Firmware 3.40 issues

Started 6 months ago | Discussions thread
just Tony
just Tony Veteran Member • Posts: 3,848
Re: Maybe AF-S is non-optimal?

LarsHP wrote:

just Tony wrote:

LarsHP wrote:

Writern wrote:

What AF settings do you use?

AF-S + subject tracking or Eye-AF.

It doesn’t take a lot of breathing to move a body away from where it was when focus was achieved, when focus was achieved just once. Even on static subjects, if I’m shooting in close quarters AF-C is often more reliable for me. Just a thought.

Right, if there is a movement before the exposure, but in the situations I have tested, movement was ruled out. That's why I used AF-S.

Excellent, my test last night was also not dynamic so it's relevant to report.

I used my 85/1.8S set to f/2.8 on my tripod mounted Z7 using AF-C eye AF. I was the subject sitting about 6 feet away. I used a radio trigger and allowed 1 second on half press before firing the shutter each time. I oriented my head in several different directions both side to side and up and down, and for one sequence I kept the same orientation but changed the direction of my eyes.

Without my glasses it nailed on the correct iris 35 times out of 36. The dof is close to 1" and the one shot that wasn't perfect was out by about 1.5 inches. I can't know why it missed because I wasn't at the eyepiece but what the camera settled on was the nearest thing with good contrast and good light, the hairs in my sideburns.

The lighting in the room was really poor, just an overhead ceiling light that put my eyes into shadow from my brows. But the details in the selected iris were vastly better than anything I had ever achieved with a DSLR. Eyelashes and eyebrows are common finds for the D800 and D810. Bottom Line: I trust the AF-C Eye AF mode in the Z7.

The experience was interestingly different when I put my eyeglass on. It nailed the shot 25 times out of a possible 29. The four that missed had something in common, troublesome reflections on the glasses. In one of the failures there were blown out reflections of the ceiling light. The other three were intriguing - there was a reflection of my monitor, with one corner of the rectangle situated over about half of my eye. This reflection wasn't strong so I could see the eye through it, but maybe this was enough to confound the eye recognition algorithm. All four of these missed shots also had the camera sharply focused on my sideburns, so there was nothing like your phrase "focusing on the background" as I understood that to mean. It didn't stray far.

If I was at the eyepiece I would have corrected the reflection situations before shooting. So I'll estimate my keeper ratio at 97% based on the no-glasses portion of the test; care is advised for eyeglass wearing subjects.

-- hide signature --

Wag more; bark less.

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow