Canon EOS R5 + any RF 35 mm = terrible back focus

Oct 10, 2020
3
0
Hi all!

I have encountered a following problem. Brand new R5 and brand new RF 35 with or without the newest firmware produce a terrible back focus for apertures above 1.8. At aperture 4.0 the back focus is the worst, resulting in the least acceptable resolution in the point of intended focus. I have also tried another copy of rf 35 and the problem persists. This problem sometimes vanishes by itself for no reason and the camera works well until shutdown. I have tried all kinds of settings to find out what might trigger it, firmware updates, all kinds of AF and modes- nothing helps. The only way I can achieve normal results is to mount a 50 mm 1.2 rf first ( flawless work) and then change it to 35 mm without turning off the camera. For some reason it will do a trick and a camera fill hit the focus at any aperture after that until shut down.

I attach a link with a couple of center crops of the target. They are both shot from a tripod at 1/1000 a second, AFS stabiliser off ( but nothing changes if it is on), aperture 4.0, iso 1600 ( yes ISO is high, no if you set it to 100 nothing changes), distance to a target is around 3 meters. One photo is taken after changing a lens from 50 mm, the other straight after I restarted a camera a moment later.

I also attach a center of a shot made at 1.8 aperture which is clearly sharper than the one made at 4.0 after the restart.

Why am I sure that is is a back focus? Because I ran several photoshoots and always saw focus on ears rather than on eyes. And because I tried gazillion shots on a ruler and I clearly see a problem.

What am I asking? I am asking the ones who have this combo to test it and see if the same thing happens. Just pick any target with small text on it like some package and shoot it in AV mode from a tripod or any steady surface at 1,8 and 4,0 apertures and watch if the resolution of the text downgrades.

What I am NOT asking for. Please do not advise to change my shooting technique or make any assumptions unless you do the test yourself. I made this test numerous times accompanied by a technician in a camera shop and I DO manage to achieve perfect results with other lenses or straight after changing the lens from 50 to 35 until I switch the camera off. I also tested another lens in another shop, same result. I already contacted an authorised Canon technician and will take the combo to him on Monday so you will have nicer pictures at 100 ISO, rulers etc.

I do know the Optical limits review, where they stated that 35 rf is prone to focus shift. The focus shift that they illustrated is incomparable to the one I get. Mine results in OOF images where at the sweet spot of the lens you get the worst results.

The crops form undedited raws are attached via the link

LINK REMOVED BY MOD_1 as we don't know the safety of the link.
 

Attachments

Last edited:

Viggo

EOS R5
Dec 13, 2010
4,591
1,294
Besides the fact that it works when mounting 50 first, my initial thought is that this is focusshift like the EF 50 L. Since the R5 focuses wide open it doesn’t matter if it’s mirrorless.
 

Mod_1

Moderator
CR Pro
Dec 11, 2018
5
3
Hi all!

I have encountered a following problem. Brand new R5 and brand new RF 35 with or without the newest firmware produce a terrible back focus for apertures above 1.8. At aperture 4.0 the back focus is the worst, resulting in the least acceptable resolution in the point of intended focus. I have also tried another copy of rf 35 and the problem persists. This problem sometimes vanishes by itself for no reason and the camera works well until shutdown. I have tried all kinds of settings to find out what might trigger it, firmware updates, all kinds of AF and modes- nothing helps. The only way I can achieve normal results is to mount a 50 mm 1.2 rf first ( flawless work) and then change it to 35 mm without turning off the camera. For some reason it will do a trick and a camera fill hit the focus at any aperture after that until shut down.

I attach a link with a couple of center crops of the target. They are both shot from a tripod at 1/1000 a second, AFS stabiliser off ( but nothing changes if it is on), aperture 4.0, iso 1600 ( yes ISO is high, no if you set it to 100 nothing changes), distance to a target is around 3 meters. One photo is taken after changing a lens from 50 mm, the other straight after I restarted a camera a moment later.

