Canon R5 backfocus.

[lukiv3]

finally i confirmed my R5 is backfocusing

i know Mirrorles shouldn't do that but probably it's a slightly sensor missalligment.
Firstly i noticed using my Sigma 35mm f1.4, Eye autofocus was locking on eye, but reviewing pictures sharpest point was ears. I updated FW, issue mostly dissapeared but still slightly exist. Then i adapted my old Tamron 150-600 i was getting missed shots when distance was greater. I've done many tests including rulers, on 20m distance was missing not by much (around 1-2cm back) But for greater distances 60m> focus sometimes was missing by meter or more. I was blaming old lens, requiring FM update (need's sending to service). I connected Sigma issue to Tamron and simply i thought it's just a lens but recently i sold Tamron and picked up Sigma Sport 150-600, and what a surprise... Same issue, slightly less than T but issue persist then i also tested rest of my lenses.
EF 70-200 2.8 IS mk1. Same small backfocus on every locked point on greater distances.
EF16-35 III. Hardest to spot as this lens have bigger dof, but after locking AF on subject, small pull on focus ring makes subject slightly sharper.
Sigma 35mm f1.4 - giving best results so far after update, very small backfocus really hard to justify. but exists.
On EVERY lens and after every focus confirm pulling back focus ring slightly makes subject tiny bit sharper

Also what i've noticed is all lenses are not focusing back to MFD's (always red square) but can focus from pulling back manually, also Focus bracketing when reach infinity still makes 3-4 extra shots for no reason, that's also giving proof sensor don't match with focusing.

I made this thread in few R5 FB groups to find people with simmilar results and maybe sollution to this?

 

[Chris.Chapterten]

That doesn’t look right at all. Since you are experiencing the problem with genuine Canon lenses, you will be able to send the camera in under warranty to get it repaired.

By comparison my R5 focus is very good. The only time I ever had back focus was with the RF 35mm 1.8. I sold that lens.. good luck!

 

[lukiv3]

I suppose there is more cases like mine but simply people don't see it on theirs cameras because it's 45mpix camera (more than 8K res) and most of people don't even look at their pictures on 4K screens so it's really hard to notice when you don't pixel peep, sometimes people can see it sharp not even noticing it's slightly OOF, that's why is so much easier to notice on greater distances with super wide open lenses. Even on my test charts, when you 100% crop when camera locked, looks like sharp till You really analize that object's behind are even sharper, not by much but it's noticeable. I hope more people will do chart test on greater distances (20m. and above)

 

[lukiv3]

Camera sent to Canon Service (UK) to further inspection.

 

[Chris.Chapterten]

Camera sent to Canon Service (UK) to further inspection.

Good luck! Let us know how you go with the repair process

 

[entoman]

I suppose there is more cases like mine but simply people don't see it on theirs cameras because it's 45mpix camera (more than 8K res) and most of people don't even look at their pictures on 4K screens so it's really hard to notice when you don't pixel peep, sometimes people can see it sharp not even noticing it's slightly OOF, that's why is so much easier to notice on greater distances with super wide open lenses. Even on my test charts, when you 100% crop when camera locked, looks like sharp till You really analize that object's behind are even sharper, not by much but it's noticeable. I hope more people will do chart test on greater distances (20m. and above)

I've always considered the AF on my R5 to be very accurate, but just as an experiment, I tried a test using focus bracketing. I used the default increments and took a series of 3 shots using my EF 100mm macro at F4, at a focus distance of about 1 metre. To my surprise the sharpest shot was the middle one in the series, indicating a small degree of back-focus. I repeated the test several times on various subjects, always using the smallest focus spot for maximum accuracy. In every case, this slight back focus was evident.

What was the outcome after sending your camera to Canon?

 

[lukiv3]

So my camera was almost 6 weeks in authorised Canon service. I got it back and done some test, everything was looking promissing. They told me, issue was not easy to repeat, but indeed they found it and commited to reset all inbody dimensions. Indeed after receiving camera focusing was much better than before but recently i was in LONG photography trip and there is big ammount of pictures slightly backfocused unfortunately. Now i'm not sure it's a camera or it's a lens or me, or maybe how R5's AF work. I found that even on smallest AF spot focusing stops metering when there is anything sharp inside te box so this can be even few pixels of tree branch behind subject(bird).

 

[Mt Spokane Photography]

The in body dimension which affects focus is the flange to backplane dimension. I thought that mirrorless compensated for that. Of course, if the sensor is not parallel to the focal plane, there will always be a issue, and I'm not sure how a IBIS Unit affects focus but it definitely does.