I was stunned when I received an email today for Canon’s “Precision Autofocus Alignment” service. At first, I thought it must be a DSLR-only service, but… no, they do this for R-series cameras and RF lenses.
I thought the whole point of mirrorless technology is to avoid this kind of thing. Has anyone heard of this, and what it is that’s adjusted?
This is a very specific scenario, ultra high magnification in a macro lens…what was worrying in the email is the thought that overtime any combo might need to be fine tuned…”Over time, however, small movements in components caused by frequent or heavy use can result in tiny changes in focus accuracy”.
Thought we were done with this kind of concern by switching to mirrorless
The “whole point of mirrorless” is to remove one failure prone mechanical component by an electronic one, speaking in layman’s terms, and if you can say that there is the one reason at all.
I fail to see how mirrorless cameras could significantly mitigate focus shift issues compared to DSLR bodies.
Personally I think this service is rarely necessary, especially compared to rangefinder cameras.
Well, not an expert in DSLRs, but the higher end ones have the AF micro adjustment functionality as at wider apertures the accuracy may vary lens by lens when shooting through the viewfinder using the non-sensor based AF system…someone please come to rescue if I got this wrong!
Sure, DSLRs have an additional AF unit, but they also have live view and CDAF just like its successor, the MILC. No difference, despite maybe some evolutionary improvements.
Also:
Again, no difference.
does not matter - it illustrates that PDAF can't fix the problem with optics design unless you have an option to force camera to focus @ aperture set for a shot ... and as is in many cases (raw histogram, etc) camera manufacturers continue to deny users options to control how they want to focus the lenses ... which is very doable thing in firmware
dSLRs got CDAF way after mirrorless P&S cameras ... those are true ancestors of dLSMs, not dSLRs
very simple - give users an option (an option - not as the only choice) to select AF focusing @ the same aperture that is selected for a shot
Interesting - although to be honest I have never encountered a problem with AF calibration on any Canon lens on any DSLR or mirrorless in just shy of 20 years and lots of lenses - any issue to me is a fault. Sigma absolutely had issues, with 30 1.4, 35 1.4 (consistency as well as shift), 150mm macro (needed recalibration) as did the wide end of a sigma 100-400.
Have seen residual spherical aberration causing focus shift stopping down - so unless AF done with the lens stopped down, don't see how any service on either a DSLR or mirrorless would fix that?
Unfortunately I wasn’t able to find out what exactly Canon is doing, but after some research it looks like they are adjusting an x offset / compensation value, something that you can do by yourself and for free on Nikon bodies, see www.nikonimgsupport.com/na/NSG_article?articleNo=000043897&configured=1&lang=en_SG&sfdcIFrameOrigin=null
That’s a feature of Nikon DSLR’s which Canon DSLR’s also support. It’s because the DSLR’s AF sensor is in the pentaprism. The problem with this approach is it enables only one offset for primes and two offsets for zooms; it helps for certain but it isn’t perfect.
But of course the concept in mirrorless is that focus is measured right on the sensor so it’s perfect… or so I thought. I’m not sure what’s adjusted here, but Nikon does not appear to offer this in their mirrorless cameras.
It’s not clear to me
Using Reikan's FoCal, I could reproduce some need for AF microadjustment for most of my lenses with my 5D mk III and IV bodies, with genuinely but mostly subtly improved results (I think it's either my second-hand 85mm f/1.2L or 200mm f/2.8L from 90's that needed a fairly significant adjustment). So far I've mostly used only the EF 100-400mm mk II with my R7 and I don't think that one has much of focus shift in the first place. Certainly the focusing does a very good job with that combination. But I guess focus shift can be problem with lenses such as the EF 50mm f/1.2L.
From my understanding it’s the offset for AF calculation. I get the feeling that we are talking about two different things.
I had my R5 in for a warranty repair (IS issue on a referb from Canon) and when I got it back it seemed like it was OK, but after a lot of testing I was actually struggling with accurate autofocus. It would consistently back focus at all distances with all my lenses, both EF (Canon/Tamron/Sigma on a canon adapter) and my RF lenses. I was also thinking that accurate auto focus was given on mirrorless so it had to be an issue with my lens, but all of my lenses couldn't be trash lol, so I sent the body in for repair/adjustment and got in back in a few days (Newport News Canon is only 2 hours away). Focus accuracy was restored.
Now since they posted the service I'll have to keep my eye out for focus drift. They probably need to restore focus adjustment to the menus again instead of charging for focus service. It's not a Leica rangefinder.
It sounds like sensor shift was the factor here. Interesting. Late last year I bought a refurbed R, which of course does not have IBIS. Recently I had the chance to use it in a low light shoot where I brought all my f/2 or better primes. All are unstabilized except for the 35. They performed brilliantly on the R without any adjustment, since it has none, even though on my 6D the adjustments for them are anywhere from -7 to +10 depending on the lens.