How can lens focus ever be "off" in a way that AFMA fixes it?

[hgraf]

Here's a question that's been bugging me.

Why does AFMA fix anything? As I understand modern phase detect focus systems, focus is a closed loop sort of thing The camera tells the lens to change it's focus, and when the camera body sees things are in focus (the phase detect sensors say focus has been achieved) it stops moving the focus on the lens.

How can this "go wrong" in a way that AFMA can fix it? The way I see it is the only way things can go wrong is if the sensor is physically in the wrong spot, meaning an "in focus" in the phase detect sensor doesn't correspond to in focus at the sensor.

Where the lens is shouldn't matter?

So what am I missing?

Thanks!

 

[Loswr]

I believe the lens plays a role - I'm not sure it's really a closed loop with the AF sensor. Roger Cicala's data showing more accurate focusing with the more recent lenses/bodies was supported by the finding of rotational encoders on the USM lens motors. So, with older lenses (pre-2000) we had a basically open loop where the AF sensor determined magnitude and direction of the move and that was transmitted to the lens (look-move) - if the motor moved a ratio slightly off 1:1 from the instructions, AFMA would compensate. The newer lenses+bodies apparently have a closed loop where the encoder reports movement (look-move-confirm). But there may be tolerances in the encoder (e.g. detecting movement as other than a 1:1 ratio) for which AFMA could correct. I'm not positive the loop is closed with the AF sensor, i.e., look-move-confirm-look.

Certainly, a misalignment of the AF sensor with the image sensor is one factor that AFMA corrects. I can personally attest to that - at one point, I dropped my 5DII to the pavement. The camera was perfectly functional, but the sensor alignment changed such that all of my AFMA values (for ~8 lenses) shifted 10 units negative relative to the initial values.

 

[Rienzphotoz]

I believe the lens plays a role - I'm not sure it's really a closed loop with the AF sensor. Roger Cicala's data showing more accurate focusing with the more recent lenses/bodies was supported by the finding of rotational encoders on the USM lens motors. So, with older lenses (pre-2000) we had a basically open loop where the AF sensor determined magnitude and direction of the move and that was transmitted to the lens (look-move) - if the motor moved a ratio slightly off 1:1 from the instructions, AFMA would compensate. The newer lenses+bodies apparently have a closed loop where the encoder reports movement (look-move-confirm). But there may be tolerances in the encoder (e.g. detecting movement as other than a 1:1 ratio) for which AFMA could correct. I'm not positive the loop is closed with the AF sensor, i.e., look-move-confirm-look.

Certainly, a misalignment of the AF sensor with the image sensor is one factor that AFMA corrects. I can personally attest to that - at one point, I dropped my 5DII to the pavement. The camera was perfectly functional, but the sensor alignment changed such that all of my AFMA values (for ~8 lenses) shifted 10 units negative relative to the initial values.

Interesting.
I find AFMA a bit difficult to do ... that's just me being lazy I guess ... for people like me fixing the AFMA values, by dropping the camera, sounds a bit more easier ;D

 

[K3nt]

On my 7D, in the very beginning, I was wondering why some of my shots seemed to miss focus even if the camera told me it has locked on. It was best seen shooting wide open, f/2.8 or larger. Suddenly the image would be in focus slightly in front or slightly behind the subject depending on the lens. With apertures of 8 or smaller, not usually a problem as the depth of field was large enough.
I think, I do not know for sure, but AFMA corrected this because the focal plane of the sensor was slightly off with regards to where the lens was projecting the focus relative to the sensor.
Now I can shoot wide open at any aperture with the lenses I have and focus is where I tell it to be. AFMA to my knowledge doesn't change anything in the lens, but rather tells the camera how to shift the sensor so the focal plane is aligned to different lenses. That's why you need to do it for different lenses.
Make sense?

 

[Rienzphotoz]

AFMA to my knowledge doesn't change anything in the lens, but rather tells the camera how to shift the sensor so the focal plane is aligned to different lenses. That's why you need to do it for different lenses.
Make sense?

Interesting ... so the chip in the lens does not make any adjustments to the lens, other than read command(s) off the sensor?

 

[K3nt]

AFMA to my knowledge doesn't change anything in the lens, but rather tells the camera how to shift the sensor so the focal plane is aligned to different lenses. That's why you need to do it for different lenses.
Make sense?

Interesting ... so the chip in the lens does not make any adjustments to the lens, other than read command(s) off the sensor?

To my knowledge that's correct. The corrections are done in the camera body and not the lens. Auto-Focus Micro Adjustment would seem to be making sure the autofocus bits of the sensor are calibrated and not calibrating the lens. Mind you, this is just a possible scenario from a geek perspective how it would be the easiest way to implement AFMA in many camera bodies.

