, the first published paper I know of was in 1974 and it’s been tested extensively since then. Today, and in fact for more than 20 years, jitter is a non-issue for consumers. Although it makes a great (false) problem for audiophile equipment makers to supposedly “fix”! Even in the late 1990’s, cheapo OEM CD/DVD players and the DACs built in to digital TVs had jitter of just a couple of hundred trillionths of a second (~200ps), which is 100 - 1,000 times lower than audibility!
Again though, commercial studios connect their ADCs and DACs directly to their computer and with many/most commercial music recordings there’s more than one round trip through them. So that’s the jitter/noise/artefacts of the ADC plus the jitter/noise/artefacts of the DAC, all times 2 (or possibly more), that’s already burnt into the recordings you’re reproducing!
G