Sigma 18-35mm infinity?

[Stagecoach]

In the second quote you suggest a relation between power and positioning of the focus motor. There is not something like that.

If you read the second quote that you refer to you will understand what I have said there is specific to the Sigma Fine Tune Docking System and is a simple explanation that the Sigma System 'Modifies' what has been sent out/or returns from the camera PDAF control system (which is done within the lens by software with the lens off the camera).

It is not intended as an accurate technical description of what type of current/voltage signals are sent/received

[george013]

If you read the second quote that you refer to you will understand what I have said there is specific to the Sigma Fine Tune Docking System and is a simple explanation that the Sigma System 'Modifies' what has been sent out/or returns from the camera PDAF control system (which is done within the lens by software with the lens off the camera).

It is not intended as an accurate technical description of what type of current/voltage signals are sent/received

The communication between camera and lens is 1 way. The camera doesn't send a proportional signal but a yes signal. like a light switch, without a dimmer of course . There's nothing to modify.
Let's stop it.

George

[Stagecoach]

The camera doesn't send a proportional signal but a yes signal.

Not quite correct George.

There are PDAF systems that send out a 'proportional' signal to the lens motor based upon the estimated distance (and of course direction) calculated by the phase differences at the PDAF sensor.

This opposed to some systems that send out a continuous signal to the motor to move which is then 'stopped' when the PDAF senses that it has reached an 'in-focus' position.

There's nothing to modify.

Not sure what you are referring to here but, with regard to 'modifying' what do you think a Fine Tune System such as the Sigma one is doing then?

[george013]

Not quite correct George.

There are PDAF systems that send out a 'proportional' signal to the lens motor based upon the estimated distance (and of course direction) calculated by the phase differences at the PDAF sensor.

Now it's my time to ask where do you get this from.

Not sure what you are referring to here but, with regard to 'modifying' what do you think a Fine Tune System such as the Sigma one is doing then?

This is still a question that is not answered yet.

George

[Stagecoach]

Now it's my time to ask where do you get this from.

https://www.lensrentals.com/blog/201...s-often-works/

"The camera reads the phase detection sensor, looks up a huge data array programmed in its chips that describes the properties of all of the manufacturer’s lenses, does some calculations and tells the lens something like “Move your autofocus this much toward infinity”.
The lens contains a sensor and chips that either measures the amount of current applied to the focusing motor, or actually measures how far the focusing element has moved, and sends a signal to the camera, saying: “almost there”.
The camera rechecks the phase detection, may send some fine-tuning signal to the lens, may even recheck a 3rd or 4th time until perfect focus is indicated. If things don’t work as planned, the infamous “hunting” may occur, but usually not.
Once the camera has confirmed focus, it tells the lens not to move anymore and then sends that little ‘beep’ and light signal we all know and love, and we push the shutter.
The whole process takes a tiny fraction of a second. It’s fast."

This is just one of a few references I have found over time that refer to the PDAF measuring the distance the lens needs to move. I suspect this approach is more modern and high end.

This is still a question that is not answered yet.

The exact method the Sigma FT unit uses to achieve this is possibly only known to them, but it is 'modifying' what the camera has sent out.

[george013]

I did read it. It's mentioning how autofocus often works. In the example with phase detection it doesn't mention for what brand this is. I still have my doubts.
All new Nikon pro camera's have an build-in focus motor. AF lenses have no internal motor, AF-s do. D lenses have a chip that provides info of the distance but is mostly meant for the flash.
I whish the manufacturers did give more info.

George

[Stagecoach]

I did read it. It's mentioning how autofocus often works. In the example with phase detection it doesn't mention for what brand this is. I still have my doubts.

Well if you have doubts George it's best to stop making definitive statements such as these in any discussion;

The communication between camera and lens is 1 way. The camera doesn't send a proportional signal but a yes signal. like a light switch, without a dimmer of course . There's nothing to modify.
George

[xpatUSA]

I did read it. It's mentioning how autofocus often works. In the example with phase detection it doesn't mention for what brand this is. I still have my doubts.
All new Nikon pro camera's have an build-in focus motor. AF lenses have no internal motor, AF-s do. D lenses have a chip that provides info of the distance but is mostly meant for the flash.
I whish the manufacturers did give more info.

