Manual Focus v Electronically Aided

[Rob C]

Funny old world. I was out doing some stuff with the D200 fitted out with a 2.8/24mm Nikkor of the old school - manual type.

Working quite close up, say about three to five feet from various things, I got the impression that there was an argument going on between the camera and myself regarding when a specific point was or was not in focus. In other words, when my eye thought I'd hit the spot the electronic focus assist (?) light would claim differently. This isn't just as simple as you might think to check out on the monitor screen because the images are of quite reclining perspectives and now, in retrospect, it's not easy to say exactly where I had intended the point of focus to be.

Is this something that anyone else has experienced to some degree?

Ciao - Rob C

[32BT]

Not meant to be funny, but on the Nikkon, how do you know that the diopter adjustment is set correctly?

[feppe]

I'm not sure what you mean. Using MF you get the focus point you want, not what the camera interprets you want, which might be the reason for the discrepancy. Why would you even look at the AF for confirmation if you're using MF - what's the point? If you mean that focusing with AF center point and recomposing yields OOF pics, here's the reason: http://visual-vacations.com/Photography/fo...mpose_sucks.htm

I also don't understand what you mean by not knowing where the intended focus point was. How can you not know? Are you shooting plain planks for perspective which have no focal point?

[Jonathan Wienke]

If you don't know where you intended to focus, how can you expect the camera to know? The discrepancy is likely caused by the AF mechanism picking something different to focus on than you did. Unless you're focusing and then reorienting the camera, in which case the focus-recompose sucks principle is the likely culprit.

[Rob C]

How to answer and make this clear?

Well, take the subject first. It's some rusty steel structure on the stern of a commercial fishing boat; it's a support for a roller over which the ropes/cables which pay out the nets get drawn. This isn't vertical, but angled in two planes to the shooting perspective (there are two supports: one port and the other starboard but the one of interest is the closer of the pair).

The interest is the roughness of the rust on the main part of the structure combined with flaked paint.  As it happened, the central focus zone of the viewfinder (the only one used for the checks) was more or less where I wanted to place focus, making it quite logical to adopt the belt and braces technique. Also, remember that the exposure was made at circa f8 which, on a 2.8/24mm lens and at the particular magnification in the shot, gives quite a wide depth of field. The eye-focusing is obviously made at f2.8 (and I understand about the possibility of focus shift on stopping down); but the electronic focus check is also having to operate at f2.8 so the two opinions should co-incide - no?

Why can't I tell where I was trying to focus? Am I an idiot? It might not be as simple (no pun etc.) as that (and the subject is not out of focus on the monitor), but it is very hard to guess, in retrospect, at which specific plane on a receding range of them (the rusty steel support) I was looking at the time of focusing. One bit of rust looks much as another and, with so much depth available, it masks problems when averything looks sharp anyway.

Really, the problem is evident at the very moment of focusing when the point inside the central a/f zone looks visually sharp but the light has failed to switch on. Ths isn't a case of allowing the camera to choose the focus - that's why I avoided buying a/f lenses even when I bought the F4s. Making my re-entry to 35mm, having dumped 35mm and the F4s for the Pentax 6x7 ll (yes, another sad day, largely driven by stock agency dogma), I avoided the F5 and bought one of the last available F3 bodies which I still have, along with a freezer full of assorted films... It and the D200 make an odd couple.

I agree that focusing on one spot and then recomposing the shot isn't best practice - that's the beauty of a screen - but I have to admit that I miss the cross-wedge system from film cameras too; great for speed and always seemed spot-on in my experience. But then, I never do sport; the closest I got to that was fashion or calendars where the big event was leaping into a wave...

Diopter setting of the pentaprism eyepiece. This is set at the point where the focus marks look sharpest to my eye.

Thanks for the suggestions.

Ciao - Rob C

[Jonathan Wienke]

If you were using center AF sensor only, the center of the image is sharp, and focus looked sharp in the center of the viewfinder at the time of exposure, but the focus indicator didn't come on, then the AF mechanism in your camera body is miscalibrated.

[Rob C]

If you were using center AF sensor only, the center of the image is sharp, and focus looked sharp in the center of the viewfinder at the time of exposure, but the focus indicator didn't come on, then the AF mechanism in your camera body is miscalibrated.
[a href=\"index.php?act=findpost&pid=92846\"][{POST_SNAPBACK}][/a]

That, Jonathan, sounds like a very distinct possibility; the other one was pointed out yesterday by a friend who said that it's vitually impossible  both to focus through a centre-spot marking, watch for a light and not move the camera at the same time;  hand-holding is probably not the way to go when you want to test anything - other than your steadiness - but there is still something happening which leads to this feeling of unease...

feppe: that was an interesting link you supplied, thank's for that. There is another problem too, particularly with w/a lenses, in that they tend to have a shell-like zone of sharp focus in front of the camera so they never can achieve the 'flat wall' style of plane of focus. This was demonstrated some few years ago by tests in The British Journal of Photography. We do not live in a perfect world!

Ciao - Rob C

[Gregory]

there are instances where the camera cannot auto-focus on the subject because of a lack of contrast in the subject within the focus area. if the camera cannot auto-focus, it cannot confirm your manual focus either. I have run into this several times in the past week.


... and I understand the part about not remembering what I was focussing on, even if I review the photos just hours after taking them. It happens to me all the time.