Canon is dead to me after they forced 3rd parties to stop all RF lenses. I bought the last AF lens by Viltrox and it's fantastic. I regret getting an RF body after they threatened them, and now only allow 3rd parties on their crop sensor bodies. Fuji, Sony, Leica, Nikon, all allow 3rd party lenses. I switched to Fuji and won't go back.
I'm feeling a little bit of buyer's remorse for that same reason. I love my Canon, but having a closed loop lens system is not ideal and is insanely pricey. All of the "adapters" that allow you to use 3rd party lenses suck.
It is a little sad honestly how Canon has now been surpassed by Sony in both sensor technology and lens optical quality and yet they double down on forcing people to use their expensive lenses. I miss the old days when EF mount was the lingua franca of autofocus mounts with a thriving ecosystem of lenses from Sigma, Tamron, and the like. Even now some of the Sigma Art lenses for SLR mounts, like the 40mm and 105mm f/1.4, are peerless in terms of image quality.
Also not pleased that third party lens vendors were excluded. However the way I look at this is perhaps we would have no Canon cameras at all if the revenue they derive from lenses had not been captured.
Wait, did they reverse their stance on this again? I thought they had finally started to open the mount up to 3rd parties. Really disappointing to hear that this is not actually the case.
I'm in the same boat. I wouldn't be as upset if the Canon range was uniformly excellent. But every one of their RF ultra-wide zooms suffers from softness or extreme vignetting. I sent back a $2200 RF 15-35mm, and for landscapes I now just use my iPhone Pro.
They weren't dead to me immediately just because there are still third party RF lenses that rock, but when Sirui did the AF anamorphic lenses and I couldn't get one for my Canon then it really bummed me out. :/
Nikon aren't doing much better on letting third-party lenses into the ecosystem, sadly - speaking as someone with three Nikon DSLRs and Sigma glass I'm very happy with.
Wait, what? That’s horrific! I love my range of lenses for my long-suffering 5d3. I wouldn’t have bought it if the only options were the canon lenses. How disappointing to hear.
I used to have to manually focus to take photos of my dog. The apparent noise from the early Ring USM AF in the 85mm/1.8 lens drove him mad and he would run out of the room.
We did initially wonder if it was some psychological effect of pointing a camera at him, but one night I hid behind a curtain and used the AF motor and he still shot out of the room.
Subsequent USM AF motors were silent to him and he was then content to have his photo taken.
My dog also developed an aversion to cameras, and I wondered if it was the USM focusing. That might have been what started it, but she still doesn't like any cameras at all, USM or not.
In case anyone else was looking for it, this article from the same author covers the more algorithm-y question of how a camera body decides to actually use these motors when you press the "focus" button: https://exclusivearchitecture.com/03-technical-articles-DSLR...
I can't find any articles on how a camera body decides what is and isn't a desired subject. I'm guessing there's some amount of machine learning-type stuff involved in that, seeing as how they can detect human (and bird?) faces?
Nowadays they use pre-trained pattern recognition AI models, yes, which has become much more impressive (and CPU-intensive) with mirrorless cameras where the entire resolution of the main image sensor is available for analyzing the scene. Some higher-end traditional DSLRs have a "high"-resolution (around 0.1 MPix or so) metering sensor that is used to assist the AF system (eg. what Canon calls iSA and iTR [1]).
Traditionally, cameras would just focus using the single focus point the photographer has selected, or if they have selected a larger area focusing mode, the camera would typically pick the closest point of a group of points, assuming that that's usually what the photographer is interested in. (Remember that traditional (D)SLRs have a discrete AF sensor with at most a few dozen focusing points to choose from!)
In tracking AF modes (eg. Canon's Servo AF), depending on settings, the camera tries to avoid sudden shifts in focus even if a foreground object momentarily occludes the original target. Tracking AF also has to predict the subject's motion to prevent the focus from lagging behind a fast-moving subject. Higher-end cameras allow configuring the AF behavior in terms of how reactive vs "sticky" it should be when tracking a subject, and how linear the subject's motion is expected to be.
With mirrorless cameras the focus switched from specialized sensors to on-CMOS contiuous exposure sensors, so movement is easy to detect. At this point the cameras have specialized AI hardware to run the models, and they also accept user input (on R5 MkII you can register up to ten people to prioritize focus on[1]). The focusing options are now very complex[2][3], and combined with lots of customization options on the camera's buttons you can have very specialized/personalized setups for different types of photography.
Sony A9 III even has a configuration setting of whether it should focus on the left or right eye of the person :) It also can remember faces and prioritize them if there other faces. Let's say someone shoots their kid on the football field and wants only them to be in focus.
I have been aggravated by (and bellyaching about) the ridiculous lack of a way to control the focusing motors in these lenses directly when shooting video, using a follow-focus wheel.
Video shooters are still strapping janky gear-tooth strips onto these lenses, and then bolting bulky mechanical follow-focus mechanisms on the OUTSIDE of lenses that have focusing motors already built in.
I can only imagine that manufacturers refuse to make direct control available in order to protect their "cinema" lens lines, where a manual normal lens sells for thousands of dollars. And yet they sell some falsely-named "hybrid" lenses that are supposedly somehow better for video... despite lacking even geared focusing rings that are compatible with follow-focus units... let alone a control port that could be used for a focusing wheel to use the autofocus motors.
I looked at the Canon camera-control SDK, and sure enough... focus control is omitted from their entire line, except for two PTZ cameras that aren't suitable for cine use.
A side project I've been neglecting is controlling autofocus on Sony Alpha cameras (specifically the A7ii and A7iii which I have) for the purpose of focus-stacking: shooting a few dozen shots with minimal-step focal differences and blending them all in photoshop later.
It took me a while to find the API; Sony made one then pretty much scrubbed it from the internet and has ignored my emails asking for one :-)
But even with the API, achieving results are tricky. Whether or not the focus motor steps properly seems to be based on how busy the camera is at the time... I'm trying various combinations of [shoot] - [sleep] - [step the focus] - [sleep] - [shoot] but the exactly which of these steps actually succeeds just seems so damn random.
BlackMagic does some pretty innovative stuff. They offered an Arduino add-on board that could communicate with and control their cameras over SDI, including access to the Canon EF mount and its commands.
Somebody did in fact create a control board with a knob to control lens focus through this thing. It can be done. The manufacturers just refuse to do it.
I've always wanted to know how the various autofocus systems worked. This page is incredible. I wish there was another one with the Nikon autofocus systems since that's what I actually have.
I still have several lenses with autofocus that don't have an AF motor in it at all, the motor is in the camera body instead there's a tiny screw on the lens mount that transfers the motor rotation to the autofocus parts in the lense. This was very slow and noisy though on my cameras.
Whereas I think "no metering" is never a problem on Canon.
Over the decades as a DSLR shooter (on Canon), I even saw Nikon shoot themselves in the foot not once but twice.
* When Nikon introduced "E" lenses (electronic aperture, like Canon and all modern designs), very few bodies were compatible with it. They didn't have the foresight to introduce compatible bodies before any lenses with E were released. For example, this was released in 2008 ( https://www.kenrockwell.com/nikon/24mm-pc.htm ), and the earliest compatible body is from 2007. It doesn't work with tons of crop (DX) bodies like the D90, as well as any film camera.
* When Nikon introduced "AF-P" (stepper motor) lenses in 2016, there were no compatible bodies before 2013. Again, not enough future-proofing.
Meanwhile, Canon took a different approach. They pissed off users twice - when transition from the FL mount (1964) to FD (1971), and FD to EF (1987). They basically got it all right with EF - fully electronic, no aperture slider, no focus screw, big diameter.
I have even successfully used a year-2017 Canon lens with Ring USM AF and IS on an original EOS 650 film body (released in 1987), and both AF and IS work perfectly. Mind you, both of those features did not even exist at the birth of EOS. Presumably the AF electronic protocol is agnostic of what motor technology is in the lens, and IS can be a lens-only thing without the body knowing.
Nikon's compatibility gotchas don't hold a candle to Canon's excellent (albeit imperfect) compatibility. Nikon keeps making the same mistakes over and over again, such as not having a screw-drive motor in the F to Z mount adapter; the company really seems to hate people with old lenses and bodies. Oh and this is not new either; Nikon's F to 1 adapter (does anyone remember that highly cropped mirrorless camera?) also lacked certain features.
This third party company announced a Nikon F to Z lens adapter that will autofocus with older screwdriver autofocus lenses. They already had a Nikon F to Sony adapter and now they have a Nikon F to Nikon Z adapter.
What annoys me is that the AF-P could probably be supported on a lot more cameras, they did add it to the D3300 with a firmware update. They just didn't do it for more cameras...
