Before we get started there are two housekeeping items to attend to. First, apologies for yesterday's long post. Second, as some correctly pointed out, a byte is technically 8 bits long and thus a 32 bit word has 4 bytes (4x8 = 32). I was hoping to not make yesterday's post longer than it already was and not get even more daunting, but....
And speaking of yesterday's post where I explained how DSD256 and DXD contain the same number of bits and information which begs the question why wouldn't we just record everything in DXD and skip the DSD256 process altogether?
I'll give you three guesses and they all start with because it sounds better. :)
We've spent countless hours comparing the two and for whatever reason the analog to DSD conversion process sounds remarkably better than the analog to PCM conversion.
Why? I have my suspicions but, the truth is, I don't know. We've spent hours comparing live recordings (no mixing or volume adjustment) in DSD256 to live recordings in DXD and find that DSD recordings are noticeably more open, live and spacious. Recording in DXD feels a bit more wooden, stiff, and a touch digital.
What's weird and unexplained is that when we capture analog in DSD and then rearrange those same bits into DXD there's almost zero difference—which I do not have a good explanation for.
Maybe others reading my post have a clue?
In any case, just some observations from the recording studio.
YMMV.
