18-200: AF fine tune & why not so soft at 200mm

Joined
Nov 23, 2008
Messages
117
Location
Australia
I've been testing my lenses with my new D7000 to set the AF fine tune.

What I have found with my 18-200 is that it requires these settings:

+2 @ 18mm
-2 @ 50mm
-8 @ 200mm

So if I left the AF fine tune at 0, it would begin backfocussing as I zoomed out towards 200mm.

The reputation of this lens is that it is sharp enough until about 135mm, after which it becomes soft.

My findings suggest that this softness may be due to AF, not optics.

In my case, most of my shots are between 18-60mm, so I'll probably set it at zero, unless I need the telephoto (e.g. trip to the zoo), in which case I'll bump it up to -8 at this time.

Thanks be to AF micro adjust - darn useful tool for squeezing more life out of a lens I reckon...
 
Joined
Apr 5, 2008
Messages
2,847
Location
Middletown, NY
I've been testing my lenses with my new D7000 to set the AF fine tune.

What I have found with my 18-200 is that it requires these settings:

+2 @ 18mm
-2 @ 50mm
-8 @ 200mm

So if I left the AF fine tune at 0, it would begin backfocussing as I zoomed out towards 200mm.

The reputation of this lens is that it is sharp enough until about 135mm, after which it becomes soft.

My findings suggest that this softness may be due to AF, not optics.

In my case, most of my shots are between 18-60mm, so I'll probably set it at zero, unless I need the telephoto (e.g. trip to the zoo), in which case I'll bump it up to -8 at this time.

Thanks be to AF micro adjust - darn useful tool for squeezing more life out of a lens I reckon...

Could very well be correct about your assumption. Professional reviews do not say whether or not the long range of a zoom is soft due to focus inaccuracy or not. But I agree, no reason why this could certainly be a factor.

Are you testing at different distances for all focal lengths? RE: At 200mm you should be testing focus accuracy at about 30 feet....and shorter as you zoom less. With a slow lens like this, you would hope that it would catch the zone of accurate focus at all zoom ranges.

I have a 55-200 that my focus tests showed a similar pattern as yours....front or back focusing at the short end of the zoom range, and the opposite at the other. And of cause smack on somewhere in-between. I sent it to both Nikon and a Nikon authorized repair center. Both reported the lens within factory tolerances. Upon return, I tested the lens and it acted the same way in my focus tests. However, when shooting in normal everyday scenes and forgetting about focus testing, I had nothing to complain about....go and figure.

Mike
 
Last edited by a moderator:
W

Wileec

Guest
It's my opinion that AF Fine Tune is basically useless for zooms, but I really believe it can be an asset, if needed, with primes. This is especially true of really fast (f/1.4) or long ones, where the DOF can be very narrow, so it's important that what is in focus be exactly what you intend to be in focus. For me, AF Fine Tune, when needed, makes that possible.

This said, I made a trip recently and shot with the D7000, 400mm f/2.8 and 1.4x or 1.7x TCs and got many great and very sharp shots without AF Fine for the combos. I had done some quick testing to affirm things were very close, so when I have the time, I'll tap into this for the new camera, but this confirmed what I suspected - good technique and decent samples can provide very acceptable results. Here is one example:

D7K0504_YNP_Red-Foxcarcass_web.jpg
Subscribe to see EXIF info for this image (if available)
 
Joined
Nov 29, 2009
Messages
151
Location
Slovenia
Getting AF fine tune juuuuuust right is a PITA even with fast primes. I can imagine what your eyes looked like after you managed to nail down the needed AF adjustment @f5.6.

Or does 200mm really help that much? My longest lens is 135mm and I just can't imagine seeing a difference between 0 and -8 @5.6. Seriously, how do you do it? So far the best method I've found is comparing shots with different AF adjustment with a shot taken in live view.
 
W

Wileec

Guest
Using a tool like the LensAlign is key to evaluating a combo - to see if it's needed or not - and for use to try settings. The key is shooting wide open and at a "reasonable distance". LensAlign provides an online calculator for determining a decent distance.

Without question, as you dial down the aperture, the benefits of AF Fine Tune are diminished, but not negated. The more wide open, or the greater the distance you shoot at, the more exacting intended point of focus and the narrower DOF one is dealing with. For example, if I'm shooting a black bear early in the morning, I may need to shoot as wide open as possible to try to get shutter speed as high as possible - at these distances and wide apertures DOF may be a few inches. If no adjustment leaves a combo forward or back focusing even three or four inches at 100-200 feet, that's the difference between eyes being in focus and not - if not - wasted shot. No good to have nose or ears in focus if the eyes aren't in focus.