George

Casting doubt on the words of Roger Cicala could lower one's credibility, George...

[george013]

I just try to read what he writes. If somebody writes

... looks up huge data array programmed in its chips that describes the properties of all of the manufacturer’s lenses...

then I know that's a very simplification, just like it is used to explain matrix metering. It also would mean that new lenses won't work with that camera.
What I read is that there's first a rough correction on the lens position. And then the fine tuning. And that fine tuning can only be a 1-way command. The judgement is in the camera.
I calculated a normal situation. A portrait with a 85mm lens from a distance of 3m. The difference between a eye and the nose I did put at 1cm. The difference in image distance in the camera will be 0.0085mm. Looks to me impossible to gain that accuracy with a calculated signal.

Some exif of camera/lens combination give info of the distance. Faaaaaar beside what it was.
I mentioned before, extension rings, how can we deal with them.
Converters the same. I believe Nikon converters do correct the focal length in the exif. But not all. And they work to.

I still believe the captain is in the camera. Some info of the lens may speed up focusing, but the final result is reached in the camera. Leaves open the question what is done when calibrating the lens in the camera. And what's done with the Sigma lens.

I wish Nikon and others gave some more info about it.

George

[Stagecoach]

Ok George, consider these points ...........................

I just try to read what he writes. If somebody writes

""... looks up huge data array programmed in its chips that describes the properties of all of the manufacturer’s lenses...""

then I know that's a very simplification, just like it is used to explain matrix metering.

Of course it's a simplification, I read into it that this means data in either analogue or digital terms such as ;

Manufacturer Tamron - uses Near focus to infinity travel = 0v to +6v, = 150,000 pulses, etc
Manufacturer Sigma - uses Near focus to infinity travel = 0v to +4v, = 180,000 pulses, etc
e.t.c, e.t.c

It also would mean that new lenses won't work with that camera.

I'm not sure what you mean as "that" camera? We are talking 'generalisations' here as with his article and many you find on the web, not a specific model.

I would guess the situation is that the latest and most modern Nikon lens will send 'certain' data back to my old D300 that it would not read or use but it will still work. If I were to put that lens on say a D850 it would use that 'certain' data to make the lens work better/faster/smoother? We then have the situation also where some cameras and lenses need to be re-chipped.

What I read is that there's first a rough correction on the lens position. And then the fine tuning. And that fine tuning can only be a 1-way command.

Don't forget he is talking about certain cameras, not all, and that there is serial data flowing from camera to lens and lens to camera.

The judgement is in the camera.

If you are referring to what decides if focus has been achieved or not the decision is made by the PDAF as far as I would assume.

I calculated a normal situation. A portrait with a 85mm lens from a distance of 3m. The difference between a eye and the nose I did put at 1cm. The difference in image distance in the camera will be 0.0085mm. Looks to me impossible to gain that accuracy with a calculated signal.

I have spent a career working on equipment some of which was moved around on the ocean with divers hanging below to an accuracy of a few metres by a satellite thousands of miles away, that accuracy was all achieved by 'calculated signals'. The signals being a combination of both analogue and digital.

Some exif of camera/lens combination give info of the distance. Faaaaaar beside what it was.

I don't understand this.

I mentioned before, extension rings, how can we deal with them.

In the same way as we deal with a lens that is too close to the subject. The PDAF can not determine the correct 'direction' so it moves focus in one way or the other until it finds it.

Converters the same. I believe Nikon converters do correct the focal length in the exif. But not all. And they work to.

See above.

I still believe the captain is in the camera. Some info of the lens may speed up focusing, but the final result is reached in the camera.

I think it's pretty obvious that it's the PDAF in the camera that makes the decision 'What I have measured is 'good enough' so I will now put that little green light on and allow the shutter to be actuated'

Leaves open the question what is done when calibrating the lens in the camera. And what's done with the Sigma lens.
I wish Nikon and others gave some more info about it.

You will have to keep searching, if you want to know exactly what is modified to move the optics to a different position.

