Fujifilm’s autofocus will never be as good as Canon, Sony, or Nikon’s, and that’s ok. I’ve explained this before a couple of times, so forgive me for repeating myself. There are some highly unrealistic expectations that need to be put to rest, which I hope this article accomplishes.
There are a lot of comments floating around the internet along the lines of, “Fujifilm’s autofocus sucks.” And, “Fujifilm needs class leading autofocus like Sony and Canon.” And, “If Fujifilm doesn’t address AF on the next generation, I’m leaving.” And many other similar sentiments. But these comments are out of touch with reality. Let me explain why Fujifilm’s autofocus will never be as good as the Canikony brands, and why it doesn’t matter.
There are three reasons why Fujifilm’s AF isn’t as good as the three big brands. First, those companies have been making autofocus systems for much longer. Sony (through Konica and Minolta) have been making AF since the 1970’s—they were the first (via Konica), and the first to have what was considered a “good” AF system (via Minolta). Canon and Nikon have been developing AF since the 1980’s. Since they had such a big head start, it is only logical that they’re further along. Second, the Canikony brands have more R&D funds to commit to autofocus development and improvements. They have larger teams with larger budgets, and it’s probably easier for them to attract the most experienced talent. Last but not least—and this was pointed out by PetaPixel—the big three have put most of their effort over the last handful of years into AF, and have largely ignored image quality improvements; in some cases, autofocus was prioritized to the detriment of image quality. That is a path I would highly discourage Fujifilm from pursuing, personally.
While Canon, Sony, and Nikon have put a lot of their eggs in the AF basket, we’ve long surpassed the point of diminishing returns—the Inverted U Curve, for those who have read Malcolm Gladwell’s David & Goliath (a book I definitely recommend). How good does AF really need to be? At what point is it good enough for almost everyone? I think we passed that point years ago. For almost the entire history of photography and videography, there was no autofocus. It was manual focus only. Yet today’s photos and videos aren’t necessarily better than they were then (in some ways, they might be subjectively worse). If we compare Minolta’s first “good” AF to the AF of any camera made in the last 25 years, it’s clearly not as good as even the worst examples you could find. In other words, we’re really spoiled today with incredible gear. Fujifilm’s so-called “sucky” AF is better than the best AF from 15 years ago, maybe even 10. Were we incapable of creating good photos and videos then? Are photos and videos captured today using the latest Canikony cameras better than anything previously? Of course not! Some people seem to think that photography has only been around for a few years.
It was never about the gear, it has always been about the one using the gear. Yes, some tools make it a bit easier to achieve your desired results, but if you really want a certain result, you can do it no matter your camera. You might have to try a little harder, you might have to learn a new skill, or you might have to practice a skill you’ve gotten rusty at, but you can still do it. My guess, though, is that for 98% of people, Fujifilm’s AF is more than good enough just as it is, as was Canikony’s five or ten or even 15 years ago. So does it actually matter that Fujifilm’s autofocus isn’t as good as the three big brands? It might matter to some, but it shouldn’t matter to the vast majority. If you can’t get the job done with the gear you have, the problem isn’t the gear; however, that’s a bitter pill that people don’t want to swallow. Nobody wants to hear that they’re the problem, but without introspection there’s no opportunity for growth.
With all of that said, Fujifilm has done an amazing job creating and improving the autofocus on their cameras, all things considered (aside from that infamous firmware bug last year). With a smaller budget and smaller team (and with a lot less time), they’re not terribly far behind Nikon. I’m sure they will continue to make strides, and—who knows—with AI they might have some breakthroughs that would be difficult otherwise, and which might level the playing field a bit. For most people, these improvements will have little-to-no practical benefit because the AF is already more than sufficient—we’re near the top of the upside-down U. Seriously, how good does it need to be? It’s been more than good enough for me since X-Trans III, and even the older models are decent in daylight situations. Only in extreme cases do the Canikony brands have a clear advantage, yet even those situations aren’t insurmountable if the one behind the camera doesn’t allow them to be. While I’m sure Fujifilm’s autofocus will continue to improve, the difference it will make for most people is very small. Yes, we’re rooting for Fujifilm to be at the head-of-the-pack in every aspect of camera-making, but it’s important to keep expectations realistic, or else we’ll inevitable be disappointed, which will lead to unnecessary resentment—something that’s easy to spot all over the internet.
