[toni-a]

Reading a good article about AF microadjustment
https://www.the-digital-picture.com/Phot...-tips.aspx
I came through the concept of focus accuracy vs precision he states  "The terms accuracy and precision are sometimes (improperly) used interchangeably - accuracy is 'closeness to true' whereas precision is repeatability."
they posted this explanatory diagram


which explains why besides good AFMA I still have few missed shots
my guess is precision is a particular body/lens  specific thing and not system related, or does mirrorless have any advantage here ?

[obican]

Go search for "Lensrentals Autofocus Reality" and read every single part.

---

Also yeah, mirrorless have many advantages there.

[Mirko]

Hey! Thank you so much for sharing the article about micro-adjustments with us. I have found it very insightful, and I found many helpful tips there. Also, now the difference between accuracy and precision is more evident. Concerning the mirrorless, I would say that it might have some advantages. Since we are talking about precision and accuracy, I wanted to suggest you look at AI Image Sharpener. This online tool, which btw is free, makes blurry images crisper and enhances photo edges. I have been using it for quite some time now, and I am delighted with the results.

[Arthur Macmillan]

Go search for "Lensrentals Autofocus Reality" and read every single part.

---

Also yeah, mirrorless have many advantages there.

Obican, nice reference article(s).  I like Roger Cicila's summary of MFA and dslrs, but of course pay attention to the qualifications in his summary:

"The conclusion is pretty obvious: If you want to shoot wide aperture prime lenses and you don’t want to use microfocus adjustment, you just refuse to cope with reality."

I've always be fascinated with ultra fast lenses, and yet I only have one or two that are not old film era manual focus lenses.  I say one or to not because I can't remember which lenses I own.  Rather because I am still not sure of what qualifies as ultra fast.  Since Roger did not calibrate the 50mm f/1.4 I'm not sure if he checked that calibration and was happy with it, thinks the inferior focus motor makes calibration that still leaves the lens incapable of accurate focus, or (least likely) that 50mm at 12 feet and F/1.4 was maybe not a narrow enough DOF.  

Here is what the depth of field is according to the DOF calculator on this site:
https://www.dofmaster.com/dofjs.html

d =12 feet, lens is 50mm, f/1.4, and camera is Canon 5D Mk II:
DOF = 1.47 feet, where in focus starts at 11.3 feet and ends at 12.8 feet.

This might seem surprising or boring to you depending on what you like to photograph.  But I just love stuff like this, and if I got it wrong, please correct me!  Now for comparison we got to the 50mmFf/1.2L,

d=12 feet, lens is 50mm F/1.2L on 5D Mark II:
DOF = 1.24 feet.  In focus starting at 11.4 feet and ending at 12.7 feet.

Of course, what Roger Cicala, whom I respect immensely is telling us that the Imatest is able to perceive sharpness well beyond what is considered in focus by the calculator.  The calculator says in focus or not.  The Imatest can compare how sharp an image is.  But my point is to wonder do I have a fast lens that needs MFA adjustment?  Well, I have the 135mm F2L USM.  A pretty old design with no OIS, and opinions of it range from the best bargain in the EF lineup to, well, I don't need to repeat words that are so far gone.  For fun:

d=12 feet, lens is 135mm F/2L USM:
DOF = 0.28 Feet.  In focus starting at 11.9 feet, and in focus ending at 12.1 feet (this is how they round off).

So, anyway, by the standard that people often use you might say that I don't have a single ultra fast lens.  But in my opinion I probably have at least seven.  Moreover at the distances that these lenses are normally used at I don't consider a 50mm F/1.4 MF lens hard to focus at all.

Still, I need to finish reading the series of articles, and do love them so far.  I just find for me, it is actually not that easy to see focus problems with a 50mm lens, where you are likely to see them easily on longer lenses.  Again, I know that the testing system makes that not relevant for his purposes, it does make me wonder why some people worry about real world results with short lenses.

-Mac

[toni-a]

To make it clearer, if you are using a standard 85f1.8 not even f1.4, at 1.5 meters depth of field is 2cm before or after subject, focus differs whether camera focuses on eyes or nose...
So if you have mirrorless without eye autofocus or an SLR without AFMA done perfectly well and autofocus point perfectly chosen you have very high chances of having out of focus portraits.

[Klaus]

Too simplistic - DSLR PD AF can't help with RSAs for instance. In mirrorless systems, it can be handled if stopped down AF metering is used (some do (e.g.Sony), others don't).
Accuracy is also heavily depending on the AF motor in the lens. Some stepping motors are really coarse, others are extremely fine-grained.
DSLR PD AF is heavily relying on the calibration of the AF system and lenses - which is just not necessary in mirrorless cameras.

[Arthur Macmillan]

To make it clearer,  if you are using a standard 85f1.8 not even f1.4, at 1.5 meters depth of field is 2cm before or after subject, focus differs whether camera focuses on eyes or nose...
So if you have mirrorless without eye autofocus or an SLR without AFMA done perfectly well and autofocus point perfectly chosen you have very high chances of having out of focus portraits.