Using the LensAlign makes it very possible to distinguish between a couple of adjustment units. I typically start in tens, then work my way down as I determine what seems to be working best. Then, once I've settled upon a setting, I try my target at different angles AND defocus and refocus for each shot. If that pans out, then I try shooting at different distances to see if it's still giving me optimum results. It takes some time to accomplish this, to be sure. And, I shoot from tripod with a remote release - but once a combo is dialed in, when one gets a poor shot, 99% of the time it's on technique and/or user choices for the lens/camera. For me, I do my best to take gear out of the equation, so that I can work on the most important part of the equation - me. ;-)
 
W

Wileec

Guest
Well, yes - I rushed my reply and misspoke - the greater the focal length (same distance, same aperture), the narrower the DOF. So 400mm is narrow, 400 + 1.4x or 1.7x is narrower yet - making correct point of focus distance being very critical. With this FL, it's easy for the focus to be a few inches off and that has a direct impact on keeper rate.
 
Joined
Aug 25, 2009
Messages
4,043
Location
Missouri
I wasn't under the impression that AFFT would produce great results throughout the entire FL with zooms. Then again, I don't use zooms, so I might not be the one to ask.

Thankfully there is a cure for your 18-200 woes, and it's called the 70-200 ƒ2.8 VRII :cool:.
 
W

Wileec

Guest
I wasn't under the impression that AFFT would produce great results throughout the entire FL with zooms. Then again, I don't use zooms, so I might not be the one to ask.

Thankfully there is a cure for your 18-200 woes, and it's called the 70-200 ƒ2.8 VRII :cool:.

You're correct - I've checked a couple of zooms and they tend to need different things at either end - or nothing at all. I had a Tamron 200-500mm lens and I tended to use it at 500mm most of the time, so I checked it at that FL and it needed nothing at all, wide open, so it made it that much easier to use. I didn't check it at the low end, since at that end, I typically would use the 70-200mm f/2.8, as it was a far better lens. But, we use what we can, until we can afford better . . .
 
Joined
Nov 23, 2008
Messages
117
Location
Australia
Getting AF fine tune juuuuuust right is a PITA even with fast primes. I can imagine what your eyes looked like after you managed to nail down the needed AF adjustment @f5.6.

Or does 200mm really help that much? My longest lens is 135mm and I just can't imagine seeing a difference between 0 and -8 @5.6. Seriously, how do you do it? So far the best method I've found is comparing shots with different AF adjustment with a shot taken in live view.

Yep - major eye strain headache!

I was also checking to make sure my replacement D7000 wasn't backfocussing after a problem with my 1st copy, so I wanted to be thorough within my limited means & technical understanding.

I used the Liveview vs Viewfinder method initially. This confirmed whether or not I had a problem. Then, I took a series of shots of a flat plane with lots of contrasty detail at different AF fine tune settings (e.g. Liveview, -8, -6, -4, 0, +2...+8).

The problem with this though, is that you don't know where a pic sits in the DOF - the pic may be sharp but right at the edge of the DOF, so there is little margin for real-world error.

So I repeated the process using a focus chart printout which allowed me to see the full extent of the DOF. By now I had a better feel for the AF fine tune 'zone' that each lens needed, so I my test range was narrower.

I wasn't always happy with my testing tho & ended up doing three different sets of focus chart tests.

In the end, I lined-up the results from all my tests in a table & estimated a final AF fine tune setting per lens. There was enough consistency across the tests to give me a reasonable level of confidence in the final setting.

For my zooms (Nikon 18-200, Tammy 10-24, Tammy 17-50) I tested at the long & short ends. And also at 50mm for the 18-200. Only the 18-200 showed significant variation across the range.

With a wide DOF (e.g. at 200mm & f5.6) the focus chart was the most useful test because the full DOF is visible. In this case I tried to identify the setting that aligned best to the liveview pic by looking for the corresponding spots above & below the focus point where the image began to lose focus.

Who knows how much difference all this will make in the real world, but it is interesting that the AF adjust started to creep fwds at the lengths that the 18-200 is reputed to be soft.

Hope this makes sense...
 
W

Wileec

Guest
I'd suggest the key to this kind of testing is that there be a target that is parallel with the camera, to set a clear focus point, but the evaluation area really needs to be at an angle and lined up with the lens. The angle is key because what is being adjusted is the distance from the camera. We set focus on one spot and either the camera actually focuses on that point or it doesn't. I think that's why the LensAlign is used and/or been copied because it provides a tangible, relatively easy to read, evaluation of what is happening. No eye strain is needed and using LiveView and manual AF is contrary to what AF Fine Tune is about. The key is to understand how the camera is seeing and calculating the distance. So, it's important to use the AF, then consider the results. I spent many many hours pursuing all kinds of ways to evaluate this - with very mixed results, until I got the LensAlign. With LA, one has a tool that allows for repeatable and measurable results - I didn't find that with any other process I read about, or came up with on my own.

NOTE: I'm not related to anyone, or the company behind LensAlign - only mention it as I found it to be the simplest and most time efficient way to check for the need for AF Fine Tune - and if so, a way to help me determine what that setting should be for a given combination.
 

Latest posts

Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Nikon Cafe is a fan site and not associated with Nikon Corporation.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Forum GIFs powered by GIPHY: https://giphy.com/
Copyright © Amin Forums, LLC
Top Bottom