Hey Ritchie, well done. I agree with you. I just have one thing to point out. I have no problem with Fuji AF. I have two cameras and 6 lenses…no problem with AF for what I do… I was at B&H recently and just curiously stopped at the Olympus Booth…I picked up one of the top cameras and could not believe the speed of the AF…I was blown away…very impressive speed… practically instant. Even seeing that, I still don’t see myself leaving Fuji now…Just thought I’d throw that out there…thank you for your insight and hard work.
I don’t have personal experience with Olympus (other than a camera I had to return many years ago because it was broken out of the box—that was a long time ago). Was it a stacked sensor model, like the OM-3? A stacked sensor can help to speed up AF because it allows for faster processing. For example, the X-H2s has faster AF than the X-H2.
I concur fully with you Ritchie!
10 years ago when switching from my Nikon F6 and D3S to the Fuji X-T1, I did stopped using AF-C as default as the X-T1 was really not at the sale level.
But AF-S was more than good enough for me. I would have not switched if not so. Now I have no problem with Fuji AF for I’m doing. It simply delivers as I want, even in low light.
I have 9 Fuji cameras (10 when I’ll receive the X-E5) and more than 30 Fuji lenses, not counting third-parties.
But that saying say, when I’m picking up the Nikon Z9 or Zf, I really appreciate their AF soeed.
Amazing compared to my Fujis. And I would wish to get that in my cameras.
But AF is definitely not the focus (sic !), many others things counts. Especially the film simulation, the image quality (you are more than right, it has been forgotten by the other brands), versatility, dials, huge customisation, quality of the lenses, third-parties option, reduce size for the X system and with all that in consideration the actual price ratio out of it.
Hence I would trade my Fujis against something else.
For those in need of outstanding AF performance, I can understand they come and go from Fujifilm or simply stick to Canikony. And each brand as its avantages and defaults. It is also good for diversity and competition.
But like shooting RAW vs JPEG or Full Frame vs APS-C, it would be great that the shit storm stops at last and people move on to something else that really matter. When things are badly made say so, but do not make it a rule / dogma. That’s not helping.
Big typo 🙀 (still difficult to write with my right hand being impaired 😓🤕 and no photography possible for the next month still 😭) :
“Hence I would **NOT** trade my Fujis against *anything* else!” 😉
But Iwill picking up every goodies and AF code updated that van bring Fuji in firmware updates. Even move up on models because if it.
Now I’m a bit for the first time circonspect to update to latest firmware versions of the X-T5, X100VI and RF…
The omitted word does make a difference, but is completely understandable with your injury. Hopefully the recovery is going well!
Can Fujifilm’s AF be improved? Yes. Will it be improved? Yes. Is it already pretty good? Yes. Does having incredible AF speed and accuracy make a difference for some? Yes.
But, for most people, it’s already more than good enough (and has been for some time now, no matter the brand), so any improvements are generally speaking gee-whiz and bragging rights. But they are certainly welcomed by those who do, in fact, need it. I think the number of those who actually need “class leading AF”, though, has been greatly exaggerated.
I appreciate the input!
The omitted word does indeed make a difference! Thanks for my recovery. Reeducation has just started and it will several weeks.
The light weight combined with excellent image quality and build quality is the reason I switched from Sony and Canon to Fujifilm three years ago. I still have a Canon, but to be honest, I haven’t used it in three years. The build quality of the Canon feels plasticky compared to Fujifilm. It feels cheap, not premium like Fujifilm. And the few APS-C lenses Canon has for APS-C cameras are, to say the least, junk in terms of image quality. So what do I care about AF? And even with the Sony, the brilliant AF hasn’t prevented me from taking out-of-focus photos in the past. And when I get out-of-focus photos with my Fujifilm, it’s usually my fault.
I have a Sony A7 IV that is just collecting dust. I need to sell it, as I don’t use it and haven’t in the last two years—it’s practically brand-new. I anticipated the AF blowing me away, based on everyone’s comments about Sony’s “amazing” AF. But it made no difference to me, it was the same as Fujifilm’s. I’m sure with certain lenses and in certain use-cases, there is a noticeable difference, but we’re getting really nitpicky, I think, when the often barely perceivable (if perceivable at all) difference is the line between “sucks” and “class leading”. It just makes no sense, other than trolls stirring the pot and click-bait content for views.
