adamdea
You are not a sound quality evaluation device
These are all broadly speaking the sorts of things that someone might say while stroking his beard philosophically. But where do they actually take us? Empirical knowledge is not certain in the Cartesian sense, but aside from tautologies and possibly maths, nothing is. Without condescending to the particulars you can't get to grips with where on the spectrum of real-world certainty we can place the conviction that anecdotal reports from a self -selecting group of people (whose hobby is listening to bits of wire) that they can hear differences between different usb cables is unlikely in fact to be result of actual differences in the sound pressure waves reaching their ears when the different cables are in circuit.But it might be... the simple answer is that we don't know for sure. Measurement may suggest that there is no difference but this relies on the accuracy of the measurement and that we are actually measuring the right things. Ironically there is an element of explanational bias applied to measurement results as well (i.e. we don't expect them to be wrong!). I'm in no way suggesting that all measurement results are bunkem but we do need to have an understanding of the limitations of measurement.
I would venture to suggest that it is a reasonably sound summary that based on an understanding of
-how digital systems work and the function of a usb cable
-the sorts of differences found in particular cables and the explanations proferred for the supposed benefits
-the limitations of human hearing (including for example just noticable amplitude differences and distortion levels, the ability to pick up signal at a certain level given a certain level of noise, masking effects etc)
-measurements of the outputs of real world dacs with different cables in them
-the (un)reliablity of anecdotal reports of sound quality comparisons (including a number of points mentioned above).
-the fact that differences in (and improvements between) data cables operating within spec are not generally observed or exploited in other more critical fields of instrumentation (medical scanning machines, radar, seismology, astronomy....).
......it is far more likely that the reports are based on something other than detection of differences in sound pressure waves reaching the ears of the listeners. In fact one could go so far as to say that there are no real grounds (other than Cartesian doubt) for supposing that the explanation lies in unmeasurable properties of the dac output and/or hitherto unidentified hearing faculties in the listeners.
I concede that I have no expertise in any of this; but I really do care about the answer, so I have spent a long time trying to learn about this stuff because it bugs me when I can't understand something. I started out assumign that what could be described as the standard hifi magazine position (human hearign is unlimited, our understading is in its infancy, there are loads of examples of things that sound better but measure the same, all cables sound different, bits are not bits, humans can detect unmeasrable levels of jitter) must be correct. I have not found pretty musch anything to support it. A very long time ago I asked John Atkinson (editor of Sterophile) for recommedations to read to understand digital audio. He pointed me to textbooks by Pohlmann and Watkinson. Imagine my surprise when these texts both mocked audiophile reports of sound quality differences between cables! It is rarely acknowledged in the mags that much of their content is considered laughable by experts. (cf my point before that a hifi forum is like a flat earth convention.) I'm genuinely interested in this stuff and happy to learn new things. Sadly I have not yet heard any thing apart from "hey its a hobby" and "but science is sometimes wrong" to support the standard hifi magazine position. I appreciate that people may seem annoyingly evangelical on cable threads, but the reason for this is (at least in part) that it is only through hifi forums that enthusiasts get exposed to the "real world" position as opposed to the hifi magazine guff. if you are actually interested in this stuff though, there is the exciting opportunity to find out stuff from people who really do know what they are talking about (not me)
Equally ruminating on "an understanding of the limitations of measurement" is great. But what limitations? Either way what are these "limitations on measurement" and how are they relevant? Incidentally if you really are interested, it might surprise you to know that one person on this forum has written a leading university text on the subject. If you are interested I would recommend reading it. If you ask nicely he may also explain things to you.
Exactly- all the dac needs to know in order to convert the recorded signal is the sample values and the sample rate. If it has these then the cable has done its job. Equally the cable cannot give any more information.For a USB cable we are measuring the right things; bits in, equals bits out. It can't be anymore than that. What the DAC does with that is up to the DAC.
In addition we can confirm this by measuring that the dac output is the same with both cables down to -130dB or more (and essentially down to the level of variation shown by repeated measurement of the output using the same cable)
Probably yes. But of course a cable which does not have a filter attached and merely consists of conductors can't do anything about this. If it has better conductors it will just conduct the noise to the dac better.I'd be more worried about the power and gnd lines in the USB cable.