A couple of tiny avian friends gave me the chance to try the a6700’s claimed “AI” autofocus, specifically with bird eye detection. Having used earlier iterations of bird eye detection on the a7IV, R5, and R7, I was prepared to be underwhelmed.

The Canon bodies were quick to detect the critter’s eye in the open, but would quickly run the focal plane to the most distant subject matter in the frame much of the time if the bird happened to be in the midst of some foliage. Especially with the fast-focusing RF 100-500mm, it got to be maddening.

The a7IV could “see” through foliage a bit better, but was often slow to spot the critter’s eye and didn’t always do a good job of tracking. It didn’t help that my last copy of the 100-400mm GM had an infinity focus defect that could be found only after the lens had spent a few weeks back in Japan, and that it would have cost a fortune to repair, being gray market goods. THREE lessons learned in that chapter.

The a6700, with limited time using bird eye AF so far, appears to be better than any of the aforementioned, though not perfect. The 70-350mm lens could be at least partly to blame, though, as critically accurate focus is occasionally missed, but only in dim light. In good light, it acquires quickly, tracks well, and focuses accurately.

Which leaves me at a loss for what to acquire as a dedicated bird lens. An encore with the 100-400mm GM, for which production was moved to China a few years ago, is not an option for me. The 200-600mm G is a spectacular lens, but bigger than what I’d like for travel. The Sigma 500mm…well, I guess it would be okay if I managed to avoid getting one of the admitted 10% defective copies. But Sigma always has a “but…”

So, forget lenses for the moment . . . just let me find some birds!

a6700/70-350mm G . . .