Sony A900 vs. Fuji X-T1, a Surprising Conclusion

Apropos Ctein's post above, I want to tell you my conclusions of trying the 8-year-old (Sept. 2008) Sony A900 (24 MP, full frame) and Sigma 35mm ƒ/1.4 Art against the current (Jan. 2014) Fuji X-T1 (16 MP, APS-C reduced frame) and Fuji XF 23mm (35mm angle of view equivalent) ƒ/1.4.

Why just the conclusions? Because the proof is tedious, and...well, isn't proof. I'll give examples of what I mean by that below.

What I did was to wander around inside and out and make about 40 matching shots with both cameras. I matched ISOs, apertures, and focus points, and let the cameras choose their shutter speeds. I then opened a number of the matching shots in Photoshop so that I could toggle back and forth between the two images and examine them at either 100% or size-matched. (The latter put the Sony images at about 85%.)

There are some distinctions. At ISO 400 outdoors, the Sony is just a little more crisp—you see it especially with complex subjects like bramble at a distance or evergreen trees against the sky. Both camera/lens combinations are very sharp and clear and both hold highlights and shadows equally well. Here's one comparison at ƒ/8:

(Sorry about the guidelines in two of those.) Those are both from the same frames, which look like this:

Is the Sony a little better? Is the Fuji a little better? Overall, and after looking at twenty such comparisons, I'm not sure. I could give the edge to one or the other here or there. But in all cases, the distinctions are minor—much more so than I expected.

The reason these don't prove anything is that it's just with one raw developer and with neither output optimized. For instance, I bet where the Sony seems to have a little better detail, I could get the Fuji's detail closer with Iridient Developer rather than ACR. I'm just testing with my own stuff and doing things my own way, to see for myself.

Here are a few more comparisons, from indoors at ISO 1600 and ƒ/5.6. Tell me which looks better to you, Camera 1 (the labels are consistent) or Camera 2.

Some ninjas can get exifs from screen grabs, but c'mon, play the game, just eyeball these. Just feeling the gestalt, does one camera have an edge over the other, you think? Which one do you prefer? Which one do you think is the Sony and which one the Fuji? (I'll add the reveal tomorrow.)

[UPDATE Sat. Dec. 26: 1 is the Sony A900/Sigma 35mm ƒ/1.4 Art, 2 is the Fuji X-T1/XF 35mm ƒ/1.4.]

Verdict
My overall conclusion: there's truly not a lot of difference between them. Not a lot of reason to choose one over the other.

So what I think this shows is the pace of sensor development over the past three-fourths of a decade. It took seven or eight years, but right now, I think we're getting more or less the same image quality from a half-frame 16-MP Fuji as we were getting from a class-leading full-frame 24-MP resolution-champ Sony back then. Yeah, I can see distinctions, but, mostly, they're not enough to matter to me. I could live with one or the other, forget IQ, and just shoot.

More importantly, I'll bet I could show different pictures made from one or the other of these cameras and most people wouldn't be able to tell which ones were shot with the bigger camera and which ones with the newer one. You'd need to be told.

Of course, Sony has much better FF sensors now. I'm not looking at that comparison. But what I am looking at really underscores what Ctein is talking about in his X-mas post above: sensor development is proceeding, and we're not at the end point of technical development yet. What it says is that "the best" is only the best for now. If the market continues to support technological improvement, today's best will mostly likely be overtaken by lesser devices in a handful of years. That's long been known, of course, but here I found myself looking at an actual example that handily demonstrates it.

Progress...quite a gift!

Mike
(Thanks to LensRentals for the loan of the Sigma 35mm)

Original contents copyright 2015 by Michael C. Johnston and/or the bylined author. All Rights Reserved. Links in this post may be to our affiliates; sales through affiliate links may benefit this site.

TOP's links!

(To see all the comments, click on the "Comments" link below.)
Featured Comments from:

No featured comments yet—please check back soon!

Comments

[Ken Bennett]

I'd be pretty happy with the output from either camera. On the set of indoor photos, I would guess that #1 is the Sony simply for the cooler color temperature and the harsher shadow noise, but they are very close.

