Autofocus issues

[Saorsa]

Another consideration might be that your lenses are D types which rely on the camera's focus motor to achieve focus. The accuracy of the mechanical linkage may not be equal to the precision of the focus module when compared to lenses with an internal focus motor.

[Manfred M]

A D-lens is a lens with a cpu in it that gives information about the Distance

No; not quite the way that this works. It is a lens with electrical connections so that the camera body and mechanisms on board the lens can communicate with each other. The contacts also provide power to the lens components are are actively powered. The actual algorithms that control the lens actions are run on the camera's CPU. This includes aperture actions, autofocus and image stabilization. There will be control circuitry in the lens to manage the autofocus motor and the in-less stabilization mechanism.

The G and E series are much the same way, but the technology is evolving. The main two differences between the D and G series is that the D series has no built in autofocus motor and relies on the camera's built in motor, rather than one in the lens used in the G series and newer. The other main change is that the G series of lenses lost the traditional aperture ring, so these lenses are not backwards compatible with the older camera bodies. To use the command wheel on the camera body, the D series lenses have to have the aperture set and locked to the maximum f-stop value, so the aperture ring is largely redundant moving forward.

[george013]

No; not quite the way that this works. It is a lens with electrical connections so that the camera body and mechanisms on board the lens can communicate with each other. The contacts also provide power to the lens components are are actively powered. The actual algorithms that control the lens actions are run on the camera's CPU. This includes aperture actions, autofocus and image stabilization. There will be control circuitry in the lens to manage the autofocus motor and the in-less stabilization mechanism.

The G and E series are much the same way, but the technology is evolving. The main two differences between the D and G series is that the D series has no built in autofocus motor and relies on the camera's built in motor, rather than one in the lens used in the G series and newer. The other main change is that the G series of lenses lost the traditional aperture ring, so these lenses are not backwards compatible with the older camera bodies. To use the command wheel on the camera body, the D series lenses have to have the aperture set and locked to the maximum f-stop value, so the aperture ring is largely redundant moving forward.

Look at http://www.photosynthesis.co.nz/nikon/serialno.html. You'll find D-lenses being AF and AF-S.
On the same website http://www.photosynthesis.co.nz/nikon/lenstype.html.
Look at https://www.nikonimgsupport.com/ni/N...d=1&lang=en_US. They forgot the AF lenses, lenses without build-in motor. That are lenses that use the camera motor. But the D-lens is explained.
Look at https://photographylife.com/nikon-lens-nomenclature. It's mentioning the AF-D but it should be AF D. That distance chip is independent from the AF.
One more https://www.nikonusa.com/en/learn-an...your-dslr.html

I own the AF-S 28-70 2.8 D ED and the AF-S 80-200 2.8 D ED. Both D-lenses with build-in motor.

But for Marco this isn't important. He has a question just for this camera and lens. His lens happens to be a AF lens, so relying on the camera motor.

George

[Stagecoach]

If the best out of 3 is shown, that means the result might be even worse.
If you want to correct AF for the close distance and wide aperture, the focus shift, then you might introduce problems with longer distances and/or smaller aperture.
George

That's obvious George and the problem you come about with tuning zoom lenses.

But, you have to start somewhere and practice would suggest check one end, record it, then check the other end and record it. Consider results and actions and at the same time consider if you have a lens/camera that can not be set to produce expected good enough results and send for repair.

[george013]

That's obvious George and the problem you come about with tuning zoom lenses.

But, you have to start somewhere and practice would suggest check one end, record it, then check the other end and record it. Consider results and actions and at the same time consider if you have a lens/camera that can not be set to produce expected good enough results and send for repair.

I still put my money on focus shifting. If you try to correct normal but fault behaviour of a spherical lens then you introduce more problems as then you solve.
With fine tuning you don't correct the lens but the control system in the camera.

I wonder what Marco is thinking.

George

[Stagecoach]

I still put my money on focus shifting.

I suspect you are putting your money on 'focus shifting' based more upon what you have read on the internet George, not based on factual diagnosis using good test procedure and protocol of Marco's specific lens on his specific camera body.

Every article I have read suggests 'focus shift' should not be noticeable in an image (and that does not mean not present) with a f/2.8 upwards lens as it will be masked by increasing DoF. So the reality of what I'm questioning is, are the errors Marco is reporting due to a) 'focus shift' OR b) 'focus shift' being noticeable 'because' the AF focus at f/2.8 is wrong.

