Fujifilm recently introduced their latest 40-megapixel APS-C sensor, the X-Trans CMOS 5 HR. Along with announcing two new cameras featuring the X-Trans 5 chip, a list of lenses that are "optimized" for the new sensor was also released. But do you need the optimized lenses to get the most out of the new cameras? I address this and more in my latest video.
The list of optimized lenses has caused two main questions for users. First, how much better are the optimized lenses when paired with a Fujifilm X-T5 or X-H2? And, second, will lenses that are not on the optimized list work well, or at all, with the new cameras? I have been asked these questions by my YouTube viewers quite a few times, so I set up my own test to find out. I used two lenses, a Fujifilm 16-55mm f/2.8, which is on the optimized list, and a Fujifilm 35mm f/1.4, which is not, and took sample images using an X-T4 and an X-T5. In Capture One, I compared the images to see how each lens fared with the two cameras, and I was surprised not only by the results of each lens, but with how the older 26-megapixel X-T4 compared to the new 40-megapixel X-T5. This test was done purely as a practical comparison and was not intended to make any definitive technical assessments, so bear that in mind as you watch the video. Enjoy!
Great video as always Pete! I watch it previously on Your channel when it emerge.
Neat video! I think I'd have liked to have seen the first test done closer to f8 than f4, since some text (at least on my screen with probably slightly compressed video), looked sharper on the X-T4 image, and some sharper on the X-T5, which would typically make me think there might have been a difference in point of focus?
Anyway, great content.
Thanks, Derek! I thought of a bunch more variables like that but it would have made for a very long video lol. I focused in the same spot for the samples but the XT5 was indeed sharper, although there was no major differences that I found.
The confusion is being caused by guys like this who did not read the actual statement by Fujifilm:
"The list specifies our selection of lenses that have high resolution performance from edge to edge at maximum aperture"
That means the 35mm f1.4 lens cannot resolve 40mp at f1.4. Maybe in can at f2 or f2.8. Surely it can at f4 where he's "testing".
"guys like this" lol
The video seems to miss some of what Fujifilm and others have talked about in regards to "optimized" lenses. For example in the video he kept using 2 different cameras, thus making the test a comparison of the X-T4 and the X-T5, when really the test that should have been done, was comparing an "optimized" lens, to one that was not "optimized" using the X-T5 for both lenses. A comparison using the same body would make it easier to see how much of an improvement the "optimized" lenses offer, as well as spot contradictions, for example, will an "optimized" zoom lens at 35mm offer less fidelity than anon-optimized 35mm prime lens? Will the optimized zoom lens at any focal length out-resolve the non-optimized prime?
On many cameras, a higher res sensor will capture more detail, even on an older lens, since resolving capability of a lens varies throughout the frame (dropping as you get closer to the edges).
As an extreme example, the Nikon AF-P 70-300 f/4.5-5.6 (full frame entry level lens), looks better on the Nikon D850 than it does on the D780, though it in no way means that the AF-P 70-300 is a good lens for the D850, as even though it looks better, it is acting as a massive OLPF for the D850 sensor.
How do you create a fair comparison using two different lenses, which have completely different optical properties? If I used an older 35 vs a newer 35, and conclude that yes, the newer lens is better on the newer camera, the obvious problem is that people would say "of course it is." I thought about this, and to me there's no way to prove that the "optimized" lens is better in this situation, since the newer 35 would perform better even on the older camera.
This is why I figured it would be interesting to see how each lens performed on the older vs newer sensor. I also wanted to see how much of an improvement the 40 megapixel sensor is practically speaking. To do what you are suggesting I guess the best way would be to take 2 35mm's for example, and then an XT4/XT5, and compare both lenses on the older and then newer sensor side by side. But this also would have its own challenges.
No matter what, any test like this creates more questions and a lot of arm chair quarterbacking, which is fine, but everyone who says "you should have done this" is welcome to do their own tests too. I would enjoy seeing more of these kinds of comparisons.
For 2 different lenses, it will certainly not be a fair comparison of the lens since newer lens tech will have improvements, instead it would be a test of the sensor, as well as a test to see if the benefit of an optimized lens is worth while. For example if the benefit is marginal compared to a non-optimized lens, then it changes the way someone looks at lens selection if they are switching to Fujifilm.
Beyond that, the video showed that the lenses used that were not optimized, showed good improvements in detail, thus showing that those lenses were made very well and were unlikely to hold back the 26 megapixel sensor.
Only thing that is more of a curiosity is how much will they hold back 40.2 megapixel APS-C sensor.
Beyond that, since their optimized terminology is more of marketing, there are no objective standards for it, such as a minimum LPM or other measure of resolving capability. That is the main reason why I was interested in seeing comparisons such as a non-optimized prime vs a optimized zoom set to the same focal length as well as examinations at different focal lengths, though that likely goes beyond the point of the video.
It makes absolutely no sense to compare a lower mp to a higher mp and magnify the lower mp to 200%, 300%, 400%; or 1.5x of the higher mp's magnification and then say, yep, the higher mp is sharper and have more detail. Well, of course it's going to. With this, it's not even about the lens anymore.
It would have made more sense to use the same body. This way, even if you zoom a ludicrous 400%, it's still a fair and common-sense LENS comparison.
Of course I thought about all of this, and here's the problem. How do you create a fair comparison with two DIFFERENT lenses? If I took the XT5 and 2 Fuji lenses, which both have completely different properties, you'd be complaining that it's not fair based on the lenses I chose, and that I should have instead picked two other lenses, etc.
Choose two different lenses (same focal length) and test them on the two different bodies to eliminate both variables. Does the 35mm f2 and 35mm f1.4 both set to f2.8 resolve differently on the X-T4 vs. the X-T5? How about f4?
Dave, I actually just mentioned this in a comment above haha. Yes, this makes sense but it has its own problems and would create other questions. For instance, you're still dealing with 2 sensors, in which case the newer sensor would most likely perform better, so how do you prove that the new lens/camera combo is better because it's "optimized" vs just being on the new sensor? But I don't think this is any better than what I did in my test. It's just 6 of one, half a dozen of the other. Again, even if I did this, people would complain that I should have used the same lens with 2 cameras LOL. there's no winning.
--- "How do you create a fair comparison with two DIFFERENT lenses?"
Like I mentioned, use the same body. This way the only variance are the lenses...which was the point of the comparison. Yes?
--- "If I took the XT5 and 2 Fuji lenses, which both have completely different properties, you'd be complaining that it's not fair based on the lenses I chose, and that I should have instead picked two other lenses, etc."
Wrong. Why would I be complaining about it? That's what I'm suggesting...exactly. A lens comparison should be about lens comparison, especially since they are the same mount.
Rule of thumb:
a) When comparing Lenses, use the same body.
b) When comparing Bodies, use the same lens
Great video! I think this confirms what most of us would expect - higher MP = more available detail, regardless of the lens being used. Seems like shooters will benefit primarily from optimized lenses when doing deep crops - again, this is what we should expect, but nice to see an example of the degree to which good glass resolves more of that 40mp potential.
Seeing some differences in where the focus landed on some of the pics. Particularly the books and the one with the Ewok figure. The X-T4 was backfocused compared to the X-T5, leading the perceived in-focus areas to look soft.