Happy Christmas and other assorted winter holidays to you, Mike, and all your readers.

[LeftCoastKenny]

In my early morning pre-caffeinated state I don't see much difference between the Sony & Fuji beyond a bit more red in the Sony, or cyan in the Fuji. With just a smidge more adjustment to one or the other they would likely be indistinguishable. I'd be happy wandering around with either but would prefer the lighter, smaller Fuji to carry on extended outings. Thanks for the comparison! Now, back to my morning cup...

[Stephen Scharf]

Progress is indeed, quite a gift. We are suffering from an embarrassment of riches these day with the cameras and lenses available to us.

Just a couple of points for clarification: the APS-C is 2/3 the size of a full frame sensor, not half the size (that is the size of the Olympus and Panny 4/3 sensors).

And yes, you could pull quite a bit more sharpness out of the X-T1 with Iridient Developer.

Merry Christmas!

[John Sparks]

I recently bought a Sony A6000 (it arrived the day before you recommended it). I did some informal comparisons with my Sony A99 (which really only noticeably improves on the Sony A900 at high ISO). I am amazed at how similar the images look. The A99 has a slight edge at 100% in bright light, in some cases, but the difference is extremely small. In the largest prints I can make (17"x25"), I don't think anyone could reliably tell them apart.

[Patrick Cooney]

For my taste in IQ, camera 2 has a very slight edge over camera 1.

My best guess is that camera 2 is the Fuji. Why? I thought I saw more chromatic aberration, color moire, and more shadow noise in camera 1's images, more of each than I expect from the Fuji.

[adamct]

I consistently liked the images from camera #2 better, but I wouldn't have the foggiest guess which camera is which.

Merry Christmas!
Adam

[tex andrews]

Hi, Mike, and happy Yule to you. This post actually makes me feel better---having just bought 2 new cameras in 2014, if it takes a few years for them to be matched, not to say exceeded, then I feel better about my purchases now than I did a decade ago, when you had the strong nagging sense that the camera you were purchasing right then, as you paid your money, would be quickly obsolete. Now I feel like we can buy with greater confidence that we can stand pat with our purchases until they break. Or at least until we know them inside and out. That's a good thing, I'd say.

[s.wolters]

The colors that these cameras produce are so different. Maybe they are both wrong, but they can’t be correct both of them. (I suspect it is the old Sony that is the one that is far away from reality).
Better a 6 megapixel camera that produces great tones and colors than a 60 megapixel one that turns the world into a strange planet.

[David Anderson]

Having had a Sony A900 in the past and now having a Fujifilm X-T1 and X-Pro 1, I came to my own conclusion that the presence of the AA filter in the A900 and the inherent difficulty of getting autofocus to work consistently across a range of lenses even with micro focus adjustments, negated any benefit of the larger sensor of the A900. I found the consistent quality of the X series lenses the clincher. To start with I missed the IBIS, but the X zooms have excellent OIS and the primes are fast and don't need it. I don't miss the A900 and the extra weight and size at all now. The only Sony lens I really missed was the 135/1.8.

Having said all that I will probably make a fool of myself and say that I feel that camera 2 is the Fujifilm. It seems the shadow noise and CA is less. In my defence I don't use PS for processing the Fujifilm files so the results may look different to what I normally see, instead I am using either Aperture for raw or accept the excellent OOC JPGs in some circumstances.

Merry Christmas.

[MarkB]

While I do have a preference for the #2 camera in the unlabeled images, I've also gone through this exercise with a Sigma SD1 Merrill (14MP Foveon) and my then brand new Pentax K-3 (24MP Bayer), both with Sigma 70mm Macro lenses. Pretty much indistinguishable at the 14MP size, with fewer artifacts in the K-3 files, and much easier processing. And yet, I still kept the SD1M, mainly for esoteric reasons I can't quite explain... almost as if I *like* technical constraints!

[Richard]

Camera 1 is the Sony, and 2 is the Fuji. I can only tell, because I can see the CA. I have become very sensitive to it since I got an EF 85 f1.8, which I affectionately refer to as my purple lens. As for the quality of the files, both are pretty close, once the CA has been addressed. I like the 2 files better.

