Given what you just said, i took a look at the parasound. Apparently it uses a relatively new B-B (TI) chip based resistor pair volume control. I have a bunch of experience with such chips, but NOT the BB. They are very good, but my experience drove me to design my own discrete equivalent. Maybe the BB unit is better than the several i designed with.
So.... if the ESS let’s you do volume control via digital math without the "truncation problem" - i would tend to use the ESS volume. On the other hand the BB chip in your Prasound is basically a bunch of chip resistor pairs with chip switches selecting one pair vs another which is also pretty darn good.
So i don't think you have a clear cut winner either way.
If you want to learn more about digital volume control and its norml issues and why ESS is different they have a good PDF slide presentation on it available publiucly.
Warning: math.
Further thought (edit/add): the downsides of these "digitally controlled analog volume array chips" (what a mouthful) are the basic material used, and the fact that 90% of them depend on the inclusion of a chip opamp gain/buffer stage. Its one reason i ruled out many of them. here’s the point that’s relevant to you: those distortions exist whether you employ the volume in the Parasound or not. Its not like by setting it to full volume, the opamp is eliminated. Its still there as a buffer. I am not criticizing the design BTW - there are convenience / cost / sound / tade-offs everywhere.