It is very hard to explain from an objective stand-point how devices that have nowadays reached lab-grade performance in accuracy can sound different from each other.

Yet, expensive equipment with with noise floors that barely allow for accurate 16-bit reproduction and distortion on audible levels (“euphonic”) is highly praised.

People who claim to hear remarkable differences among different well-performing DACs often disagree on “what sounds better than what” and they resort to insult because, of course, they must be right and who disagree with them must be “deaf”.

I believe bias from sighted listening plays a big part in all this. However I have no trouble believing that mildly noisy DACs with somewhat added euphonic distortion can sound different and capture some of the listener preferences out there.

Can you achieve a better speaker placement?

There is no actual science, just a bunch of different narratives pushed by manufacturers and marketers and propagated by audiophiles.

Some say digital audio transmission is inherently lossy, but it can be verified this is not the case.

Some say it’s about the jitter, but it can be shown that jitter levels are reduced to negligible terms in modern hardware.

Some say it’s about what kind of noise or interference the streamer hardware induces on the system. There may be some truth to this but only under certain circumstances.

DSD it is actually the result of a primitive implementation of what is sigma-delta modulation (SDM), which is now the industry standard in A/D and D/A conversion. The reason it became a “high end” standard is commercial, not technical.

As a transport encoding, it makes little sense, as you can retain most of the advantages of SDM when encoding the result in PCM.

Merging Technologies, that pushes a lot on DSD, even had to invent a new brand for high-res PCM and call it DXD in order to justify that… PCM is a better format for production after all (their whitepaper on this is fun to read).

Of course it is perfectly possible that DSD releases are on average better recorded, as there is very little room for editing afterwards. I believe they would sound equally good if stored as PCM.

A bit of education allows you to spot obvious design mistakes or shortcomings in a given product. Speaker measurements help you understand what kind of value you are getting for the money.

There is huge variability in terms of value-for-money: certain brands overcharge for simple and traditional designs because they can, others put a lot of effort on R&D and it shows.

You can avoid expensive potentiometers or stepped attenuators by using digital attenuation.

You can avoid an expensive preamp using your DAC instead, provided you do not have analog sources.

You can avoid an expensive CD transport by converting your CDs to FLAC and using off-the-shelf hardware as a digital player running open source software.

You can avoid heavy power supplies and huge heatsinks by buying a class D power amplifier.