RF100mm F2.8 L macro - Focus Shift is Fixed/Addressed

[homebodyMacro]

It would be very interesting if other RF 100mm macro owners chimed in with their firmware version and presence or absence of focus shift.

I think we could likely determine where the change occurred if we had enough RF 100mm owners (probably a bit of a stretch with the size of the userbase).

Currently: firmware 1.1.1 is 2/2 with no focus shift.

Bit of a sidetrack there, but still interested in this (on topic) if there's anyone out there with the RF 100 macro lens.

 

[Shooters on My Squad]

It would be very interesting if other RF 100mm macro owners chimed in with their firmware version and presence or absence of focus shift.

I think we could likely determine where the change occurred if we had enough RF 100mm owners (probably a bit of a stretch with the size of the userbase).

Currently: firmware 1.1.1 is 2/2 with no focus shift.

Bit of a sidetrack there, but still interested in this (on topic) if there's anyone out there with the RF 100 macro lens.

I own one, and I heard about the focus shift issues after I bought it. It might happen with my lens, but I’ve never noticed it in real-world use. I’m very satisfied with it, though I tend to reach for the 85mm more often, so the 100mm doesn’t get as much action. Maybe I’ll try testing it with a sheet of paper or a ruler just to see.

 

[homebodyMacro]

I own one, and I heard about the focus shift issues after I bought it. It might happen with my lens, but I’ve never noticed it in real-world use. I’m very satisfied with it, though I tend to reach for the 85mm more often, so the 100mm doesn’t get as much action. Maybe I’ll try testing it with a sheet of paper or a ruler just to see.

If you do take a look, please let us know your results (and the firmware version of your copy of the lens)!

Would love to hear from more folks on here.

 

[Shooters on My Squad]

I own one, and I heard about the focus shift issues after I bought it. It might happen with my lens, but I’ve never noticed it in real-world use. I’m very satisfied with it, though I tend to reach for the 85mm more often, so the 100mm doesn’t get as much action. Maybe I’ll try testing it with a sheet of paper or a ruler just to see.

If you do take a look, please let us know your results (and the firmware version of your copy of the lens)!

Would love to hear from more folks on here.

Weekend’s coming, and I’m curious—still got some hope left. Which focus distance did you use? Lens MFD of 26 cm?

--
You are talking to crazy. With too many cameras.

 

[Shooters on My Squad]

Unless RF100mm macro lenses with firmware 1.1.1 are shown to have the same focus shift issue that was very, very clear in the TDP review, I think the simplest explanation is that it simply did not meet their internal risk threshold for a product advisory.

For example, in that list of RF lenses and firmware versions, there are specific firmware versions for only some cases. We have a 1.0.0, a 1.0.3, 1.0.6, 1.0.8, 1.1.0, 1.1.1, etc., etc. The versions increase from the right i.e. 1.0.0 < 1.0.1 < 1.1.0 < 1.1.1, etc.

For the RF 24-105mm F4L IS USM, it is at 2.0.6 - yikes. But if we use a search engine, we see releases for only a few of these:

site:https://www.usa.canon.com/support/canon-product-advisories/ RF 24-105mm F4L IS USM

or this search, if you want to be broader:

site:https://www.usa.canon.com/support/ RF 24-105mm F4L IS USM "firmware"

Firmware update notices for this lens (24-105 F4L) appear only for versions 2.0.0, 2.0.1, 2.0.6. Their software engineers are not just generating random firmware/software version numbers (though that would be a bit funny at this point), so there are clearly many versions that are not publicly announced, including the entire 1.X.X series!!!

This is something I don’t get. What do you mean by "unannounced"? There should still be firmware downloads available on Canon's page, but I don't see anything except the manual on the RF 100mm support page. If there’s no firmware download, does that imply there aren’t any firmware updates?

https://www.usa.canon.com/support/p/rf100mm-f2-8-l-macro-is-usm

 

[Shooters on My Squad]

The Canon website doesn't show any firmware downloads for the RF100mm macro. That's the only thing that doesn't quite pass the sniff test. However, an official firmware fix also requires admitting to a significant issue which may have a level of liability they don't like. Maybe they consider it minor and didn't officially release it as a "fix". Corporations tend to be terrible, so this seems very possible. 1.0.0 is "completely new firmware"; mine shows as 1.1.1, and there are lenses that are still at 1.0.0 here: https://rfshooters.com/blog/lenses/

This really doesn’t mean much. Sometimes a firmware is finalized when pre-production samples are sent out, other times bugs are found, and the release number is bumped up before the lens even hits the shelves. That’s why you might see a higher number on release. The link you provided says "Release Firmware" for the RF 100mm, which suggests there haven’t been any firmware updates. So something doesn’t add up here.

I think there’s just some confusion around the firmware updates, and it got me confused too, haha. No worries, though—I’d still like to replicate the tests with my copy and whatever firmware Canon shipped. I’d appreciate more info on your test setup so I can replicate it as closely as possible.

--
You are talking to crazy. With too many cameras.

 

[Shooters on My Squad]

The Canon website doesn't show any firmware downloads for the RF100mm macro. That's the only thing that doesn't quite pass the sniff test. However, an official firmware fix also requires admitting to a significant issue which may have a level of liability they don't like. Maybe they consider it minor and didn't officially release it as a "fix". Corporations tend to be terrible, so this seems very possible. 1.0.0 is "completely new firmware"; mine shows as 1.1.1, and there are lenses that are still at 1.0.0 here: https://rfshooters.com/blog/lenses/

This really doesn’t mean much. Sometimes a firmware is finalized when pre-production samples are sent out, other times bugs are found, and the release number is bumped up before the lens even hits the shelves. That’s why you might see a higher number on release. The link you provided says "Release Firmware" for the RF 100mm, which suggests there haven’t been any firmware updates. So something doesn’t add up here.

I think there’s just some confusion around the firmware updates, and it got me confused too, haha. No worries, though—I’d still like to replicate the tests with my copy and whatever firmware Canon shipped. I’d appreciate more info on your test setup so I can replicate it as closely as possible.

Okay, so I tested from f/2.8 to f/11. My lens has firmware 1.1.1, and I believe this is the only released firmware for it. Personally, I’m not seeing any focus issues, similar to what others have reported, e. g. in this thread.

So, I’m not sure what’s going on here or about Canon’s "official" statement regarding focus shift issues being expected with this lens. At least to my eyes, I can’t spot any focus shifting in my images. Correct me if I’m wrong, but focus seems to always be dead center on the 5. I reacquired focus between every aperture change.

What’s your take?


f/2.8


f/4


f/5.6


f/8


f/11

Maybe the reviewers chatted among themselves and ended up with a similar conclusion, which happens more often than we think. Still, it’s a bit shady if Canon actually gave that statement.

But whatever, my lens seems solid. Time to move on and shoot! No point wasting time on this over the weekend—better to use that time snapping more photos, haha.

--
You are talking to crazy. With too many cameras.

 

[Shooters on My Squad]

So, I’m not sure what’s going on here or about Canon’s "official" statement regarding focus shift issues being expected with this lens. At least to my eyes, I can’t spot any focus shifting in my images. Correct me if I’m wrong, but focus seems to always be dead center on the 5. I reacquired focus between every aperture change.

Alright, here’s another test, this time without reacquiring focus between aperture changes (which seems to be what they mean by "focus shift" in this context, I’m more familiar with focus shift in zoom lenses where the focus shifts after zooming, so I wasn’t really familiar with this concept on primes). But here it is, and again, I’m not seeing any issues at all.

This time, I just focused on the 5 once, then used the Canon app to change apertures and snap away. Still looks solid to me.


f/2.8




f/4




f/5.6




f/8




f/11



--
You are talking to crazy. With too many cameras.

 

[homebodyMacro]

Weekend’s coming, and I’m curious—still got some hope left. Which focus distance did you use? Lens MFD of 26 cm?

I didn't exactly measure it. I did mark one of my posts as being as close as I could get to the MFD (1.4x magnification). The distance was constant though for all matched sets of photos (i.e. per post).

This is something I don’t get. What do you mean by "unannounced"? There should still be firmware downloads available on Canon's page, but I don't see anything except the manual on the RF 100mm support page. If there’s no firmware download, does that imply there aren’t any firmware updates?

https://www.usa.canon.com/support/p/rf100mm-f2-8-l-macro-is-usm

Shooters on My Squad wrote:

This really doesn’t mean much. Sometimes a firmware is finalized when pre-production samples are sent out, other times bugs are found, and the release number is bumped up before the lens even hits the shelves. That’s why you might see a higher number on release. The link you provided says "Release Firmware" for the RF 100mm, which suggests there haven’t been any firmware updates. So something doesn’t add up here.

I think there’s just some confusion around the firmware updates, and it got me confused too, haha. No worries, though—I’d still like to replicate the tests with my copy and whatever firmware Canon shipped. I’d appreciate more info on your test setup so I can replicate it as closely as possible.

Canon has a 3-component version system, i.e. Major.Minor.Patch versions.

Without being on their software team, we cannot say for sure exactly how they handle things.

However, generally speaking for X.Y.Z, a given X.Y will have a few patches (Z values), and a major version X. will have a few minor versions.

This means that each major version will probably have a handful to a dozen or more firmware versions at least on the Canon end.

From the RF landscape ( https://rfshooters.com/blog/lenses/ ), most lenses don't have a known firmware version. A standard practice would be to release to production at least at 1.0.0. As we can see, there are known versions indicating that patch releases are fairly common, and minor versions seen are 0 or 1. That doesn't mean there can't be more minor releases; we see 2.0.6 for the RF 24-105 F/4 L, but we don't know anything about the 1-series major release and how many patches and minor releases it had.

...

When I say unreleased, I mean that it appears that Canon only posts firmware downloads to their website for a given lens when there is a corresponding "Firmware Notice", e.g. https://www.usa.canon.com/support/c...-RF400mm-F2-8-L-IS-USM-Firmware-Version-1-0-6

...

Definitionally, firmware notices correct some issue that Canon felt necessary to provide a download for. Flashing firmware can be risky - it can ruin devices it something goes wrong. It is also somewhat an admission of fault on their part. However, we cannot find the prior firmware version(s) from production for cases even where Canon has posted a firmware download. From the RF product line, it is clear that they do not normally disclose the firmware version during production.

...

All the moving pieces together, we have the following:

1. We only see RF firmware versions when a Firmware Notice is given and a download is available.

2. The RF lineup firmware versions available indicate normal software development practices with a 3-component versioning system.

3. For some lenses there are clearly MANY firmware versions that occurred prior to a notice making the current (at time of notice) firmware Major.Minor.Patch version clear.

4. Firmware Notices occurring at widely varying version states indicate that firmware development occurs during production, and firmware versions can occur without any Firmware Notice/download announcing this.

5. It may be the case that not all firmware versions are released to production. Even if production is less than the full set of versions, it is still the case that known firmware versions that have a notice are a tiny subset of total firmware versions.

6. Thus, it is extremely likely that there are updates to production-state lens firmware that Canon does not announce with a "Firmware Notice", meaning that an issue like focus shift for this RF 100mm macro could be addressed with no announcement and no download available for those with an earlier copy.

Onto your other posts:

Okay, so I tested from f/2.8 to f/11. My lens has firmware 1.1.1, and I believe this is the only released firmware for it. Personally, I’m not seeing any focus issues, similar to what others have reported, e. g. in this thread.

So, I’m not sure what’s going on here or about Canon’s "official" statement regarding focus shift issues being expected with this lens. At least to my eyes, I can’t spot any focus shifting in my images. Correct me if I’m wrong, but focus seems to always be dead center on the 5. I reacquired focus between every aperture change.

What’s your take?

Maybe the reviewers chatted among themselves and ended up with a similar conclusion, which happens more often than we think. Still, it’s a bit shady if Canon actually gave that statement.

But whatever, my lens seems solid. Time to move on and shoot! No point wasting time on this over the weekend—better to use that time snapping more photos, haha.

Shooters on My Squad wrote:

Alright, here’s another test, this time without reacquiring focus between aperture changes (which seems to be what they mean by "focus shift" in this context, I’m more familiar with focus shift in zoom lenses where the focus shifts after zooming, so I wasn’t really familiar with this concept on primes). But here it is, and again, I’m not seeing any issues at all.

This time, I just focused on the 5 once, then used the Canon app to change apertures and snap away. Still looks solid to me.

Thanks for your testing! It looks good to me at least from a relatively quick look. From a closer look and trying to bisect the blur to get a focus plane location, f/2.8 and f/5.6 look the exact same to me (as I saw in my own testing).

Also - I'm glad you mentioned focus prior to / after changing aperture. Just to note my process during testing:

I did reacquire focus in all my testing. The TDP review says "Canon cameras autofocus with the aperture wide open, not accounting for the stopped down depth of field or resulting plane of sharp focus", so presumably it should not make a difference when the autofocus is applied for this test. But still worth noting.

Regarding the reviewers, here's what I said previously on the TDP review and its focus shift assessment:

That's the clearest illustration of the focus shift issue that I've seen ( https://www.the-digital-picture.com/Reviews/Canon-RF-100mm-F2.8-L-Macro-IS-USM-Lens.aspx ).

If you scroll about halfway down, there's an image of some letters "GRI" illustrating the issue. f/2.8 and f/11 are clear, and the focus shift issue exists in between. I have linked to images from this set above, i.e.:

f/2.8 clear: https://media.the-digital-picture.c...F2.8-L-Macro-IS-USM-Lens/Focus-Shift/f28.webp

f/5.6 quite soft: https://media.the-digital-picture.c...F2.8-L-Macro-IS-USM-Lens/Focus-Shift/f56.webp

At the very least, in the pro testing they did and the examples they have, I think the issue is extremely clear. I don't think they faked the images or botched an easy test. The shift is real and obvious in their images.

I also think it is highly unlikely that they just got two "bad" copies of the lens that happened to align with a statement from Canon's engineers, as with Dustin Abbott etc. I think my testing is at least comparable to theirs, and the focus is equal in all my photos to a very high degree of confidence; I did extreme crops and looked down to the pixel on all my shots. I also did both planar and depth-based tests with high certainty on both results. So, I do still think it is most likely to be related to software.

On the DPR link you sent, I hadn't seen that old thread - thanks for sharing. Very interesting that the people on the forum couldn't replicate the issue, even back then.

In any case, the firmware version of people's lenses and the presence or absence of noticeable shift is the only way to determine if it changed at some point. The Canon firmware landscape indicates such a change is absolutely possible.

So far, we have RF 100mm F/2.8 L Macro firmware 1.1.1: 3/3 no focus shift observed.

Again, thanks for testing and sharing the results! I agree to "just go shoot", though that is not why I made this topic; I made it because I saw recent comments by others who were knocking or avoiding the lens due to this widely-reported focus shift issue (at least in technical lens reviews, which have a fairly niche audience).

Basically, I knew mine was good, and from my investigation here I think it is highly likely that new copies of the lens no longer have the issue. I shoot with it constantly and absolutely love it, so I hope more people will see our testing and have fewer reservations about getting the lens. Since I have a ton of fun with it, I think it's a shame if outdated information prevents people from enjoying it.

 

[Shooters on My Squad]

...

All the moving pieces together, we have the following:

1. We only see RF firmware versions when a Firmware Notice is given and a download is available.

2. The RF lineup firmware versions available indicate normal software development practices with a 3-component versioning system.

3. For some lenses there are clearly MANY firmware versions that occurred prior to a notice making the current (at time of notice) firmware Major.Minor.Patch version clear.

4. Firmware Notices occurring at widely varying version states indicate that firmware development occurs during production, and firmware versions can occur without any Firmware Notice/download announcing this.

5. It may be the case that not all firmware versions are released to production. Even if production is less than the full set of versions, it is still the case that known firmware versions that have a notice are a tiny subset of total firmware versions.

6. Thus, it is extremely likely that there are updates to production-state lens firmware that Canon does not announce with a "Firmware Notice", meaning that an issue like focus shift for this RF 100mm macro could be addressed with no announcement and no download available for those with an earlier copy.

I think points 1–5 are overcomplicating things, and point 6 draws a false conclusion.

As I mentioned before, any firmware that gets onto a lens is either shipped directly with the lens or made available for download on Canon’s support pages. Your wording makes it sound like Canon has secret firmware versions they don’t announce, and that there are mythical Canon lenses out there with these magic firmware versions.

But it’s not like that.

As I said, every manufacturer aims to ship a bug-free version 1.0.0. If that’s not possible, some lenses get a higher initial version. There might be pre-production units floating around with version 1.0.0 when the only public version is 1.1.1. The only way to clarify this would be to ask the reviewers who experienced focus shift issues which firmware they tested on, but since this was years ago, they likely won’t remember.

Okay, so I tested from f/2.8 to f/11. My lens has firmware 1.1.1, and I believe this is the only released firmware for it. Personally, I’m not seeing any focus issues, similar to what others have reported, e. g. in this thread.

So, I’m not sure what’s going on here or about Canon’s "official" statement regarding focus shift issues being expected with this lens. At least to my eyes, I can’t spot any focus shifting in my images. Correct me if I’m wrong, but focus seems to always be dead center on the 5. I reacquired focus between every aperture change.

What’s your take?

Maybe the reviewers chatted among themselves and ended up with a similar conclusion, which happens more often than we think. Still, it’s a bit shady if Canon actually gave that statement.

But whatever, my lens seems solid. Time to move on and shoot! No point wasting time on this over the weekend—better to use that time snapping more photos, haha.

Shooters on My Squad, post: 67883927, member: 2282938"]
Alright, here’s another test, this time without reacquiring focus between aperture changes (which seems to be what they mean by "focus shift" in this context, I’m more familiar with focus shift in zoom lenses where the focus shifts after zooming, so I wasn’t really familiar with this concept on primes). But here it is, and again, I’m not seeing any issues at all.

This time, I just focused on the 5 once, then used the Canon app to change apertures and snap away. Still looks solid to me.

Thanks for your testing! It looks good to me at least from a relatively quick look. From a closer look and trying to bisect the blur to get a focus plane location, f/2.8 and f/5.6 look the exact same to me (as I saw in my own testing).

Also - I'm glad you mentioned focus prior to / after changing aperture. Just to note my process during testing:

I did reacquire focus in all my testing. The TDP review says "Canon cameras autofocus with the aperture wide open, not accounting for the stopped down depth of field or resulting plane of sharp focus", so presumably it should not make a difference when the autofocus is applied for this test. But still worth noting.
[/QUOTE]
I think for this test, the focus shouldn’t be reacquired between aperture changes. However, the definition of "focus shift" in this context isn’t exactly clear, even after re-reading the reviews.

In any case, the firmware version of people's lenses and the presence or absence of noticeable shift is the only way to determine if it changed at some point.

There’s only version 1.1.1 for this lens. Anything else would be pre-production or beta.

Again, thanks for testing and sharing the results! I agree to "just go shoot", though that is not why I made this topic; I made it because I saw recent comments by others who were knocking or avoiding the lens due to this widely-reported focus shift issue (at least in technical lens reviews, which have a fairly niche audience).

I agree. The only thing we can do now is ask around to see if others are experiencing this issue. From my perspective, some threads are popping up because potential buyers are unsure and want a second opinion. But considering this is one of the most popular Canon prime L lenses out there, we’d be hearing a lot more about it if it were a widespread issue.

Basically, I knew mine was good, and from my investigation here I think it is highly likely that new copies of the lens no longer have the issue. I shoot with it constantly and absolutely love it, so I hope more people will see our testing and have fewer reservations about getting the lens. Since I have a ton of fun with it, I think it's a shame if outdated information prevents people from enjoying it.

Exactly. That’s why I’d love to see even more similar tests where people can confirm that focus shift isn’t really an issue with this lens. The more data we have, the clearer the picture will be for everyone.

 

[homebodyMacro]

I think points 1–5 are overcomplicating things, and point 6 draws a false conclusion.

As I mentioned before, any firmware that gets onto a lens is either shipped directly with the lens or made available for download on Canon’s support pages.

The latter statement is trivial/vacuous - any firmware on any lens is shipped with it or downloaded onto it.

Your wording makes it sound like Canon has secret firmware versions they don’t announce, and that there are mythical Canon lenses out there with these magic firmware versions.

But it’s not like that.

[citation needed]

As I said, every manufacturer aims to ship a bug-free version 1.0.0. If that’s not possible, some lenses get a higher initial version. There might be pre-production units floating around with version 1.0.0 when the only public version is 1.1.1. The only way to clarify this would be to ask the reviewers who experienced focus shift issues which firmware they tested on, but since this was years ago, they likely won’t remember.

...

There’s only version 1.1.1 for this lens. Anything else would be pre-production or beta.

[citation needed].

As I've said in the thread, I encourage everyone to review the evidence and come to their own conclusions.

My own conclusions are the simplest explanation of all of the evidence that I'm aware of. My conclusion, as I've stated, is not absolute by any means. It's an abductive conclusion based on the lens observations, corporate production practices, software engineering practices, the RF firmware landscape, and the fact that RF firmware versioning is only visible when a notice has been published. It is a valid hypothesis that is entirely falsifiable. For instance it is possible, though unlikely based on the TDP review statements and quotes, that reviewers all had an inferior pre-production version of the lens.

You can certainly conclude based on opinion or intuition that my hypothesis is "overcomplicated" or false, but I don't find that to be a compelling argument nor evidence for one.

As I've stated, I'm certainly interested in falsifying my hypothesis, or on the off chance we get a large data sample, confirming it. I would love to get more replies and testing to see what firmware versions we find along with the presence or absence of focus shift.

I agree. The only thing we can do now is ask around to see if others are experiencing this issue. From my perspective, some threads are popping up because potential buyers are unsure and want a second opinion. But considering this is one of the most popular Canon prime L lenses out there, we’d be hearing a lot more about it if it were a widespread issue.

Exactly. That’s why I’d love to see even more similar tests where people can confirm that focus shift isn’t really an issue with this lens. The more data we have, the clearer the picture will be for everyone.

We agree on all these points. If you know other folks who have the lens IRL or on these message boards, I'd love to hear about the firmware version and presence or absence of focus shift.

The only thing we can do is get more data; personally I am interested in learning more, regardless of the conclusion. It is generally more interesting to be wrong because that means we learned more as things progressed.

...

The current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 3/3 no focus shift observed.

 

[homebodyMacro]

We're still at this current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 3/3 no focus shift observed.

Would love to hear from other owners of the rf 100 macro regarding the presence or absence of focus shift and the firmware version of their lens.

 

[David Yule]

I've got lens firmware Version 1.1.1 which the lens came with (purchased in August 2024) and I don't see any focus shift.

 

[homebodyMacro]

I've got lens firmware Version 1.1.1 which the lens came with (purchased in August 2024) and I don't see any focus shift.

Thanks for sharing, David!

Updated current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 4/4 no focus shift observed.

 

[homebodyMacro]

Periodic fishing for more data:

Updated current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 4/4 no focus shift observed.

Please comment if you have the RF 100 F/2.8 L Macro lens, what your firmware is, and if you've observed focus shift (if so, please post some samples)!!

 

[jrscls]

Periodic fishing for more data:

Updated current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 4/4 no focus shift observed.

Please comment if you have the RF 100 F/2.8 L Macro lens, what your firmware is, and if you've observed focus shift (if so, please post some samples)!!

Brand new RF 100 f/2.8 Macro here with Firmware 1.1.1 on a R5 Mark II and no focus shift observed using similiar tests described here on a tripod between f/2.8 and f/11.

 

[homebodyMacro]

Periodic fishing for more data:

Updated current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 4/4 no focus shift observed.

Please comment if you have the RF 100 F/2.8 L Macro lens, what your firmware is, and if you've observed focus shift (if so, please post some samples)!!

Brand new RF 100 f/2.8 Macro here with Firmware 1.1.1 on a R5 Mark II and no focus shift observed using similiar tests described here on a tripod between f/2.8 and f/11.

Awesome! Thanks very much for testing it out and sharing your results.

Updated current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 5/5 no focus shift observed.

...

Continued call for data:

Please comment if you have the RF 100 F/2.8 L Macro lens, what your firmware is, and if you've observed focus shift (if so, please post some samples)!!

 

[homebodyMacro]

Have posted in some recent threads about this: https://www.dpreview.com/forums/thread/4777500

and just now: https://www.dpreview.com/forums/thread/4785338

...

Always a great time for any other RF 100 macro owners to share their experience!

Current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 5/5 no focus shift observed.

...

Continued call for data:

Please comment if you have the RF 100 F/2.8 L Macro lens, what your firmware is, and if you've observed focus shift (if so, please post some samples)!!

Brute force data collecting is our only real option! Let's keep it going.

 

[homebodyMacro]

From the recent thread on this, R2 responded with some data: https://www.dpreview.com/forums/post/68028240

So far we are 5/5 in that topic for people with Firmware 1.1.1 (everyone who has posted with as having the lens) with no focus shift upon testing.

Make that 6/6 with v1.1.1

I just tested my RF 100L Macro on my R5 @ f/5.6 at several different distances from macro through 15 feet. No focus shift noted.

Which is interesting, as I did observe some focus shift from this lens when I tested it right after release. It's gone now.

As far as answering the OP's question, IME the RF version is the most capable, given the improved IS and slightly faster AF vs. the EF 100L Macro.

But the EF version is still a wonderful lens (I've shot extensively with all 4 of Canon's 100 Macros).

R2

My comments:

"Thank you for sharing!

That's very interesting. Perhaps there was a body firmware fix that accounted for the shift in the lens, bypassing the need for a lens firmware announcement? That's certainly a new angle/possibility.

I'll add your comment to my thread that I linked above"

...

To the enumeration of possibilities here, I think that's a significant addition. It's definitely not certain, but it is at least possible, along with a lens firmware update. A body update is an odd way to address a lens issue, but it's also a way to bypass a lens firmware announcement.

...

Current state: RF 100mm F/2.8 L Macro firmware 1.1.1: 6/6 no focus shift observed* (*R2D2 reported focus shift in the past that is no longer present).

...

Continued call for data:

Please comment if you have the RF 100 F/2.8 L Macro lens, what your firmware is, and if you've observed focus shift (if so, please post some samples)!!

 

[mvert]

I can't replicate this any more. My lens is from first batch delivered to Europe and when I got it, I tested focus shift and it was there. Camera and lens are same. Camera (R5) was on firmware v1.3.1, and now v.2.1.0. Lens says version v1.1.1 but I don't recall upgrading its firmware. My test images from 2021 and 2024 (minumum focus distance):

https://mve.kuvat.fi/kuvat/Testikuvia/RF100mm+F2.8L+Macro/

Edit: I had wrong lens FW version there first (v1.0.4). I don't know why my camera showed that yesterday when I wrote this post.