Canon EF-RF Adapter Video Autofocus Speed/Quality
- Thread starter ahalpert
- Start date
soarprod
Veteran
NorBro
Major Contributor
It's going to be great for many lenses, but native lenses will naturally work best as the glass technology (software and hardware) continues to improve and work better with the camera system.
Many old EF lens designed for stills only when video didn't really exist in stills cameras will most likely not work as well. (On a side note, this is a real shame with several older Fujifilm lenses.)
Many old EF lens designed for stills only when video didn't really exist in stills cameras will most likely not work as well. (On a side note, this is a real shame with several older Fujifilm lenses.)
ahalpert
Major Contributor
That is interesting - do you have any idea of a technological change in the autofocus system of EF lenses that I can refer to? Or maybe a production year that designs transitioned? Or do you think there are just continuous developments in that area and so the newer the better?
NorBro
Major Contributor
I read various articles and posts on forums over the years, but at this point you'd have to google/research if you're interested in finding anything specific.
However, it makes a lot of sense if you simply think about it.
In short, lenses are just like other older electronics, and they improved over time. Just like cameras. Just like AF technology (via different methods including artificial intelligence).
Initially, lenses weren't designed to consistently track moving objects in video - at least without hunting - because there was no video. Just 1 frame.
You could confirm this by using some older glass and seeing how badly they couldn't even focus in difficult stills capturing. Not only did they hunt but the focus was slow and the hit or miss ratio varied rather than a 99% hit with top-of-the-line equipment these days.
Plus, there are the physical parts; the motor and other mechanicality working slower or not engineered to complete a certain task more efficiently (like AI AF).
Or the coding...firmware updates weren't as common and therefore lenses weren't programmed as they are today. How many firmware updates do we have fixing or updating a product's performance every other month?
___
So when you consider all of that technical ideology, I think it makes sense there was an improvement (just like with most electronics...cameras, computers, etc).
Side note: You pointed out DXOMARK testing on another thread for something else, and if you ever check out EF glass throughout the years the differences in performance is astonishing with some lenses from 10-12-15 years ago vs. even the cheapest glass today (which is almost always sharper because of better glass, formulas, and a different market forcing better products for less money).
However, it makes a lot of sense if you simply think about it.
In short, lenses are just like other older electronics, and they improved over time. Just like cameras. Just like AF technology (via different methods including artificial intelligence).
Initially, lenses weren't designed to consistently track moving objects in video - at least without hunting - because there was no video. Just 1 frame.
You could confirm this by using some older glass and seeing how badly they couldn't even focus in difficult stills capturing. Not only did they hunt but the focus was slow and the hit or miss ratio varied rather than a 99% hit with top-of-the-line equipment these days.
Plus, there are the physical parts; the motor and other mechanicality working slower or not engineered to complete a certain task more efficiently (like AI AF).
Or the coding...firmware updates weren't as common and therefore lenses weren't programmed as they are today. How many firmware updates do we have fixing or updating a product's performance every other month?
___
So when you consider all of that technical ideology, I think it makes sense there was an improvement (just like with most electronics...cameras, computers, etc).
Side note: You pointed out DXOMARK testing on another thread for something else, and if you ever check out EF glass throughout the years the differences in performance is astonishing with some lenses from 10-12-15 years ago vs. even the cheapest glass today (which is almost always sharper because of better glass, formulas, and a different market forcing better products for less money).
ahalpert
Major Contributor
Interesting. What I have read most about AF (not something I ever worried about in the past) is the motor technologies and different methods of focusing the lens more or less quickly and silently. The problem is that manufacturers are not very clear about the technology they are employing, aside from marketing labels for different implementations. And each implementation touts the same benefits (speed, accuracy, silence), so you really have to read performance reviews.
In any case, I *think* that my EF lenses work well enough to focus on DPAF EF mount cameras for my purposes. So what I'm most interested in right now (since I'd rather not buy new lenses) is if I can achieve the same performance in an RF camera with an adapter.
In any case, I *think* that my EF lenses work well enough to focus on DPAF EF mount cameras for my purposes. So what I'm most interested in right now (since I'd rather not buy new lenses) is if I can achieve the same performance in an RF camera with an adapter.
NorBro
Major Contributor
Chances are good you won't notice a difference unless you have something from like 1995 that was $200. Not sure what you're thinking about using, but if you're talking about glass like the 70-200mm f/2.8L IS II, etc...you probably won't notice a difference for most of your work.
With that said, it's all relative. Contrast in the environment (like a strong sun behind your subject) can affect the best AF camera and lens combo in the world.
Personally, I'd think about the big picture; how many times in your life will you find yourself in an environment and subject(s) needing the absolute best AF? Even the most rudimentary performance (simple face-tracking from current technologies) is enough for most people and what they are filming.
With that said, it's all relative. Contrast in the environment (like a strong sun behind your subject) can affect the best AF camera and lens combo in the world.
Personally, I'd think about the big picture; how many times in your life will you find yourself in an environment and subject(s) needing the absolute best AF? Even the most rudimentary performance (simple face-tracking from current technologies) is enough for most people and what they are filming.
ahalpert
Major Contributor
Yes I have the 70-200 2.8 IS II, as well as the original 24-70 2.8, the Tokina 11-16 2.8 (which fills FF at 14mm and longer), and a Tamron 35mm 1.8 IS. All my other lenses are manual focus or APS-C.
Man how great would it be if they also implemented a S35 crop mode on the R5 with 4K 60P RAW capture. They won't do that, though.
cpreston
Veteran
All of the RF lenses and the RF mount have been designed for continuous autofocus with DPAF. The system is supposedly faster and allows more options for speed. I would assume that RF lenses would generally be better than EF lenses for AF. In regards to EF lenses, I have never noticed any difference in performance when switching between my EOS R and a regular EF camera. Some lenses aren't so great for AF on the R, but they weren't great on my EF cameras either.
hasselblad
Well-known member
It makes sense! Thanks so much for you input!
cookinghusband
Well-known member
I will say it perform very well on the EOS-R, no issue at all.