You could also ponder over this total 'guess';

Camera logic :
PDAF module gives signal to say it has reached acceptable focus. (Note this will be somewhere within the dead band.)
Signal read by CPU and it looks to see if a 'correction' value has been input. (Fine Tune Saved Value).
If a correction value is stored e.g. +5, lens is instructed to move 'x' pulses clockwise. (- = anticlockwise)
CPU checks to see if PDAF still measuring that the result is within it's deadband. (it could possibly say ignore if not)
CPU gives signal that shutter can be activated.

[george013]

Ok George, consider these points ...........................



Of course it's a simplification, I read into it that this means data in either analogue or digital terms such as ;

Manufacturer Tamron - uses Near focus to infinity travel = 0v to +6v, = 150,000 pulses, etc
Manufacturer Sigma - uses Near focus to infinity travel = 0v to +4v, = 180,000 pulses, etc
e.t.c, e.t.c



I'm not sure what you mean as "that" camera? We are talking 'generalisations' here as with his article and many you find on the web, not a specific model.

I would guess the situation is that the latest and most modern Nikon lens will send 'certain' data back to my old D300 that it would not read or use but it will still work. If I were to put that lens on say a D850 it would use that 'certain' data to make the lens work better/faster/smoother? We then have the situation also where some cameras and lenses need to be re-chipped.




Don't forget he is talking about certain cameras, not all, and that there is serial data flowing from camera to lens and lens to camera.



If you are referring to what decides if focus has been achieved or not the decision is made by the PDAF as far as I would assume.



I have spent a career working on equipment some of which was moved around on the ocean with divers hanging below to an accuracy of a few metres by a satellite thousands of miles away, that accuracy was all achieved by 'calculated signals'. The signals being a combination of both analogue and digital.



I don't understand this.



In the same way as we deal with a lens that is too close to the subject. The PDAF can not determine the correct 'direction' so it moves focus in one way or the other until it finds it.



See above.



I think it's pretty obvious that it's the PDAF in the camera that makes the decision 'What I have measured is 'good enough' so I will now put that little green light on and allow the shutter to be actuated'



You will have to keep searching, if you want to know exactly what is modified to move the optics to a different position.

You could also ponder over this total 'guess';

Camera logic :
PDAF module gives signal to say it has reached acceptable focus. (Note this will be somewhere within the dead band.)
Signal read by CPU and it looks to see if a 'correction' value has been input. (Fine Tune Saved Value).
If a correction value is stored e.g. +5, lens is instructed to move 'x' pulses clockwise. (- = anticlockwise)
CPU checks to see if PDAF still measuring that the result is within it's deadband. (it could possibly say ignore if not)
CPU gives signal that shutter can be activated.

It looks we agree. The AF system in the camera is the captain, it sends a command, checks the result, send again etc. Apart from a speedy movement at the start to save time. You may correct me but if a pulse is equal with a step, than only the amount of commands is changing before a satisfied focus is reached.

George

[Stagecoach]

It looks we agree.

I really don't know what it is that you are "agreeing" with George, but that's beside the point.

If you consider that I would agree with this statement of yours ...........................

The AF system in the camera is the captain, it sends a command, checks the result, send again etc.

........ I would suggest that very much depends upon camera/lens manufacturers, age of technology, models and combinations. You can't make a specific statement like that without knowing exactly what is happening and specifying camera/lens combination.

Your suggestion regarding the AF system in the camera being 'Captain' depends very much what you consider the function is that the 'Captain' is controlling.

If you are talking about the 'Fine Tune Correction' function, which is unclear, you would need to consider two scenarios;

a) Taking my D300 and a Nikon Lens as example I would suggest it's fairly certain the 'command' to 'Fine Tune' the lens position by a certain amount comes from the Camera. It is the camera where these 'Fine Tune' values are held, what that 'command' and its value 'modify' can only be speculated.

b) Taking my D300 again and using a Sigma Art lens on it which has the ability for the user to store 'Fine Tune' data/values (on firmware within the lens body) and modify signals from the camera is a different scenario.

'Fine Tune' correction data/values are then stored in two separate places, the camera and the lens.

It could be that the camera reads both of these OR it could be that the camera only reads the one in camera and then when the lens receives a signal (derived from the PDAF) that all is ready to fire off a shot the lens reads the 'Fine Tune' correction data/values within the lens chip and makes these adjustments prior to aperture control.