But as you mentioned, everything has its pluses and minuses. Those who sold their Sony or Canon gear to move to Fujifilm (who are now complaining about the AF) did so for a reason. What was wrong with the system that they were in? Obviously something, and perhaps—like you—it was build quality or image quality, or it could have been something else. For me, it was the traditional tactile controls. But no camera is perfect, not even Fujifilm; however, every camera is pretty amazing nowadays.
So true. Technology has come so far. How much more could we really need? How much faster, sharper, smaller, more MP? If you listen to the voices in forums one could think that camera standards are still not on par with today’s requirements.
It’s always been like that and will probably always be like that. It’s good for sales of the latest gear. As long as we keep that in mind we’ll be fine.
We need to know what’s most important to us and act on that. And then go and take awesome jpegs ❤️
“We need to know what’s most important to us and act on that. And then go and take awesome jpegs ❤️”
Great reminder! Thanks!
1. Gear: Photography has far too much focus on gear, comparing spec sheets and ‘my gear is better’ syndrome. (My car, my tv, my xxx, etc.)
2. Youth: if you’re, say 21 now, you were 11 ten years ago. Tech from then is ancient history if you’re young.
3. Film: if we take film photography as a benchmark then all digital camera brands reached that benchmark a good few years ago now, in iq and other measures. (Digital photography had to catch up with film.) Anything else is a bonus that we don’t really need.
Can’t agree more with your comments @Taigen especially on point 3. coming myself from film.
I think those are all excellent points.
Thank you Ritchie for the time you invest to make this research to inform us all. We should not forget that if people are ignorant about the history of camera development, it’s easy to critisize technical aspects of today’s cameras. Being creative taking photographs is more important than nitpicking about AF speeds in MHO…
I agree with your humble opinion. As Ansel Adams famously said, “There’s nothing worse than a sharpe image of a fuzzy concept.”
This discussion reminds me of another Ansel Adams maxim which I try to keep telling myself: “The single most important component of a camera is the twelve inches behind it.”
Yes, so true! One of Adam’s best pieces of advice, which is unfortunately ignored far too often nowadays.
I haven’t done much wildlife photography over the last year and that was where fast AF counts most the big issue was that sometimes the subject was close say 15-20 feet and the camera would focus on the background say 30-40-50 feet behind the subject and get stuck there. Of course the subject in the foreground was maybe invisible due to the blur (at 400mm). Once this happened there was nothing you could do to get the lens to hunt, seems to me half pressing the shutter should cause the camera to hunt again and half pressing the shutter several times in quick succession should be a sign that the background is not the subject but nothing worked except to aim the camera at leaves or grass closer to get the focus in the ballpark and try again, this is annoying. Anyway that seems fixed now or much better last time I tried unless I just got lucky. As a software guy that has worked in embedded systems all me life I feel Fuji should have done better, I realize AF is a hard nut to crack and I thing getting the right approach is probably the toughest problem and if you don’t have the right approach no amount of tweaking will get you to the top of the heap. AI is another variable and resets the game although AI is slow so again you need the right approach to get this to work at peak performance
The size of the focus box was probably the issue. If you make the box smaller, it is less susceptible to that kind of problem. Pal2Tech has a video about this specifically.
Of course, if we’re talking about a year ago, that could also relate to the AF firmware bug that plagued a lot of people around that time.
I think probably the biggest difference between Canikony brands AF and Fujifilm’s is that Fujifilm’s is less intuitive. Canikony cameras tend to perform well out of the box (they don’t expect their users to fine-tune AF for their specific needs, it is automatic), while Fujifilm expects their users to dig into the menu and set it up for their specific use cases. This requires steps by the photographer (who may not even be aware that steps should be taken), and Fujifilm doesn’t make it clear how one should set up AF on their cameras for their specific needs. Pal2Tech has some helpful videos, as does (surprisingly) Camera Conspiracies (who recently figured out that many of his AF issues were actually user errors). Once you figure out Fujifilm’s AF, it’s pretty much as good as anyone else’s (not 100%, but it’s pretty close), and certainly more than good enough for 98-99% of photographers (and even that 1%-2% could make it work for them if they wanted).
Fujifilm af is okay until you shoot with a canikony camera. My XT4 stays in single point af-s at almost all the time and it does the job for me. But it’s impressive what other brands offer for the price. For sub-thousand dollar cameras fujifilm af is decent but after Fuji’s latest price bracket, I feel like we deserve better tech.
At the same time, all the other brands are way behind when it comes to jpeg quality and customization. So it’s more about what do you care when you select your next camera.