[ctein]

Dear Mike,

I would like to add that it's not just improvements in sensors. Improvements in the readout electronics have a substantial impact on image quality, for a variety of reasons. It's part of why two different contemporaneous cameras using the same sensor may very well yield different measurements on the test bench.

~~~~

Dear Stephen,

I think there's been a vocabulary confusion. You are indeed correct that measures of image quality scale with the linear dimension of the sensor, not the area. Which is why it's appropriate to call micro 4/3 half-size, relative to full frame, whereas people who talk about it having one quarter of the area are making a meaningless comparison.

But, that's not what “half frame” means. It's a term from film use for an uncommon still-photo format that split the 35mm frame in two, yielding to 18 mm x 24 mm frames (it's not too much different from 35mm motion picture format). That's what Mike was referring to when he talked about it being a half-frame camera. I can see how that would be confusing if you don't know the old terminologies.

~~~~

Dear s.wolters,

It is a fool's errand to attempt to judge the accuracy of tone and color in web images. You have no idea of what is or isn't accurate nor how faithful they are to the original. Many things can confound the rendition.

For example, on my calibrated sRGB monitor, viewing them in Firefox, under Mac OS X, there's very little difference in color rendition between the two cameras and what there is doesn't clearly suggest one is better than the other. If anything, my display favors the Sony. Further, this is entirely dependent on the exact RAW development settings used on the camera files; it would be not at all difficult to make them virtually identical.

The truth is that neither of them is “correct” in the sense that you mean it, and we are not close to talking about great tones and color vs. surreal ones.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com
-- Digital Restorations http://photo-repair.com
======================================

[Frank]

First,camera 1 and 2 are similar, and images from them are almost the same good.

Then, to my eyes, images from camera 1 look a very little better. (Again, they are actually the same.)

[Ben Rosengart]

I prefer camera #2 for its gentler contrast.

[Rod Thompson]

As an extension of your comparison, the Fuji stands up well against the D750?
I am guessing the Fuji is bottom from color warmth (I have noticed the Sony sensors tend towards cooler tones?)

[Robert Roaldi]

Comparisons such as these should have the effect of increasing the price of 2nd-hand cameras and maybe depressing the price of new models. Luckily, by and large, that's not what happens, and so guys like me enjoy life much more frugally by always staying a generation (even 2) behind.

[Steve Jacob]

To be fair, Sony never quite achieved the same noise performance from their sensors that Nikon did. I think the improvements are largely down to improved processing rather than physics.

Even so, comparing Fuji XE2 and Nikon D800 shots in a 24X16" print it is very hard to tell the difference. If you don't make prints bigger than that, much of the sensor performance data becomes somewhat theoretical.

And the latest version of LR CC seems to do a better job too, with a slight bump to the detail slider.

[s.wolters]

Ctein, thanks for your comment on my comment. Of course there are a lot of factors that influence the color reproduction in web images and probably we will never find one that is completely accurate. One could say that all reproductions are wrong, but surely some of them are less wrong than others. Men from Mars are as we all know green, not greenish blue. That’s not a matter of opinion or personal taste, but a fact.

These huge color differences worry me. Fortunately the newer Fujifilm looks much much much better (and the more recent Sony R-range is great as well). Hopefully that’s progression.

Most images we see daily are web images and the majority of them certainly comes straight out the camera. Therefore we should pay more attention to jpegs than to gallery prints. In other words, we are not dealing with the quality of the champaign here, but with the quality of the drinking water.

As an extremely lazy shooter I just want to take fuzz-free pictures and do not want to re-invent digital photography. Hardly ever go beyond jpeg-fine. In my life as art-director however I am very critical. Over 35 years I worked with and commissioned many kinds of photographers. Sometimes we had issues about the content, but hardly ever about the quality of reproduction. Not that it’s unimportant, it's just common ground. But if an expert on reproduction tells me that blue grass could be just as plausible as green grass, I become very uncertain. Surely there must be a bandwidth within things are acceptable.