OR

You may have put your money on 'focus shift' based on this comment of Marco's in post 46

I really am very grateful you stopped by, I put the camera in LV, focused with contrast detection at f2.8 and then changed my aperture to f5 and the focus shifted as suspected! At least now I know I'm not crazy or something

................. for which I would wonder is that a result that is representative of results achieved throughout the entire range of FLs and SDs using the same method?

If you try to correct normal but fault behaviour of a spherical lens then you introduce more problems as then you solve.

Another theory based upon what you have read, not a diagnosis based upon good test procedure and protocol of Marco's specific lens on his specific camera body to see what's actually happening.

With fine tuning you don't correct the lens but the control system in the camera.

Correct, with the addition it is applicable to that particular lens and camera body combination when the 'Lens input' adjustment is used.

Out of interest how would you go about 'confirming' to what degree a lens/body combination image result is being affected by 'focus shift' using the example of 70mm 2.8 lens, 3.5m SD between f/2.8 and f/5.6?

[george013]

Graham,

You know one should have a theory before he can examine it.

Focus shift is explained by the fact that the rays of a spherical lens don't converge to just 1 point. In a drawing it's shown as a point with 2 dimensions: the minimal and maximal distance to the lens. From the drawings and theory the extension of that "point" is moving to the lens. Phase AF is choosing a focal plane to close to the lens. If the image distance is decreasing, then the subject distance is increasing. The real world focus is laying further away from the lens. I always mix up if that's called back or front focus.
And how much is the difference? That depends on the differences in image distance. I've no idea how much that is. In the drawings showing the explanation of the fault in a spherical lens, the subject plane is always in infinity, the image distance is the focal distance. In this example the subject distance is close, so the image distance is increasing and probably the fault of the rays too.
That looks to the problem Marco describes.

From the specs I think that are for his lens I can't see anything of aspherical or ED glass. The lens was produced since 1992. In 1993 only the D-chip was upgraded. http://www.photosynthesis.co.nz/nikon/serialno.html

If one wants to compare this behaviour with his lens, be at least sure that lens doesn't have aspherical glass.

Enough justification of my theory?

And it's just a theory. It doesn't have to be the explanation.

Originally Posted by george013 View Post
If you try to correct normal but fault behaviour of a spherical lens then you introduce more problems as then you solve.

Another theory based upon what you have read, not a diagnosis based upon good test procedure and protocol of Marco's specific lens on his specific camera body to see what's actually happening.

That seems logical. If you try to correct focus shift with a wide open aperture, then you introduce focus faults with smaller apertures. You correct the camera system, not the lens system. Read it in the context I wrote it.

George

[Stagecoach]

Enough justification of my theory?

And it's just a theory. It doesn't have to be the explanation.
George

Exactly George, it's just a theory.

For a flagship lens made by Nikon with so many raving reviews regarding its performance, even on digital, to me it seems unusual that such basics as an inability to AF on the eyes acceptably in portrait work has not been mentioned in what I have read. This especially when so many talk about its great sharpness.

If you try to correct focus shift with a wide open aperture, then you introduce focus faults with smaller apertures.

That may well be the case, but in the first instance you need to correct (or check/confirm) for the AF accuracy wide open.

[george013]

Exactly George, it's just a theory.

For a flagship lens made by Nikon with so many raving reviews regarding its performance, even on digital, to me it seems unusual that such basics as an inability to AF on the eyes acceptably in portrait work has not been mentioned in what I have read. This especially when so many talk about its great sharpness.

A design nearly 30 year old.

Without a theory you don't come anywhere. Mine is build upon the link Dem gave. The only advise I can give you is to read more. And wonder why and when the aspherical glass has been introduced.
My lens, the AF-S 28-70 2.8 D ED was produced from 1999 till 2007 and does have the addition ED on it.
Let's quit.

George

[Stagecoach]

A design nearly 30 year old.

Without a theory you don't come anywhere. Mine is build upon the link Dem gave. The only advise I can give you is to read more. And wonder why and when the aspherical glass has been introduced.
My lens, the AF-S 28-70 2.8 D ED was produced from 1999 till 2007 and does have the addition ED on it.
Let's quit.

George

George,

So from that I can assume you disagree with my theory that inadequate testing may have been undertaken to confirm if 'correct' AF tuning will/can mask or minimise the affect of any 'focus shift'.

Only a yes or no answer is required

[Manfred M]

Time to close down this thread:

We do not have the knowledge or expertise to diagnose this problem without access to the equipment and to do some controlled testing.