I also attach a center of a shot made at 1.8 aperture which is clearly sharper than the one made at 4.0 after the restart.

Why am I sure that is is a back focus? Because I ran several photoshoots and always saw focus on ears rather than on eyes. And because I tried gazillion shots on a ruler and I clearly see a problem.

What am I asking? I am asking the ones who have this combo to test it and see if the same thing happens. Just pick any target with small text on it like some package and shoot it in AV mode from a tripod or any steady surface at 1,8 and 4,0 apertures and watch if the resolution of the text downgrades.

What I am NOT asking for. Please do not advise to change my shooting technique or make any assumptions unless you do the test yourself. I made this test numerous times accompanied by a technician in a camera shop and I DO manage to achieve perfect results with other lenses or straight after changing the lens from 50 to 35 until I switch the camera off. I also tested another lens in another shop, same result. I already contacted an authorised Canon technician and will take the combo to him on Monday so you will have nicer pictures at 100 ISO, rulers etc.

I do know the Optical limits review, where they stated that 35 rf is prone to focus shift. The focus shift that they illustrated is incomparable to the one I get. Mine results in OOF images where at the sweet spot of the lens you get the worst results.

The crops form undedited raws are attached via the link

LINK REMOVED BY MOD_1 as we don't know the safety of the link.
Please attach jpeg images. I deleted the link as you are a first time poster and we do not know the safety.
 
  • Like
Reactions: zim

Mt Spokane Photography

I post too Much on Here!!
CR Pro
Mar 25, 2011
16,364
1,370
I've used the two quite a bit now. AF has been amazingly good, I don't recall getting even one out of focus shot. I'd tend to think there is a lens issue.
 
Oct 10, 2020
3
0
I've used the two quite a bit now. AF has been amazingly good, I don't recall getting even one out of focus shot. I'd tend to think there is a lens issue.
I highly doubt that it has anything to do with a lens. I tested the body with another copy, identical results. And I already said that I can make the combo work if I attach the lens after I unmount 50mm. 99% sure that the body brains are f**d up. Now wondering how I am going to explain that to canon. This all sounds as an absurd but here it is.
 

SHAMwow

I'm New Here
CR Pro
Sep 7, 2020
15
16
Yeah now that that AF is on board, if I have huge focus issues I'd assume its a lemon lens and return it. Depends how large the miss is though as they have been patching small focus issues through firmware. This doesn't sound like that though.
 
Oct 10, 2020
3
0
Yeah now that that AF is on board, if I have huge focus issues I'd assume its a lemon lens and return it. Depends how large the miss is though as they have been patching small focus issues through firmware. This doesn't sound like that though.
This is not most surely not a lemon lens, as I tested another brand new with similar results. And there is away to make a lens work, described in the original post: switching to it from the rf 50 mm 1.2 ( switching from zooms does not help). This looks like some crazy bug on the board.
 

SHAMwow

I'm New Here
CR Pro
Sep 7, 2020
15
16
This is not most surely not a lemon lens, as I tested another brand new with similar results. And there is away to make a lens work, described in the original post: switching to it from the rf 50 mm 1.2 ( switching from zooms does not help). This looks like some crazy bug on the board.
Okay then. Based on this and the OP, I"m not sure what the goal of posting about this was. I'm glad you found a work around, but that is incredibly random and arbitrary. Good luck with the issue though.
 

YuengLinger

Sufficiently Pixilated
CR Pro
Dec 20, 2012
3,109
1,374
USA
Unless I missed it, please let us do some simple trouble-shooting:

Does the R5 in question work with other lenses without the same problem?

Does the 35mm in question work on another Rf-mount body without the same problem?

Please be clear. Thank you!

ALSO: Your "flat" charts aren't helping a lot. We need to see a shot of a target that is at a 45 degree angle to the plane of the sensor, so that the far end, the target point, and the close end are not all in the same plane of focus. Like the one shown right at the top of the article here: