My guess is it will be about another year for a III, maybe longer. I'd only expect modest upgrades. Maybe some AF improvements, and faster processing but not much else.
Not until after the XH3 set of cameras with they're newest processors. And considering how I'm still shooting the GFX100II in AF-S mostly with the small AF point like it's a DSLR like I was with the 100S I don't think the new AF tech is going to matter much.
I am more convinced each passing day that Fuji doesn't care about their AF catching up to CaNiSo... They seem more interested in making simpler profits off of their retro-look fans. We will see when (if?) the XH3 is actually released if I'm (hopefully) wrong. As of now given Fuji's relative lack of response to the still market, I have my doubts an XH3 will offer any truly significant improvements over the XH2. As such, I suspect it will be at least 2 more years before a GFX 100iii; and then it's probably going to be a higher resolution sensor as the main feature upgrade...
Oct 16, 2025 at 11:42 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
Jack Flesher wrote:
I am more convinced each passing day that Fuji doesn't care about their AF catching up to CaNiSo... They seem more interested in making simpler profits off of their retro-look fans. We will see when (if?) the XH3 is actually released if I'm (hopefully) wrong. As of now given Fuji's relative lack of response to the still market, I have my doubts an XH3 will offer any truly significant improvements over the XH2. As such, I suspect it will be at least 2 more years before a GFX 100iii; and then it's probably going to be a higher resolution sensor as the main feature upgrade... ...Show more →
I actually think about it kind of the opposite way. I think the CaNiSo AF capabilities are so good, there really isn't much room for improvement. They might improve things a little here or there, but I think for many types of shooting AF focus for these systems has been so close to bullet proof that they haven't advanced much. In contrast, Fuji has had some work to do and they have made their own slow progress, but because the other systems can't improve much Fuji has caught up substantially. I expect that to continue over the next 5 to 10 years. Sure there will be some types of action photography where the big guys will still be noticeably better, but Fuji will be darn close even there. I actually don't think the current gap is all that big to be honest.
I also think this is the most natural improvement for the GF system. The AF is behind, but I expect they can begin to close the gap as that system just has so much more room for improvement. My best guess is the 100 III has an AI chip and a lot more computing horsepower and that will be the major upgrade. The 100 made the jump to a higher resolution sensor and that sensor still has as much pixel density as an FF sensor. I don't see them going with higher resolution just yet. The 100 II added a better base ISO, better IBIS, and a better EVF, so I don't expect and can't really see how they would improve any of those things much more. That leaves AF as one thing that needs to be improved and probably can be improved and would justify for at least some getting a new model. They might be able to improve a few other things like frame rate, flash sync speed, and I suppose video as well.
It strikes me as kinda funny going through some of the threads and YouTube's recently. Tony Northrup skewering Canon for saying they want to build iconic cameras, the "Porsche 911 of cameras", noting that Canon has focused on frame rates and improving performance in areas they are already good at while building blob shaped cameras that don't inspire. Meanwhile Sony appears to be launching an A7V with improved AF on an old sensor, and people are already disappointed and its not even released yet. Fuji releases gorgeous camera like the XE-5 that has the same performance as other models like they did with 100VI, and X-T5. My point here is that companies tend to focus on what they are good at instead of what they aren't. Kind of a corporate comfort zone.
Somewhat opposed to this thinking, I went out for a golden hour walk yesterday eve and thoroughly enjoyed my Contax 35/3.5 on Gfx100S. Came away with some great shots and feeling great. It crossed my mind that Fuji might, someday, release an AF system in new models that wouldn't allow me to strap old glass on my new AI driven, laser focused, latest, greatest Gf super X with 150mp. That would be disappointing. I guess I am happily living in my old tech.
Steve Spencer wrote:
That leaves AF as one thing that needs to be improved and probably can be improved and would justify for at least some getting a new model.
Yeah but... They've had at least 3 years to do that and haven't. Instead they've released several retro look, fixed lens and small form factor options, while their main workhorse GFX and XH lines get limited revival...
But credit where it is due -- we've gotten a raft of good performing albeit optically slow zooms, along with a handful of faster, really good primes for both platforms
JadedWriter wrote:
Not until after the XH3 set of cameras with they're newest processors. And considering how I'm still shooting the GFX100II in AF-S mostly with the small AF point like it's a DSLR like I was with the 100S I don't think the new AF tech is going to matter much.
That's remarkable and understandable at the same time (using it like you're used to). I shoot my GFX100II in AF-C mostly. Like I did with the A7RIV before. The fact that AF-C on the 100II is so good, was even one of the reasons why I didn't hesitate to switch from Sony to Fuji. It didn't keep me from changing systems. It's not better, but very good and certainly good enough for my usage.
If only I could use that (very) small AF point (like in AF-S).
I honestly do not know how people have their 100II's setup. Whenever I'm in AF-C there's a very good chance that the AF is just going to detach from the subject, lose focus and then basically hunt till there's nothing in focus. So I basically just turn it off. I don't know if it's a lens thing or what, it happens a lot on my LM lenses like the 110 and 250. Now I basically just use AF-S to reduce that. Mujabad123 wrote:
That's remarkable and understandable at the same time (using it like you're used to). I shoot my GFX100II in AF-C mostly. Like I did with the A7RIV before. The fact that AF-C on the 100II is so good, was even one of the reasons why I didn't hesitate to switch from Sony to Fuji. It didn't keep me from changing systems. It's not better, but very good and certainly good enough for my usage.
If only I could use that (very) small AF point (like in AF-S).
I use the 20-35mm and the 110mm. In AF-C both lenses stick to their prefocussed point perfectly (with portraits I prefer to use eye tracking in AF-C). Even in less than ideal lighting. Only this morning I had a portraitshoot in a dimly lit boxing school, with no additional studio lights. No problem with AF on moving subjects, although some images were shot at ISO6400 and one even at ISO12800 (DXO PL knows how to work with those files).
Don't know about your settings ofcourse, but for me there are no problems with AF-C. Did you try AF-C tracking?
My 2 Cent: Legacy lenses (= pretty much all of the current GFX catalog) are the road block.
The catch (experienced by Hasselblad users as well) is that repurchasing many lenses is a big financial obstacle that would delay/prevent upgrades to some extent.
On the technical side, 1. The focusing noise (and speed) of many current GFX lenses feels like there is a small animal trapped in there and actively resisting the attempts of the camera to acquire a subject. On FF, recent releases (e.g. by Sigma) have much faster autofocus motors.
2. 'Good, not great' designs like the 55 f/1.7 would not look pretty on higher resolution sensors?
Steve Spencer wrote:
I actually think about it kind of the opposite way. I think the CaNiSo AF capabilities are so good, there really isn't much room for improvement. They might improve things a little here or there, but I think for many types of shooting AF focus for these systems has been so close to bullet proof that they haven't advanced much. In contrast, Fuji has had some work to do and they have made their own slow progress, but because the other systems can't improve much Fuji has caught up substantially. I expect that to continue over the next 5 to 10 years. Sure there will be some types of action photography where the big guys will still be noticeably better, but Fuji will be darn close even there. I actually don't think the current gap is all that big to be honest.
I also think this is the most natural improvement for the GF system. The AF is behind, but I expect they can begin to close the gap as that system just has so much more room for improvement. My best guess is the 100 III has an AI chip and a lot more computing horsepower and that will be the major upgrade. The 100 made the jump to a higher resolution sensor and that sensor still has as much pixel density as an FF sensor. I don't see them going with higher resolution just yet. The 100 II added a better base ISO, better IBIS, and a better EVF, so I don't expect and can't really see how they would improve any of those things much more. That leaves AF as one thing that needs to be improved and probably can be improved and would justify for at least some getting a new model. They might be able to improve a few other things like frame rate, flash sync speed, and I suppose video as well. ...Show more →
Focus speed/accuracy varies so much between my GF lenses on my GFX100S II, it leads me to wonder how much their lag behind CaNiSo is related to the camera and how much is related to the lenses. Of course, I expected this shortfall when I bought into the GFX system, so I still shoot a little over half my shots on Sony FF, with little expectations of significant change on Fuji. I’m sure I will more than likely buy the GFX100S III when released, but my expectations are certainly tempered.
Since Fuji can’t catch a break with autofocus speed on GFX cameras, and the common complaint about G lenses is their size, it just makes sense for them to make nice manual focus G lenses next - compact and without autofocus issues! I don’t think they’ll improve autofocus much without redesigning the lenses.
The 100 II still has very slow sensor readout speed compared to FF35, especially the stacked and partially stacked sensors, which presumably an upper bounds on how close Fuji is going to be able to get to any of those cameras until Sony improves the miniMF sensor offerings. I think it's telling that supposedly (I haven't used any of them) Fuji's own smaller-sensor cameras are better AF performers--that tells you there's more going on than, like, programming know-how.
1. Processing speed: same processing unit as in the APS-C cameras; more data (100 MP) to process
2. Lens size (distance of lens elements to travel)
3. Lens motors (speed, power)
4. Lens aperture when dark (not as many fast lenses as on FF)
5. Software (not Sony-level)
6. Fuji's priorities/attitude towards fixing (1.-5.)
and of course -
7. User skill
Oct 16, 2025 at 06:12 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
mzbe wrote:
GFX auto focus is inherently limited by
1. Processing speed: same processing unit as in the APS-C cameras; more data (100 MP) to process
2. Lens size (distance of lens elements to travel)
3. Lens motors (speed, power)
4. Lens aperture when dark (not as many fast lenses as on FF)
5. Software (not Sony-level)
6. Fuji's priorities/attitude towards fixing (1.-5.)
and of course -
7. User skill
These are mostly things they could address in a 100 III.
#1 - there is no reason they have to use the same processing unit as their APS-C cameras. They could bring more processing power to the 100 III.
#2 - they can't change the lenses that exist (at least not easily without a recall or something), but some have really nice linear motors and others only move smaller elements. Improving speed for even some of the lenses would very much be welcome and of course future lenses could address these issues.
#3 - see above some have quite decent motors and some only move smaller elements and they can certainly make use of some of the lenses now and eventually address this more with future lenses.
#4 - when considering lens aperture the important bit is the entrance pupil of the lens and not the f/number, which is a function of focal length. For example, the 110 f/2 has an entrance pupil of 55mm, which is pretty close to the entrance pupil of an 85 f/1.4 which is about 60mm. There isn't really much to that difference in providing light to work with. Keep in mind with a larger sensor the light is spread over a larger area, so a lens with a bigger f/number still has as many photons hitting the sensor as a lens with a lower f/number focussing that light on a smaller area. The way to compare is by looking at the entrance pupil which is the hole that actually lets the light in and not the f/number. If two lenses have the same field of view and the same size entrance pupil, then they will provide the same amount of photons to turn into signal even if they have different f/numbers. If you compare this way, then you will see FF has at most minimal advantage and the advantage for sensor size depends on the field of view with sometimes FF having a slight advantage and sometimes 44 X 33 having an slight advantage.
#5 Software -- Fuji can obviously do better and I believe they have as we have seen slow but clear progress in the development of their AF. I hope they put in a big effort here with the 100 III.
#6 Well we will see about whether they prioritize fixing the above. I tend to be optimistic, so I think they will actually really work on these issues for the next version of the camera.
#7 User skill- hopefully we can all do our part as well.
Steve Spencer wrote:
#1 - there is no reason they have to use the same processing unit as their APS-C cameras. They could bring more processing power to the 100 III.
There are costs for using a different and more powerful processor in the lower-sales GFX series.
Oct 16, 2025 at 09:47 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
Steve Spencer wrote:
Of course there are. If they want to improve AF, however, increasing the processing power is one of the things that they could do.
It will be interesting to see if Fuji wants to keep improving the video capabilities of the Gfx. Now with the Eterna, they could focus video on that camera and leave the Gfx to stills. If they want to continue to improve video that will mean more data processing needed (assuming they would be increasing sensor read-out speeds).