So I would suggest things are not straightforward regarding "Captains" or commands with respect to 'AF Fine Tuning' with certain camera/lens combinations.

Apart from a speedy movement at the start to save time.

I have no idea if this comment is with respect to "Captains" or the following

You may correct me but if a pulse is equal with a step, than only the amount of commands is changing before a satisfied focus is reached.

This comment makes no sense to me and I have also not got a clue what you mean by the term "satisfied focus". Is "satisfied focus" the optics position (determining sensor image) before or after any movement caused by a 'Fine Tune' optics position correction?

[george013]

Let'sgo to the beginning of our discussion.
Part 1.
Post 19.
AF fine tuning doesn't correct sensor misplacement. If the sensor was misplaced any lens would produce a fault.
I corrected myself about that other way the light has to travel. That's also concerns any other lens.
The above would use a camera calibration.
It still leaves me with the question WHAT is calibrated when doing lens calibrating in the camera.
sytem is needed.

Part 2.
The camera doesn't focus with a command to the lens like: go to that position and than we're finished. It's a try and check game. Your link to lensrentals adds a quick way to come somewhere in the neighborhood to start with only. With try/check it's not that much important what the lens itself is doing, except for moving in the right direction. Can be done with a uninterrupted or a interrupted signal.

Everybody can see we both have no idea how that communication works. Knowing more would give us the answer.

From Nikon. https://www.nikonimgsupport.com/ni/N...d=1&lang=en_US
It's always about how to fine tune, not why but this part says a lot.

Service note: If your lens has a focus problem you should return it to Nikon Service as AF Fine-Tune is not intended to solve optical problems which will generally be outside of scope for this tool. If you have a focus problem which is always present with different lenses then this would indicate a camera setting issue or that the camera has received impact damage which is causing general defocus problems. Again return it to service.

George

[Stagecoach]

Let'sgo to the beginning of our discussion.

Part 1.
Post 19.
AF fine tuning doesn't correct sensor misplacement. If the sensor was misplaced any lens would produce a fault.
I corrected myself about that other way the light has to travel. That's also concerns any other lens.
The above would use a camera calibration.
It still leaves me with the question WHAT is calibrated when doing lens calibrating in the camera.
sytem is needed.

AF Fine tuning will correct for sensor misplacement, and by misplacement I refer to minor tolerance variations of it's placement along the axis as long as it's perpendicular to the axis. This in exactly the same way as AF Fine Tune will correct for other minor mechanical tolerance errors as long as they are within the span of 'tuning' allowed.

I think it's pretty obvious that any positional tolerance errors within the camera, e.g. sensors, mirrors and flange are going to be there whatever lens is fitted.

What are you referring to as "camera calibration" as opposed to user available 'AF Fine Tuning' and 'AF Default Tuning'?

Part 2.
The camera doesn't focus with a command to the lens like: go to that position and than we're finished. It's a try and check game. Your link to lensrentals adds a quick way to come somewhere in the neighborhood to start with only. With try/check it's not that much important what the lens itself is doing, except for moving in the right direction. Can be done with a uninterrupted or a interrupted signal.

Not sure why you are quoting the obvious, neither I nor Roger Cicala in his article suggest that

Everybody can see we both have no idea how that communication works. Knowing more would give us the answer.

That's why I'm always careful to not make definitive statements without being sure I can support them adequately otherwise I use the word "could" .

From Nikon. https://www.nikonimgsupport.com/ni/N...d=1&lang=en_US
It's always about how to fine tune, not why but this part says a lot.

""Service note: If your lens has a focus problem you should return it to Nikon Service as AF Fine-Tune is not intended to solve optical problems which will generally be outside of scope for this tool. If you have a focus problem which is always present with different lenses then this would indicate a camera setting issue or that the camera has received impact damage which is causing general defocus problems. Again return it to service.""

I don't think it "says a lot" other than if you can't correct or don't want to correct the camera errors with the 'Default' fine tune input (that interestingly is twice the span of the individual lens input options) send it back.

My washing machine and car manual have similar jargon in them

[Manfred M]

This thread is going in circles. Time to close it down.