I am not trying to take sides on DSLR or MILC this time.  I know some features are available with mirrorless cameras.  I was just trying to figure out the method as outlined which stated that the AF was set for a single point, the center point.  I'm not trying to dissect the Imatest and how they come up with sharpness.  I guess I was just surprised that one can really have a focus problem with a 50mm lens at 12 feet.  I mean, are we testing focus ability or the intrinsic characteristic of the lens?  And honestly, how could could you possibly beat live view manual focus?  I could tether the camera to a million to one contrast 24 inch monitor then zoom in 10x, and the cameras contrast focus can beat that?  I just was not able to get it focus and sharpness only go as far as the lens goes.  If you want a baseline for the test you have to be able to achieve the best focus manually, otherwise how do you have a standard by which to judge the autofocus systems.  I think Roger C. was right there were problems in the methodology because changing the lenses only proved that the autofocus tests were not appropriate.  For one thing, in general you would always calibrate the MFA for a F/1.4 or F/1.2 lens on a DSLR.  But I still think I am missing something.  And looking ahead at Klaus answer he mentioned something that has bothered me because I happened to be confronted with them in related situations.  So it leads me to another question, that again, I hope is not off topic because it came up when Roger C. decided to change the Micro motor 50/1.4 for the USM 50/1.2.

Just wondering...I do get your point about eyes.  It is a game changer.

-Mac

[Arthur Macmillan]

Klaus said:
Too simplistic - DSLR PD AF can't help with RSAs for instance. In mirrorless systems, it can be handled if stopped down AF metering is used (some do (e.g.Sony), others don't).
Accuracy is also heavily depending on the AF motor in the lens. Some stepping motors are really coarse, others are extremely fine-grained.
DSLR PD AF is heavily relying on the calibration of the AF system and lenses - which is just not necessary in mirrorless cameras.

Sorry, I don't understand the bulletin board method to quote so I just copied and pasted to quote you, Klaus.

The RSA thing is way over my head. I would have thought that again, it is the lens design that determines RSA. Of course stopping down will lessen the problem. If the Sony is running a sort of editing on the fly its pretty awesome, but is that the same as AF? You might just as easily say that a DSLR PD could do focus stacking and call that better focusing. Seems like multiple images/ multiple settings are being merge, or composited. Don't get me wrong. I don't deny they greatness of the mirrorless system. I feel totally obsolete now, but I don't mind because I do just very basic shooting and it relaxes me. I do just use the center point for AF. Where is the art in that? Well, it gets worse, I use single shot mode most of the time. But it does work to focus on what I am pointing at. I live in terror of that AF that tracks where my eye in the viewfinder is looking. I mean, could that be used in court? How could I explain to my girlfriend the things that end up in sharp focus?

Sorry! So here is something sort of cool. I downloaded this free tool GUI for Exiftool. I originally was hoping to find a tool that would edit my EXIF so that the stuff I care about is on top and I don't have to scroll down to see what focal length and what lens were used. I soon lost interest. But what I found, I hope I'm not repeating myself, was an estimation of focus distance from object to image plane. How far away something was! Am I the only one excited about this? Maybe the only one who hasn't known this already? But I also saw something that probably answers the next question. There was the "increment number". And actually it was not a huge number. For an owl 80 feet away, around 24 meters for most of the world, the value was 31 increments. The lens was the better than expected for the price Olympus 75-300mm and it is at 300mm. I assume it would be focusing wide open at F/6.7. So does that mean there are 31 steps from close focus to 24 meters away? You'd think there would be more maybe? And yet some things are hard to see when dealing with mirrorless cameras. The manual focus is by wire. So there would be no way for me to tell how much my manual adjustment has. Is it discrete and not continuous.

Go back to a DSLR lens. Here manual focus means in some case very fine control. But I digress, this was my question.

What does "Linear Focus Motors" mean. I'm going to try to look it up. It's good right? Fast? If the elements actually move in a linear way...well that would not be all good. But by counting increments you would have a built in distance, which is nice. I have to warn you that if you edit the file first before looking at GUI for Exiftool you will probably find that the image editing program also decided what items to display, and the distance and increments will be gone.

I really do have enthusiasm for this new technology. And I'd love to not worry about MFA. I guess I am lucky I have only two fast lenses!

-Mac

[Klaus]

Just think how contrast-detection AF works - it tries to detect the maximum edge contrast pixel level.
It literally reads out the main sensor.

RSAs are focus shifts when stopping down. They do ALWAYS occur - the level can range from negligible to very substantial.

If the camera stops down prior to engaging the CD-AF, RSAs are simply irrelevant. There is no bracketing necessary (I suppose you meant bracketing and not stacking? Stacking is really more meant to increase the DoF in an image).

In theory, a DSLR could use bracketing around the PD-AF point - and do a contrast-detect post-analysis to select the best-focused image.
However, bracketing requires time - time you may not have. Bracketing would also introduce a wobbly viewfinder image.

Funnily, the latter has been criticized in Panasonic mirrorless cameras which use CD-AF only rather than a combo of PD & CD-AF (like Sony/Canon/Olympus/Fuji/Nikon) where this wobbling is not detectable.

That all being said - it's all a question of how holy you want to have it. An occasional misfocus is usually not the end of the world.
I do, however, appreciate "animal AF" for instance.

[toni-a]

That all being said - it's all a question of how holy you want to have it. An occasional misfocus is usually not the end of the world.

All depends on your priorities, and what you are shooting it's true in 99.99% of cases though

the problem is sometimes you take a lot of shots but you have to choose one or two, it's frustrating when you enlarge thumbnail to discover it's the misfocused one .
Experience taught me that in portraits photoshoots I take many photos there will be always a close replacement for A misfocused shot.
When shooting any social event I absolutely avoid wide apertures for important photos using DSLR, increased DOF will almost always compensate for any tiny focus error, on APS-C USING 17-55 AT f4 I don't have any issues, better have some noise than a misfocused shot.