Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Some Teasing: Fujinon XF 16-80mm f/4
#1
Well, I am terribly sorry, but I'm afraid I need a few more days to publish the review. Because:

  • I'm really out of practice Wink
  • for several days in a row now I haven't seen much else other than the test chart, Capture One, Imatest and Excel and I really need a break
  • Since my last review, basically everything is new: new cameras, new charts, new software. All or most of the work to set up everything is done, but one thing I completely underestimated is: none of the templates I used in the past can be reused. So, for the first few reviews, I'm basically starting from scratch for most charts and images.

However, a few facts in advance at least:

I guess no one really expects a consumer zoom (and that's what the XF 16-80 in my opinion is, even with its high price point) to be stellar. Still, the lens has some surprising negative sides.

Vignetting, for example. At 16mm wide open, it exceeds the scale of our chart by a full stop, reaching 2.5 EV.

And then there is distortion. You could see it in some sample images already: pronounced pincushion distortion for most of the upper zoom range and hefty barrel distortion at 16mm.

'Hefty' in numbers means: 5.36 %

Of course there is software correction, and if you shoot JPG or use a 'Fuji friendly' raw converter, the actual distortion will be hidden (as the camera hides it by default in the viewfinder and on the rear display).

But software correction takes its toll, in this case it's a whopping 20% resolution in the image corners.

Just the corners, though, the effect of the auto correction is a lot less pronounced at the image borders and of course basically zero at the image center.

As mentioned already in a different thread, I will provide MTF charts for both native lens performance and autocorrected (with correction in Capture One).
Editor
opticallimits.com

#2
Come on just get it out of the gate already. Smile I'm not a Fuji user, but even I am jumpy with anticipation. Big Grin
#3
Unfortunately this is the new trend: rely heavily on software correction for vignetting and distortions, almost all new mirrorless lenses follow this trend and I don't expect it to be any worse than competition, check otther lenses 2% distortions is being rather acceptable since most zooms are within this league, however some lenses like Sony 16-50 show rather extreme values
I think you need to take this into consideration for the end verdict and rating, just my 2 cents
#4
I definitely will. And it's also the reason why I'm currently rechecking the results. With C1, there occasionally seem to be noticeable differences between the manufacturer profiles for and the C1 profiles (one can choose in C1). I went for the C1 profile in the first analysis of the 16-80, because it's the default and there was no visible difference to the Fuji profile.

However, while doing analysis of some other XF and XC zooms (note to self: checking more than 30 lenses in parallel is not a healthy approach Wink ) I noticed very visible differences between the two profiles with the XC 16-50. I am currently rechecking results for both zooms with both profiles, because I assume that either the C1 profile for the XC 16-50 is wrong or buggy or there is a bug in C1 when applying the C1 profile to images of that particular lens.

I know I'm really stretching your patience now, but I would prefer to make sure the results are not flawed before I publish the numbers. Publishing wrong numbers and then later having to admit the results would be a nightmare... lesson I learned from the first Nikon AF-D 18-35 review Wink
Editor
opticallimits.com

#5
Even 5% is not the limit now if the figures from - say - the new Tamron 20/2.8 are anything to go by (8,1% barrel as measured by Ephotozine for their review); the 24/2.8 isn't far behind. Smile

(EDIT: I was replying to toni-a but I was so slow that Markus got his post up in the interim). Smile
#6
Fuji has released lenses with even more distortion before, the XF 16-55 for example.

On the other hand, it's not that they don't know how to correct distortion optically. Some of the primes show almost none at all. But even within one product line, there is no consistency. Take the "Fujicrons" for example: the XF 23/2 and 50/2 show a very low amount of distortion, while the XF 35/2 distorts heavily (and so does the XF 16/2.8, btw.)
Editor
opticallimits.com

#7
Something i am afraid of is that people like to compare and need numbers or stars rating, while a lens star rating might be severely sanctioned for vignetting or distortions.... As for resolution I suggest it to be measured on distortions corrected pictures
#8
(11-13-2020, 07:52 PM)toni-a Wrote: Something i am afraid of is that people like to compare and need numbers or stars rating, while  a lens star rating might be severely sanctioned for vignetting or distortions.... As for resolution I suggest it to be measured on distortions corrected pictures

This is not so easy as it sounds - the various interpolation engines use different algorithms.
Chief Editor - opticallimits.com

Doing all things Canon, MFT, Sony and Fuji
  


Forum Jump:


Users browsing this thread:
1 Guest(s)