 

[Loswr]

The OP's point seems to be that if the fundamental problem corrected by AFMA is one in the camera body, the lens shouldn't matter. While I offered a couple of possible reasons the lens does matter, whether or not those reasons are correct, it's an empirical fact that the lens does matter. Different lenses (including different copies of the same type of lens) require different AFMA values to achieve optimal focus, zoom lenses require different AFMA values at different points in the zoom range to achieve optimal focus. In fact, subject distance - something not determined by camera or lens - also affects the needed AFMA value.

As for the execution of AFMA, nothing physical is shifted in the camera. It's an electronic correction factor - if the AF system would prompt the lens to move the focus elements a distance of x, the AFMA value modifies that command to x+n.

 

[hgraf]

I believe the lens plays a role - I'm not sure it's really a closed loop with the AF sensor. Roger Cicala's data showing more accurate focusing with the more recent lenses/bodies was supported by the finding of rotational encoders on the USM lens motors. So, with older lenses (pre-2000) we had a basically open loop where the AF sensor determined magnitude and direction of the move and that was transmitted to the lens (look-move) - if the motor moved a ratio slightly off 1:1 from the instructions, AFMA would compensate. The newer lenses+bodies apparently have a closed loop where the encoder reports movement (look-move-confirm). But there may be tolerances in the encoder (e.g. detecting movement as other than a 1:1 ratio) for which AFMA could correct. I'm not positive the loop is closed with the AF sensor, i.e., look-move-confirm-look.

Certainly, a misalignment of the AF sensor with the image sensor is one factor that AFMA corrects. I can personally attest to that - at one point, I dropped my 5DII to the pavement. The camera was perfectly functional, but the sensor alignment changed such that all of my AFMA values (for ~8 lenses) shifted 10 units negative relative to the initial values.

Interesting, I've always assumed it was closed loop, since well, that's how I would have designed it, and all the pieces are there to close the loop. Perhaps the focus speed is hit too hard, but frankly I can't believe that since the second "look" would be nearly instantaneous.

That said, I'm no expert, I'm sure if it's true that focus using the phase detect is open loop, Canon has a very good reason for that.

TTYL

 

[Rienzphotoz]

As for the execution of AFMA, nothing physical is shifted in the camera. It's an electronic correction factor - if the AF system would prompt the lens to move the focus elements a distance of x, the AFMA value modifies that command to x+n.

If its only an electronic correction, couldn't the manufacturers make some software that's kinda "idiot proof" for people like me who find AFMA a bit too tedious and complex? e.g. I shoot some text using the center focus point, with the camera connected to a computer and the software analyses if Af is accurate or not and makes AF adjustments accordingly? ... does that make sense?

 

[Loswr]

As for the execution of AFMA, nothing physical is shifted in the camera. It's an electronic correction factor - if the AF system would prompt the lens to move the focus elements a distance of x, the AFMA value modifies that command to x+n.

If its only an electronic correction, couldn't the manufacturers make some software that's kinda "idiot proof" for people like me who find AFMA a bit too tedious and complex? e.g. I shoot some text using the center focus point, with the camera connected to a computer and the software analyses if Af is accurate or not and makes AF adjustments accordingly? ... does that make sense?

Camera manufacturers could, but they haven't. Reikan did, it's called FoCal. Worth every penny (or pence, in this case).

 

[Rienzphotoz]

As for the execution of AFMA, nothing physical is shifted in the camera. It's an electronic correction factor - if the AF system would prompt the lens to move the focus elements a distance of x, the AFMA value modifies that command to x+n.

If its only an electronic correction, couldn't the manufacturers make some software that's kinda "idiot proof" for people like me who find AFMA a bit too tedious and complex? e.g. I shoot some text using the center focus point, with the camera connected to a computer and the software analyses if Af is accurate or not and makes AF adjustments accordingly? ... does that make sense?

Camera manufacturers could, but they haven't. Reikan did, it's called FoCal. Worth every penny (or pence, in this case).

IC ... Thanks! will look at FoCal

 

[ahab1372]

Interesting, I've always assumed it was closed loop, since well, that's how I would have designed it, and all the pieces are there to close the loop. Perhaps the focus speed is hit too hard, but frankly I can't believe that since the second "look" would be nearly instantaneous.

That said, I'm no expert, I'm sure if it's true that focus using the phase detect is open loop, Canon has a very good reason for that.

TTYL

If it was closed loop, with a second look, we would be only a few steps away from a self-learning AFMA built in into the camera. Now that would be nice, wouldn't it.

 

[Rienzphotoz]

If it was closed loop, with a second look, we would be only a few steps away from a self-learning AFMA built in into the camera. Now that would be nice, wouldn't it.

More than nice ... it'd be AWESOME!

 

[Mt Spokane Photography]

Here is how I understand it. I repair simple issues with lenses, and have dissected a few ones that were totalled.