Anyway, I would have rejected the examples with the number 1 on it (as they appear om my screen) for the use in serious visual communication.

[David Bennett]

All the images look good, certainly good enough. I like Fuji colours and Olympus colours. whereas I am not familiar with Sony colours. Therefore, other things being equal, I would go for the Fuji.

But things are not equal. The Sony and Sigma combination is, I think more than twice the weight of the Fuji.

That tips the balance for me, if it needed tipping, because the main drivers for me nowadays are weight and size.

[Steve Deutsch]

The apparent depth of field of #1 seems wider than #2 (look at plant stalks outside of windows). I was thinking this meant #1 was the smaller sensor (the Fuji), so I'm bit surprised it was the Sony.

[I've puzzled about that myself. I went back to look at the original files at up to 600%. I suspect it has something to do with either a) a slightly different focus point (probably) or b) the fact that the grass fronds are being imaged through window-screen--something about the interference could make them seem more in-focus in the Sony shot. That's possible, but I suspect it's just the former. I "asked" each camera to focus on the chair, and most of the objects in the scene were more or less the same distance from the camera as the chair, but it's possible there's just enough variance in how the AF worked that a slight disparity in focus is apparent.

Anyway my written conclusions were based on comparing many images. The grabs from these two are really just for illustration. --Mike]

[Tom Legrady]

In image 1, I like the warmth and richer shadows of #1.

Image 2, #1 is too blue, #2 has better whites.

Image 3, there's something about the noise in #2 which is more acceptable.

Image 4, #1 has a quality I like.

But it's like comparing first and second place finishers at an Olympic event. Husain Bolt may have a fingertip lead over his competitor, but from my viewpoint, they both win, compared to 99.99 % of humanity.

Both cameras are good, and if you aren't taking advantage of the infintisimal advantages of one over the other, who it doesn't much matter which is on your should.

I like the price tag and weight of my Fujis.

[Tim]

Examples as depicted, I'm a pretty consistent Fuji or #2 across the board. Only exception is the red/cyan in the closeup of the tree-trunk. Otherwise, both the roof-tiles are sharper and the green of the distant (cypress?) trees is more believable.

That opens up the whole realm of areas whereby such differences can occur: variable white-balance at source is an obvious contender; deviations in RAW-converter's mapping to target (ie Adobe/DxO/other's profiling of the respective sensors); your manual intervention.

So, here's a question born from both the over/mis-processed Butters in an earlier post and the chair and microwave above. Have we (collectively) abandoned knowledge of tone-placement (cf Zone-system)? I mean, if it were me posting that photo of the chair, I would process it every which way until I was happy with the brightnesses of both the light top and the front face in relative shadow as well.

[Marco Ristuccia]

I like prefer by far the resuts from the Fuji camera. Sony images clearly shows a blue/red chromatic aberration on high contrast borders. And because of this the camera 1 is clearly recognizable as the Sony one.

Of course this is more a problem of the lens rather than the camera, and of course can be easily corrected with the modern digital darkroom softwares. But if we have to produce a verdict based on the images you posted here, then in my opinion Fuji clearly wins.

[Corky Miller]

One design consideration is price. Sony: $398. Fuji: $810.

[Tom Bell]

Love this whole thing. It is extremely encouraging for me for as I grow older and my left knee suffers from walking I stay one generation behind.

I have a OMD EM5 and love the colour rendition and the size with small primes or the quality of the 12-40 zoom

I love my X100 for people and streets ....

I bought the XPRO1 with 18/27 lens ... so lovely but I just couldnt cope with the desperate focusing .... So I await the drop in price of the XT1 or even XT10 ....

Coming from Pentax I realise that three things are special to me

1/ compactness
2/ image stabilisation
3/ fast focusing

.
They transform the pleasure of photography. More images caught, less need for tripods .... And a lightness which enhances my pleasure of a day in the hills !

[Gerry]

This article neatly illustrates why I am still happy with my 5D Mark 2 (released in January, 2009) and have not bothered to replace it with anything newer.