Newer Nikon lenses have a few motor types but if I'm not mistaken they support old mechanical (body driven) aperture and focus drive as well (the D6 and somewhat recently F6 film camera were still for sale)
Nikon has a complex ecosystem keeping generally the same mount for many decades and general support for it, where others developed a specific autofocus mount when they moved towards those systems.
Yeah the in-body motor thing is only for the older Nikon lenses, I don't think AF for that is still supported if you're using Z-mount adapters. Nikon seems to have a few similar AF motor technologies like the canon ones in this article.
Cool videos, thanks! The driver shown in the first video is indeed a piezoelectric actuator, but not the type Canon uses in their lenses. The driver shown in the second video is probably related to Canon's Nano USM.
As far as I know, Canon controls a large part of the supply chain and has numerous factories to manufacture quality-critical parts (especially their lenses, zoom mechanisms, and barrels) themselves. But there are numerous components that are more efficient to obtain from third parties, such as microchips. I am not sure about the motors and actuators, though.
A note of possible interest for this crowd: all these motor types are open loop, so you can't actually command a specific state with any accuracy / repeatability. In other words, if you send a signal to the lens to focus at, say, 3m, it won't necessarily be in exactly the same state as the next time you focus at 3m.
This is why camera calibration can be tricky, and you often don't want to touch the cameras after you've done a calibration pass.
It’s open loop in that the measurement of it being in focus is reliant on the subject matter, and a different measurement.
I’m not making this up, camera manufacturers have told me to my face that focus is open loop, period. They can’t guarantee repeatable focus.
Notably the measurement isn’t of the state of the motor/gearing. Furthermore, being “in focus” means the subject matter’s out of focus blur is below some threshold; there is a range of focus states that qualifies — but those seemingly small differences can affect camera calibration, with >pixel-level differences in effective focal lengths.
Correct. The AF sensor continually takes more samples even while the motor is moving. You can tell because sometimes the motor overshoots and then you can see it come back.
Readit News
I was hoping for the new Sigma BF camera, simple and elegant. Unfortunately no Bluetooth, which is a deal breaker for me.
Dead Comment
We did initially wonder if it was some psychological effect of pointing a camera at him, but one night I hid behind a curtain and used the AF motor and he still shot out of the room.
Subsequent USM AF motors were silent to him and he was then content to have his photo taken.
Some animals just know and some like being the center of such attention. Others just don’t.
In case anyone else was looking for it, this article from the same author covers the more algorithm-y question of how a camera body decides to actually use these motors when you press the "focus" button: https://exclusivearchitecture.com/03-technical-articles-DSLR...
I can't find any articles on how a camera body decides what is and isn't a desired subject. I'm guessing there's some amount of machine learning-type stuff involved in that, seeing as how they can detect human (and bird?) faces?
Traditionally, cameras would just focus using the single focus point the photographer has selected, or if they have selected a larger area focusing mode, the camera would typically pick the closest point of a group of points, assuming that that's usually what the photographer is interested in. (Remember that traditional (D)SLRs have a discrete AF sensor with at most a few dozen focusing points to choose from!)
In tracking AF modes (eg. Canon's Servo AF), depending on settings, the camera tries to avoid sudden shifts in focus even if a foreground object momentarily occludes the original target. Tracking AF also has to predict the subject's motion to prevent the focus from lagging behind a fast-moving subject. Higher-end cameras allow configuring the AF behavior in terms of how reactive vs "sticky" it should be when tracking a subject, and how linear the subject's motion is expected to be.
[1] https://www.canon.com.hk/cpx/en/technical/pa_Overview_of_65-...
[1] https://cam.start.canon/en/C017/manual/html/UG-04_AF-Drive_0... [2] https://cam.start.canon/en/C017/manual/html/UG-04_AF-Drive_0... [3] https://cam.start.canon/en/C017/manual/html/UG-04_AF-Drive_0...
I have no idea how this worked, but would have loved to see the photos they used for “training” this system 25 years ago.
Edit: Wow - there's a whole collection of Canon lens technology articles there: https://exclusivearchitecture.com/03-technical-articles-CLT-...
I have been aggravated by (and bellyaching about) the ridiculous lack of a way to control the focusing motors in these lenses directly when shooting video, using a follow-focus wheel.
Video shooters are still strapping janky gear-tooth strips onto these lenses, and then bolting bulky mechanical follow-focus mechanisms on the OUTSIDE of lenses that have focusing motors already built in.
I can only imagine that manufacturers refuse to make direct control available in order to protect their "cinema" lens lines, where a manual normal lens sells for thousands of dollars. And yet they sell some falsely-named "hybrid" lenses that are supposedly somehow better for video... despite lacking even geared focusing rings that are compatible with follow-focus units... let alone a control port that could be used for a focusing wheel to use the autofocus motors.
I looked at the Canon camera-control SDK, and sure enough... focus control is omitted from their entire line, except for two PTZ cameras that aren't suitable for cine use.
https://www.nikon.com/company/news/2025/0213_imaging_02.html
It took me a while to find the API; Sony made one then pretty much scrubbed it from the internet and has ignored my emails asking for one :-)
But even with the API, achieving results are tricky. Whether or not the focus motor steps properly seems to be based on how busy the camera is at the time... I'm trying various combinations of [shoot] - [sleep] - [step the focus] - [sleep] - [shoot] but the exactly which of these steps actually succeeds just seems so damn random.
Somebody did in fact create a control board with a knob to control lens focus through this thing. It can be done. The manufacturers just refuse to do it.
I still have several lenses with autofocus that don't have an AF motor in it at all, the motor is in the camera body instead there's a tiny screw on the lens mount that transfers the motor rotation to the autofocus parts in the lense. This was very slow and noisy though on my cameras.
Whereas I think "no metering" is never a problem on Canon.
Over the decades as a DSLR shooter (on Canon), I even saw Nikon shoot themselves in the foot not once but twice.
* When Nikon introduced "E" lenses (electronic aperture, like Canon and all modern designs), very few bodies were compatible with it. They didn't have the foresight to introduce compatible bodies before any lenses with E were released. For example, this was released in 2008 ( https://www.kenrockwell.com/nikon/24mm-pc.htm ), and the earliest compatible body is from 2007. It doesn't work with tons of crop (DX) bodies like the D90, as well as any film camera.
* When Nikon introduced "AF-P" (stepper motor) lenses in 2016, there were no compatible bodies before 2013. Again, not enough future-proofing.
Meanwhile, Canon took a different approach. They pissed off users twice - when transition from the FL mount (1964) to FD (1971), and FD to EF (1987). They basically got it all right with EF - fully electronic, no aperture slider, no focus screw, big diameter.
I have even successfully used a year-2017 Canon lens with Ring USM AF and IS on an original EOS 650 film body (released in 1987), and both AF and IS work perfectly. Mind you, both of those features did not even exist at the birth of EOS. Presumably the AF electronic protocol is agnostic of what motor technology is in the lens, and IS can be a lens-only thing without the body knowing.
Nikon's compatibility gotchas don't hold a candle to Canon's excellent (albeit imperfect) compatibility. Nikon keeps making the same mistakes over and over again, such as not having a screw-drive motor in the F to Z mount adapter; the company really seems to hate people with old lenses and bodies. Oh and this is not new either; Nikon's F to 1 adapter (does anyone remember that highly cropped mirrorless camera?) also lacked certain features.
https://nikonrumors.com/2025/02/28/monsteradapter-la-fz1-off...
I've been using my old 85mm f1.4 "D screwdriver" lens and others in manual focus mode on my Z8 and Z5.
Wave motor = AF-S (1998)
Stepper motor = AF-P (2016)
Voice coil motor = SSVCM (2022)
https://imaging.nikon.com/imaging/lineup/lens/glossary/
Nikon has a complex ecosystem keeping generally the same mount for many decades and general support for it, where others developed a specific autofocus mount when they moved towards those systems.
Here's a video showing the mechanism:
https://www.youtube.com/watch?v=7iHL4ZCkCKc
And here's a video showing such a motor at full speed:
https://www.youtube.com/watch?v=JtttNnmCVmU
This is why camera calibration can be tricky, and you often don't want to touch the cameras after you've done a calibration pass.
I’m not making this up, camera manufacturers have told me to my face that focus is open loop, period. They can’t guarantee repeatable focus.
Notably the measurement isn’t of the state of the motor/gearing. Furthermore, being “in focus” means the subject matter’s out of focus blur is below some threshold; there is a range of focus states that qualifies — but those seemingly small differences can affect camera calibration, with >pixel-level differences in effective focal lengths.
It’s open loop.