Brilliant lesson in realities of AF - Roger Cicala's article

Started Aug 2, 2012 | Discussions thread
Shop cameras & lenses ▾
Flat view
Ednaz Veteran Member • Posts: 4,707
Brilliant lesson in realities of AF - Roger Cicala's article

Yeah, I know it's an article about Canon AF, but both Canon and Nikon technologies for focusing have a bunch of things in common: they both have the option of phase detect and contrast AF; they both allow for tuning of lens/body combinations; they've both gotten better over time but in ways not necessarily easily understood.

D800 AF bashers should read and learn. There are definitely those who have a legitimate problem introduced by what seems to be a failure in programming of the AF system. But for those who have turned the problem into a massive end of the world conspiracy theory and a general rhetorical bash on Nikon AF, the article explains a LOT about how AF actually works and what it's actually capable of. Pay attention to the lessons Roger is giving on AF, in quotes like:

"Microfocus adjustment makes absolutely no difference in the shot-to-shot variation in AF–it only improves the average value of the group. But I got tired of explaining that to people in the previous articles. It was easier to just do it than to answer another 50 emails."

In other words - you can micro-focus adjust until the cows come home, but there'll never be a flawlessly, perfectly repeatable result because we live in a world and time where that's not possible. You're tuning an average result, not an absolute one.

When I did microfocus testing on a few of my lenses, I did many test shots before doing the analysis, and found that there were small differences caused by: distance you're testing at, direction in which the AF is moving (from infinity or closest distance), zoom extension, age of body, whether the lens was AFS or not, age of the AFS technology... and because I looked at average performance (5-10 shots going infinity to 15 feet, 5-10 shots going closest focus to 15 feet... etc) with combinations of the variables (did it at 5 feet, 15 feet, 30 feet, for example) I only found ONE lens that really needed tweaking. All the rest were, when results were averaged, focusing properly given the state of technology today. Wasted most of a weekend finding that out, but learned a ton.

The differences between my D800e, D700, and D300 were big. The D800e had a much tighter variation range around "perfect" focus. MUCH tighter. If my sample is any indicator, the D800e focusing system is quite a breakthrough. Wish Nikon hadn't blown the great news with what appears to be a process slip up.

But, when I'm shooting with f1.4 or f1.2 lenses wide open, I'll still be live view focusing them when I can, and manual focusing when I can't, because the remaining tiny variability WILL bite you from time to time. Not always, not mostly... but just enough to make it worth trying to avoid. And my primary lenses will still be f2.8 max aperture, because that's about where state of the art lives.

I'm thrilled there are people like Roger who have time, skills, inclination, and thoughtfulness to do work like this. I find doing proper testing not much different than crawling naked over broken glass, which is why I don't do it much. (And, if you don't have his infinite patience, knowledge about what's good practice, and analytical skill - you really shouldn't be trying to focus tune your lens/body combos...)

Bravo to Roger Cicala!

 Ednaz's gear list:Ednaz's gear list
Fujifilm X-Pro1 Nikon D800E Panasonic Lumix DMC-GX7 Fujifilm X-T1 Nikon D810 +41 more
Nikon D300 Nikon D700 Nikon D800 Nikon D800E
If you believe there are incorrect tags, please send us this post using our feedback form.
Flat view
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow