You seriously call that "the right light"? That's terrible light. It's exposed properly, the histogram was about 1/2 way into the right-most histogram bar...but to actually get contrast exactly on the eye, where you want focus to occur, is difficult when the subject matter around the eye is all shadows. I had to process the image to bring up the detail in the bird such as you saw it in the previously posted version.
The only benefit of a scene like this is for tracking, really...the strong contrast between bird and sky makes it easier for any camera to track the subject...it just doesn't do anything to help you lock AF on the part of the subject you want (yet the 7D did quite well in that respect.)
This image is not properly exposed. Your histogram is displaying the exposure of the entire image. The raptor is way under exposed. Your final image would have been much cleaner if you had pushed exposure to the right a couple of stops. But I do agree with your point regarding the 7d as a much better AF performer then the 6d and with the 5diii being king (never owned a 1dx due to the cost).
I personally had a love-hate relationship with my 7d and a love-love relationship with my 5diii.
If I had pushed exposure to the right "a couple stops", the sky would have been completely blown. Additionally, my shutter speed was already getting rather low...I wanted it low enough to produce blur in the birds wings, but if I let it get any lower, it would have resulted in the entire bird being motion blurred. Personally, I don't really like how bird in flight photos end up looking with a blown sky...and when you process to lift the deeper shadows (which would have still been fairly deep), you end up with a funky noise halo around the subject (not exactly sure why...seems to be a 7D thing...right where the sky meets the bird, that little border of ever so slightly blurred pixels, there is a sudden tonal drop from 255 to 245-250...looks nasty). For the scene, in order to preserve the sky, the exposure was dead on. The histogram was all the way to the right-most bar in the histogram chart. I may have been able to eek out another third stop, however when your primary task is to zero in on and track the bird, you stick with what you originally chose on the exposure.
Also, the post-processed version of the raptor:
Turned out pretty well in the end. Exposure with a digital camera is as much about knowing what you can do in post, as it is knowing how to use the histogram and where your highlight cutoff is. The shot was at ISO 400. At that ISO, you actually have pretty good DR (still around 11 stops), but you don't have nearly as much banding noise. So you can lift the shadows by quite a bit (in this case, I think I lifted them almost two stops). The whole entire exposure was increased (including the sky a little)...there was some slight banding visible originally, but after boosting the exposure the banding in the sky faded. There is still some noise at full size on the raptor's underbelly and in the lower tail...but overall it is pretty clean. I also really like the fact that the sky is still a pale blue, rather than a blown white! But, that is also somewhat personal taste...if you don't care about the sky, you certainly could have exposed more.
(BTW, check out Art Morris' book "The Art of Bird Photography", read the chapter on exposure, and let me know if you think Art would have chosen anything different. ;P)
I'm with you on the 7dii. If it is almost as good as the 5diii at high ISO (with easy to manage luminance noise) and has a new generation AF system with 8-10 fps.... It will be the perfect nature photographer compliment to the 5diii. My hopes are high and I don't think canon will let me down.
I don't know that I said the 7D II would be almost as good as the 5D III. I did say the gap between the 7D and 5D III could be closed with the 7D II. There will still be a gap, but I don't see any reason why it would be as wide as between the 7D and 5D III today. Even if we assume that Canon doubles Q.E. to 80% (unlikely), moves to black silicon (probably also unlikely), that might bring the 7D II's FWC at ISO 100 to ~60,000e-. The 5D III would still be ~68,000e-. The 7D would need even more technology to close the gap any farther than that. Color splitting along with BSI, on top of double the Q.E. and black silicon, might actually put the 7D II over the top with 70ke- or so...but that is honestly a LOT of technology to pack into a new generation APS-C sensor.
I think Canon is an innovative company...but these innovations have already been made, many are patented, and Canon is conservative. I think we'll see a jump in Q.E. to somewhere in the 55% range at best, maybe lightpipes, and a slightly weaker AA filter. I think that will really do wonders for 7D-class IQ...the 7D II will be a lot better than the 7D. It will close the gap in terms of high ISO IQ. But, realistically, the 7D II is probably at most going to have an ISO 100 FWC of around 30,000e- tops unless they reduce pixel count to ~16mp. There would be a very noticeable improvement in 7D II IQ over 7D IQ, and instead of an approximate 2-stop difference in high ISO performance with the 5D III/6D there would only be an approximate 1-stop difference. There won't, however, be any situation where the 7D II high ISO performance is actually "almost as good as" the 5D III or 6D. Nothing can really beat bigger pixels in that sense. Maybe layered photodiodes, but I've only read theoretical papers for non-foveon type layered photodiodes, and I don't even know if it is really a viable option...it's just a theory.
So, you can get your hopes up for increased 7D II IQ relative to the 7D. Just don't get your hopes up for "nearly 5D III level high ISO noise". That is highly unlikely.