All sounds great, but given the 7D ML is still at alpha stage and has been for some time, this is more a theoretical achievement than actual.
"Alpha" is more a label to scare people who don't know what they're doing, since the ml framework is very stable by now a port to a new camera is also expected to be stable. So the "alpha" label more relates to the fact that you shouldn't *rely* on it for production and it might not be feature-complete then that it's likely to brick your camera.
Btw raw on the 5d3 will still be better than 7d because I'd expect the 1080p data rate to be too high and 720p has moire.
From what I've been following in the forums, the 5D3 is the one that is the most developed and working the best.
I wish they'd pool the resources, and concentrate on the 5D3, and get it 100% ready for "prime time" and bring that one at least to Beta, and have a straightforward method to easily load it, and also to easily get the RAW video out for a normal workflow, which appears to be straight into Davinci Resolve (after the 9.15 update).
At least get the one closest to working well out the door, and then concentrate next on the next best camera, maybe the 7D.
I'm a computer geek by trade, I do it for a living, however, after reading the ML forums, I'm not confident I could get the Alpha Stuff on, AND then correctly get the RAW video parts in, and am wary about some "flags" they say that can't be undone...etc.
I'm just not confident enough reading through the hundreds of links in the forums, etc...to get it to work, without potentially harming my camera.
So, wish they'd devote resouces to getting the first one ready for the masses, with good instructions and out the door, which seems to be the 5D3.....and then move onto the other cameras, rather than working a little on all of them, and taking forever to get any of them out into a stage that is easier to be used by the "Joe Sixpack" user crowd.