My concern is not necessarily that the flaws, shown here, would necessarily be audible under normal circumstances. But for example, Amirs lab may not be the worst case noise environment us readers may want to run our audio gear within... or some of us readers may not be as good at following equivalent best practice for setting up our complex audio rig cabling in ways that don't promote unwanted noises. There are so many variables to consider outside the lab.
That is a very valid concern. Usually everything in a test environment is properly grounded as to avoid hum etc.
IRL situations with PC's and external DACs as well as amps or active speakers all connected to mains can turn into a nightmare with low level hums, weird 'noises' etc.
For all these variables one cannot test.
Weird common mode currents etc. are hard to find and diagnose sometimes. Certainly when one would want to solve something remote.
It can be as simple as 2 devices stacked which shouldn't be stacked, faulty interlinks but also weird ground loops.
You cannot test all situations under all circumstances.
For EMC tests there are also standard ways of testing and applying signals. One has to have standards. Reality may yet differ again from those test results.
I have had to tackle this kind of problems occasionally and know it can be a real pain to figure out where something comes from.
Everyone that ever worked with a scope has seen weird pulses one knows aren't there but its origin is difficult to pinpoint.
Measuring something properly, especially when more than 1 device is being used can be a challenge to say the least.
So can connecting various A-V components be.
Fortunately in most cases it works problem free or nasties are there but below audible levels.