Why do some lenses have front/back focus issues?

Pentaxian 450 has explained this to you above in his second post.



How are my statements 'rash' ? They are based upon taking over 80,000 pictures in the real world with the cameras I have mentioned

Anyone can embed a link. You can find links on the Web to justify any kind of propaganda you want to encourage. It means nothing. I go on real world observations.

However as you seem a 'link' person here's one

http://graphics.stanford.edu/courses/cs178/applets/autofocusPD.html

And here's how Contrast AF works ...

http://graphics.stanford.edu/courses/cs178/applets/autofocusCD.html

which as said requires hunting by default.


None of which "why we need to adjust focus on a per lens basis" though.

I simply wanted something Objective, rather than your Subjective findings

From your link :

It can therefore instruct the lens to move directly to the correct position, without taking a second measurement or "hunting" around.

My findings were objective not subjective.


You forgot to add .... "In practice it usually moves close to the correct position, then takes additional measurements to refine this position. "

.... which comes right afterwards. And correlates with my objective findings.

You originally presented the claim as fact, when in fact it is simply a theory. In practise things are actually different. You also presented this theory to explain away lens front and back focussing when it has nothing to do with it and even after pentaxian450 explained that inacuraccies between the light path to the sensor and path to the AF sensor can explain FF and BF


Do you actually have a camera?

It doesn't exactly works like that. Since some lens have a longer focus throw than others, and since some lens screwdrive don't all use the same "drive ratio", the AF takes an "educated guess" to send the lens close to where the focus point should be for the AVERAGE lens, then it refines the focus in smaller steps.

But, if the camera's AF sensor is not perfectly located, or if the secondary mirror is not at the perfect angle to send the right part of the subject on the AF sensor, you still end up with a front/back focus problem.

[quote:3496ace15f="Smeggypants"
Do you actually have a camera?[/quote]Several more than You ! I even make a living from it


I just like to know how they ACTUALLY work, NOT how someone THINKS they work !

That I wouldn't disagree with

The question is : does it check it's in focus at the completion of the final move ? or does it just assume it's in the right place ?? having calculated where it should be.

This is the bit I'm having trouble researching

It's measuring a phase difference isn't it? When it finds the cancel-out point as received by the AF sensors it stops moving the lens. If the lens carries on a bit because of play in the mechanism then focus will be slightly out and could be corrected for.

However, this may vary depending upon how far the mechanism has travelled. Something whizzing from infinity to macro distances, for example, whould take more stopping than something whizzing from 1.2m to 1.0m

Then again, it may also vary depending on direction of travel. Going from macro to infinity will certainly be stopped pretty quickly when it hits the limit of travel.

Lots of variables and potential for error in an AF mechanism I suspect. However, focusing tests have shown that manual focus is also very unreliable, basically because the eye adapts and can be poor at judging the sharpest point. Try it. Focus on something, note the position of the lens. Focus again and note how different your second and subsequent choices are.

Fortunately despite all this getting normal, sharp images still seems to happen most of the time.

Thank you for your Polite answer John

If it was a closed-loop servo system then then that would answer my question, but as I (and many others) suspect it isn't ! It simply moves to 'A' position ! (right or wrong)