Inside the lens is a small resistive element with a wiper. The resistance changes as the focus element moves. This is converted to digital by the chip in the lens and the lens position sent to the camera. The lens chip has a table of resistance versus focus distances, it is used to determine the focus distance versus resistance reading. The lens chip can be reprogrammed by Canon or a authorized third party repair station. (Canon sells them the programmer and software to do this)

When the camera does a phase detect, it sends a command to the lens to move focus to a certain position. There can be inaccuracies in the system that add up to being slightly off focus. AFMA tells the camera to offset the command it sends to the lens so that the lens focuses slightly closer or slightly further away.


The problem lies with the linear accuracy of the element and programming in the lens. The lens will now focus correctly at the distance you adjusted it for, and might be off at other distances. If thats the case, and its too severe, Canon can adjust the lens to be accurate at multiple distances, but the user can't.

Canon cameras contain information about all the EF lenses and the commands to send to them. Third party lenses tell the camera that they are a Canon lens and then translate the command they receive to their lens. This can cause even another step that adds to inaccuracy, but it can be adjusted by AFMA as well.

Its a lot more complex that a person might think.

 

[Drizzt321]

If it was closed loop, with a second look, we would be only a few steps away from a self-learning AFMA built in into the camera. Now that would be nice, wouldn't it.

More than nice ... it'd be AWESOME!

The problem is you can't really, unless you also look at the actual image projected onto the sensor. AFMA fixes the situation where the AF detects "in focus", while due to small differences between lens + body combinations, the image projected onto the sensor is not quite in focus. So if the lens mount on the body is every so slightly thicker than the specs but still within tolerances, and the sensor is positioned ever so slightly closer to the back, then while the AF might be in focus, what is projected onto the sensor is just slightly off leading to images that might be a bit software than the might otherwise be. This is most obvious when you have a very shallow DoF, because the plane of focus is very thin where being slightly off from the focus is noticeable.

 

[hgraf]

Canon cameras contain information about all the EF lenses and the commands to send to them. Third party lenses tell the camera that they are a Canon lens and then translate the command they receive to their lens. This can cause even another step that adds to inaccuracy, but it can be adjusted by AFMA as well.

Its a lot more complex that a person might think.

Again, all of that would be solved by closing the loop. Then all you'd need is a correction for sensor/vs AF array, which would be body specific and programmed by Canon at the factory.

It is sounding like the whole phase detect AF system is fully open loop, which really surprised me. Is Nikon like this too? Do they also have an AFMA type feature on their bodies?

Wish there were a Canon engineer I could speak to this about, would be a fascinating discussion!

 

[Drizzt321]

Canon cameras contain information about all the EF lenses and the commands to send to them. Third party lenses tell the camera that they are a Canon lens and then translate the command they receive to their lens. This can cause even another step that adds to inaccuracy, but it can be adjusted by AFMA as well.

Its a lot more complex that a person might think.

Again, all of that would be solved by closing the loop. Then all you'd need is a correction for sensor/vs AF array, which would be body specific and programmed by Canon at the factory.

It is sounding like the whole phase detect AF system is fully open loop, which really surprised me. Is Nikon like this too? Do they also have an AFMA type feature on their bodies?

Wish there were a Canon engineer I could speak to this about, would be a fascinating discussion!

Even with a closed loop, you may not be able to guarantee that it will always be correct since there's long term wear and tear, if the body is sent back to be cleaned and things shift slightly...plus it'll be different if you're using it in cold weather vs hot weather since parts expand/contract and at different amounts.

 

[Rienzphotoz]

As for the execution of AFMA, nothing physical is shifted in the camera. It's an electronic correction factor - if the AF system would prompt the lens to move the focus elements a distance of x, the AFMA value modifies that command to x+n.

If its only an electronic correction, couldn't the manufacturers make some software that's kinda "idiot proof" for people like me who find AFMA a bit too tedious and complex? e.g. I shoot some text using the center focus point, with the camera connected to a computer and the software analyses if Af is accurate or not and makes AF adjustments accordingly? ... does that make sense?

Camera manufacturers could, but they haven't. Reikan did, it's called FoCal. Worth every penny (or pence, in this case).

IC ... Thanks! will look at FoCal

neuroanatomist, I checked Reikan website for FoCal and the price sounds reasonable ... if I get the "Pro" version, how easy or difficult is it to work with?

 

[Loswr]

It's easy. You do need a stable tripod, and lots of light in the target (sun or tungsten/halogen, not fluorescent or LED).

 

[kphoto99]

AFMA to my knowledge doesn't change anything in the lens, but rather tells the camera how to shift the sensor so the focal plane is aligned to different lenses. That's why you need to do it for different lenses.
Make sense?

Interesting ... so the chip in the lens does not make any adjustments to the lens, other than read command(s) off the sensor?

To my knowledge that's correct. The corrections are done in the camera body and not the lens. Auto-Focus Micro Adjustment would seem to be making sure the autofocus bits of the sensor are calibrated and not calibrating the lens. Mind you, this is just a possible scenario from a geek perspective how it would be the easiest way to implement AFMA in many camera bodies.

I have a Tamron lens and T4i, the lens was doing front focus. Tamron adjusted the lens to work with my body. So the lens can be modified.