I have shot Canikony cameras. In fact, I have a Sony A7 IV not more than six feet from me right now as I type this. I have not experienced any major difference in AF personally (it’s about the same to me as my X-T5); however, I freely admit and fully understand that my use-cases are not the same as everyone else’s, and Canikony (particularly Sony and Canon) are objectively better in certain situations. My guess is that my use-cases are probably very similar to most people’s, and for the majority there is not a practical difference; only a relatively small group really “needs” it and routinely takes advantage of it, and for everyone else it’s more gee-whiz and bragging rights on internet forums than anything else.
As far as what tech we “deserve” for the price, that’s for everyone to determine on their own. What I would say, though, is that all cameras are becoming more expensive, not just Fujifilm. I don’t know what you are specifically thinking of, but I can’t think of any good apples-to-apples comparisons, as each has advantages and disadvantages, and Fujifilm seems to offer a lot of great advantages for their price points. Even the Nikon Zfc vs Fujifilm X-T30 II and (especially) the Nikon Zf vs X-T5 is a bit apples-to-oranges.
As far as your second paragraph, I agree completely. Some brands are further behind than others, but Fujifilm is clearly in the lead when it comes to camera-made JPEGs. I suspect that is one of the leading sales drivers for Fujifilm right now.
“Fujifilm’s autofocus will never be as good as Canon, Sony, or Nikon’s”
There’s no reason for this to be true. There’s no reason fuji can’t have as good autofocus as the other companies. Literally none.
And when they charge as much or more than the others while advertising “class leading autofocus” in some cases then they should endeavour to follow through and try to at least match their competitors. And don’t say they aren’t competitors, because all three (canon, Nikon, Sony) sell apsc cameras in similar price ranges that offer similar features, and in the case of Nikon even similar form factors with film simulation capabilities too.
They are direct competitors, in the same class, and fuji falls behind in this aspect.
No, the autofocus isn’t shit. Yes, it isn’t as good as the competition. Yes, they should improve it. There’s no reason for them not to work on it. If you think it’s good enough as is, good for you, and even better for you if it improves. If it’s not good enough for you, then yes making it better would also be a good thing. Improving the autofocus hurts no-one, benefits everyone, and makes fujifilm a more competitive option in the market.
There’s no reason for it to be true other than the three really good, logical, and factual reasons that I gave in the article. I noticed that you didn’t attempt to refute them, only ignore them. So, yeah, there are literally three good reasons, and not none as you suggest. Literally three—tres buenas razones.
And that illustrates my point about unrealistic expectations. Sure, if Elon Musk purchased Fujifilm, he has the money and resources to make Fujifilm’s AF class leading, and it could probably be done within a year or less. But that’s not going to happen. So Fujifilm would need to invest quite heavily into the resources necessary to overcome their disadvantages, and that money is where? Fujifilm doesn’t have Sony money. Fujifilm doesn’t have Canon money. Fujifilm might have Nikon money after Nikon spent 85 million to buy RED (which seems like a great bargain, btw), but Fujifilm doesn’t have Nikon money pre-RED, and possibly not post-RED, either. It’s just so fantasyland to believe that Fujifilm has the resources available to overcome the giant obstacles of: being late to the AF party, having a smaller, less experienced team, and having a significantly smaller annual R&D budget.
I would be interested to know the specific “in the same class” cameras that you are comparing. I bet they are not in the same class after all, if we knew which ones you are talking about specifically.
I don’t understand your last paragraph—it’s kind of all over the place, and contradicts itself at times. Fujifilm is working to improve their AF, and have been for 15 years now. I think you are expecting miracles. They’re doing the best they can with the resources they have. It’s quite amazing what they have accomplished, when you give it a little consideration. Perfect? No (not one camera is). Room to improve? Sure. Plenty good enough for most? Heck, yes.
Fuji simply does not need to be spending as much of their R&D time as they currently are on making and releasing new cameras. They simply don’t. They could slow down and stop releasing new cameras until they focus in and catch up with their autofocus. Them having a smaller R&D team would matter way less if they dedicated their whole team to this. Instead they’re so weirdly focused on releasing a new camera every week even though they have products that consistently bring in money to the company. And yet they spend so much of their R&D time on making cameras like the X half and figuring out a way to hide the film sim dial behind a window? Nah man, take a lesson from Ubisoft with the assassin’s creed games. They stopped churning one out every year, and the games massively improved from it because instead of trying to make a whole new game (or camera) every year, they could focus on the development things that would make each game (or camera) way better.
