I know that the Dolby Renderer app allows one to dynamically change the number of speakers they’re monitoring on, in order to see how the object-oriented information will scale up and down. Do you ever change your monitoring output to 7.1.2 or 5.1.4 to see how the listening experience changes?
Yeah, it's a bit like when you're mixing stereo and press the mono button to make sure there’s no disasters. That being said, I would never just create an Atmos mix and then generate the 5.1 and stereo from that. I’ll always do a dedicated 5.1 and stereo.
It definitely makes a difference, especially for fans who haven’t yet taken the Atmos plunge. On Nothing But The Truth, you made the interesting decision of panning the vocal mainly to the side speakers. It sounds great on a 7.1.4 setup, with the vocal positioned dead center in the listening space, but all that information would be sent to a single rear pair on a 5.1.4 or 5.1.2 array.
On the dedicated 5.1 mix, that shouldn’t be. I remember hearing from people at the time of release that the 5.1 was back-loaded, but they must have been listening to a fold-down of the Atmos. If it were the other way around, I think you’d lose a really nice feeling of having me right in the middle. There are always compromises to be made.