Question for those who are talented in things digital audio. These forums abound with examples of people preferring one bit-perfect source to another, both playing the same source material. There’s subjective differences in the “digital chain”, with the same hardware. Some say that there are measured differences in SPL between two chains that differ only in the software used to deliver supposedly bit-perfect, and hence, by definition, identical digital chains. How can that be? If the bit-stream arriving at the DAC is the same, how can the SPL differ?
More details & an example: If two bit-perfect sources (say Volumio playing a file directly from my NAS via Pi2AES to my DAC, and Roon playing the same file directly from my NAS over the Roon plug-in on the same Volumio device via the same Pi2AES to the same DAC, and neither one has any DSP enabled, neither one is using software volume control, Roon shows the “two line pathway” and Volumio only has the Roon plug-in enabled) sound a little different on the same song, I tend to chalk it up to differences in volume. But that doesn’t make sense to me.
I realize that we might need to actually check that the two digital streams are identical. But then what do we mean by bit-perfect?
Here’s an example (not pointing to this one smart user, but I’ve read many such examples on this forum):
[Full disclosure, I’m of the “bits-is-bits” mindset, but I don’t doubt those who find differences between 2 bit-perfect sources on the same DAC/chain, and in fact a big part of me is jealous of those who can find sound quality improvement that way. Much as in other areas of my life, my “rationality” may keep me from finding pleasure in improvements whether real or subjective.]
If you deliver the same exact bits to the same DAC, and if you don’t change anything in the DAC settings, and you also don’t change the analog chain after the DAC, there’s still a variable left: the position of the analog volume knob. That’s the only difference I can think of.
If two bit perfect streaming software solutions playing the exact same file while running on the exact same hardware with the exact same settings are perpetuated to be audibly different, they can be measured with current technology if one really would be interested in finding out, but that‘s where the typical subjective listening stance comes into play and hence, no hard facts are ever established.
Those who can and do measure, don‘t find any differences.
I think it has more to do with some conglomeration of expectations and biases…
Well, I’m trying to point to a very specific and fairly straightforwardly measured difference - that of SPL. People often say “well, of course you perceived the difference, because they’re at different SPLs”. And some go further to measure SPLs and show that different nominally bit perfect streams have different levels through the same chain, again according to reports without touching the analog knob. I haven’t done this myself. Just wondering if it’s theoretically possible or even meaningful.
It would need a fixed SPL meter mount and acoustically identical conditions during measurement.
Performing acoustical measurements is subject to many environmental influences that are hard to control, so it‘ll need extended measuring times with i.e. pink noise and lots of averaging, something that has likely not been done and reported.
To circumvent the problem, measuring the DAC or AMP analog output electrically would be the better controllable method.
I still don‘t believe the anecdotal report would hold close scrutiny - just measuring something somehow and not considering proper control of the variables doesn‘t cut it.