Thanks for your interest in our developments!
MIR 3D is still work-in-progress, so there might be changes in the feature set in general or specific changes I can't anticipate, but since you asked, I'll try to give you an idea:
- MIR 3D will based on 3rd order Ambisonics throughout its whole signal path, but of course you will be able to decode in 1st order like MIR Pro does, too. We plan to allow for a painless transition from MIR Pro to MIR 3D (not the other way round, for obvious reasons).
- There _will_ be sonic differences for several reasons, so if you need bit-identical recalls you will have switch back to Pro. 3D and Pro can't coexist side-by-side, otherwise there would be no straight-forward way to migrate existing templates to the new version.
- Format-wise the sky is the limit, from a technical POV. :-) MIR 3D will enhance the existing concept of "capsule-based" decoding (i.e. the individual capsules of the virtual microphone arrays we have in MIR Pro) with an approach that could be called "coefficient-based decoding". The latter can be seen as a way to sculpt and organize the Ambisonics output from a loudspeaker-perspective rather than a microphone-perspective. Sounds a bit confusing, I know. 8-) ... In general, things get _really_ complex quickly when dealing with HOA, so there will be (or better: will _have_ to be) Output Format Presets to deal with multi-channel mixes.
- These output formats will include all typical surround- and 3D-configurations, some of them maybe even in several flavours. MIR 3D will also be able to deliver the un-decoded Ambisonics audio stream, which opens whole new areas of use-cases. But MIR 3D can _not_ output Dolby Atmos-metadata. You will most likely integrate its chosen, discrete multi-channel output in form of so-called "beds" with your Atmos mix. There are no moving "Objects" in MIR Pro, so that's a non-issue for now.
- A final word of the notion "3D" in context of MIR: As you all know, MIR is based on impulse responses from real halls. IOW: If there was no IR source position, then there's nothing we can put a signal in. Of course we found ways to interpolate and do some other nice trickery, but please keep in mind that this is _not_ an algorithmic reverb engine, so don't expect the option to have a singer flying through the hall. ;-) MIR all about a holistic, enveloping presentation of real (!) rooms and the way instruments (or other sound sources) interact with them.
... let me repeat that this info is based on the most recent developments, but things still might change for the actual release version!
Thanks again for your interest! Highly appreciated.