As I’ve mentioned, there is a big difference in amount of steps when using dB vs percentages. Therefore, there’s cases of non-exact approximations when converting the dB value to %.
For instance: with the NAD C658, the minimum value for percentages is obviously 0% and the maximum is 100%. It uses whole percentages as a unit, so there’s 100 steps of volume change when using percentage (obvious stuff).
In contrast with dB, the minimum value is -80dB and the maximum value is 12dB. And here the units are separated by half a decibel (0.5dB per step), which makes the total amount of steps when using dB 184.
That difference between 100 and 184 makes it non-linear and non-exact when converting the two values. So, NAD is just giving you an approximation in % in this case.
Roon Ready devices need the volume level to be exactly the same. This is in order to have only a single volume changer between the whole signal path to keep everything as bit-perfect as possible.
Now, for a developer, does this mean it’s impossible to implement both values within Roon? Of course not, it’s possible, but it’s also an extra hassle for something that’s quite benign. And as someone who has more than a decade of experience in the field of software engineering, I implore you not to underestimate feature requests like this. You might think it’s just a quick fix that can be implemented within a few hours or less, but in order to do this right, without making the code less maintainable and/or less scalable, it could take maybe 10x or 100x more time than you might think to do this right. Now, I’m not a product owner/manager or engineer working for Roon, and I have no idea how the code looks like, but if I was involved, I think I’d rather spend recourses on other features.
In my opinion, volume should always be displayed in dB. This was the case before digital audio was a thing (and still is). Some inputs always required a different dB adjustment than others (think vinyl vs radio vs whatever). And with the arrival of more and more types of sources, this discrepancy will only be more truer and truer. I imagine in a percentage-first volume world, it’d probably require a universally accepted chain from the source to your ears to be digital from start to end (which is impossible, unless you’re a robot or something, since soundwaves are analogue. That’s why we have DACs in the first place). dB is still the most useful unit when it comes to audio volume in practice.
Hope this explanation helps you understand the issue a bit better.