Compare them to any other brand (who, like you said, have way bigger r&d budgets) and look how many cameras they make a year.
Fuji has released 4 cameras THIS YEAR and we’re only 7 months in.
Sony? 1
Canon? 1
Nikon? 2
It’s needless.
We’re obviously not going to agree here, but please at least consider that fuji could slow down a touch and focus their small r&d budget to improving something a decent chuck of the community is asking for, instead of just making more and more cameras.
In different years, different manufacturers release different numbers of cameras.
While I do agree that cameras in-general should have longer production runs—five-to-seven years should be the normal, with some models going longer—no camera maker is routinely doing that nowadays. Fujifilm would sink the camera business if they were still selling X-Trans III models in 2025. And even if they were, that’s not likely going to equal any AF improvements beyond what they are already doing.
I just think your expectations are really out there, and not based on reality. That’s why you are so upset. Unrealized expectations lead to anger and resentment. Keep your expectations at a more realistic level, and you’ll find yourself a little happier.
Another point I’d like to bring up is that Fujifilm’s AF is better than you are giving it credit for. If you are having trouble, here’s one (of many) resources that could potentially help you get more out of it:
https://digitalcamera-support-en.fujifilm.com/digitalcameraengpcdetail?aid=000008362
Thanks this is a great article, Ritchie.
I’m in the 2%, as a wildlife photographer. I can say that the latest version of firmware is very, very good. Could it be better, definitely.
But here’s the thing….I moved from Nikon 2 years ago having used Nikon cameras for 20+ years. The reasons
-Weight – getting older I just can’t afford to carry heavier gear, my setup is now 20-30% lighter with greater capability
– Price – Nikon has done away with the bottom end of its range and now charges a significant premium, albeit for good gear. My setup cost me approx 40% less than equivalent Nikon gear
So maybe the AF isn’t as good as Nikon/Canon/Sony BUT for the reasons above it works for me.
And that’s before you talk about the other significant benefits of Fuji….image quality, wonderful jpegs straight out of camera etc….
So you’ve got to look at it in the round!
Best wishes
Peter
I agree. There’s a big picture, and people like to focus (pun intended) on one small section of the whole and ignore all the rest. Thanks for the comment!
Although I agree with the sentiment that if it isn’t on par with the crazy good AF of like the Sony cameras, that should not need to be a problem… I have to say the AF on my X100VI is extremely disappointing. I have situattions where I take a head shot, of a person sitting still 1 meter away from the camera, camera stationary, face recognition on, focus area manually put on the subjects head… it says face detected… and still totally out of focus!!! This happens a lot. Walking the streets, hunting… over 60% is out of focus … completely… often with a face being detected. And yes… I tried everything people suggested. Fujis design also doesn’t help… like the range limiter sounds nice but not if it doesn’t work together with face detection. So to simply blame the poor results on the photographer is too easy. The camera definitely takes the blame here!
I have the same camera (for over a year now), and have used it in what sounds like very similar use-cases. I have not encountered your issues. Either you have a defective copy, or there’s something that you could do to improve the situation (yes, that’s user error). If a whole bunch of people using the same gear in similar situations aren’t encountering the same problem, the only explanation other than it’s the one using the gear is if the gear itself is broken somehow. Those are quite literally the only two options. If you are really sure the problem isn’t “the twelve inches behind the camera” (as Ansel Adams famously said) then I would get your camera looked at.
I don’t know about your case specifically, but I do think a lot of the problems are indeed user-error, and it’s because Fujifilm’s implementation is a lot more customizable (which can be good), but is also a lot less intuitive (which can be bad). As with anything, there are pluses and minuses. I think people want to power on the camera and have the AF work great out-of-the-box, which is pretty much how the Canikony brands have implemented it, so it’s understandable. But that’s not how Fujifilm’s works, so you likely have to try a little harder to get it set up in a way that does work for you. Once you figure that out, it’s still not quite going to be at Sony or Canon level, but pretty close nonetheless. Interestingly enough, Camera Conspiracies is a great example of this, where he had to admit a lot of the problems he was frustrated about (in many videos) was ultimately because he didn’t set it up “right” for his use cases (was his own user-error). Once he figured it out, he was a lot less frustrated by the cameras.
https://youtu.be/2BqfaGXGzWY?si=x2-Pok3XIGzdiLgD