Sony FE 70-200mm F2.8 GM sample gallery updated

My mistake. Yes MP resolution 50.6 MP (5DSr) vs 42.4 MP (A7Ri)

At test chard distance and just for resolution this might be correct, but what about other distances, and real world subjects. And what about handlig, focus speed, focus accuracy, build quality and other aspects of the lenses? What about using an adapter that limits the camera features you can use? Doesn't this matter? And dose such minor differences really matters more than those other aspects?

Actually, I find such test chard photographs quite boring.

IMHO, the digital picture website is no so good. Between Canon 100-400 IS and Canon 100-400 IS MKII it shows very little difference, in all focal lengths, with no numerical or graphical presentation. I don't think it is trustable resource.

I think the Sony is No less sharp and it is the 20% difference pixels that makes the difference.

@jaykumarr I'm not sure what you mean, there's a pretty large difference in corner sharpness at 400mm and a significant difference across the frame at 100mm. The Mk1 version was always a pretty decent lens in the center, it's just that it wasn't as good as the 400 f/5.6 prime across the frame and was prone to decentering. Also, it's only a 9% difference in linear resolution between the A7R2 and 5DsR although that doesn't account for the 5DsR's not there but still there AA filter.

@ whumber: Tests have shown noticeable differences in performance for lenses from different brands, like test shots taken at infinity, 10 meters and 3 meters. Also, there are a lot more to image quality and perceved "sharp and clear" than just resolution. Pictures with real world subjects can often tell a lot more than just 1 bit/black & white test chards.

Sony is a good lens, but like many Sony lenses is way overpriced. I'm looking at adding A7RIII and if replaced my triumvirate, 24-70 f/2.8 II, 70-200 f/2.8 II and 100-400II, it would cost be an additional $1800 for the glass. And when does Sony run good lens rebates like Canon and Nikon.

Luckily I can use a metabones and keep my Canon glass, but Nikon users would be SOL.

Used the F4 for our a9 review, was pleasantly surprised by the quality.

Prime good wide open @f4. excellent

If you use the F4 on a 24mp FF then it is perfectly fine. On the 42mp FF it is solidly ok. On 24mp APS-C I find it a bit soft at F4.
However I shot it side by side against the F2.8 and was really challenged to see the difference at least in the sharpness.

Thats a good joke moimoi. Take a look at DXOMARK and you fond that the Canon is not as sharpness the Sony.

I can confirm as a user of both lenses that Sony does have better central sharpness so that I'm not afraid to use it wide open. I've used Canon usually @3.5. (I did not specially compared corner sharpness, but looks fine on my half body shots wide open.) [Sony A9 vs Canon 1ds3]

Looks like the Canon is worse. Weird as I do not recall such a bad IQ with that lens. I will need to double check this.

Though I do not think the Sony samples shot at 200 mm are particularly sharp wide open on the dpreview gallery.

Sillen, when I look at the Digital Picture's crops, and Lens Rentals MTFs, I believe the Sony isn't as good as I first thought. They've been making some great glass lately, like Sigma, but I think this one is pretty much average. Definitely slotting behind the class leader, Nikon's, but IMO also behind Canon and Tamron. Again its not bad, just isn't great.

not wrong

just saying, thats over £8000 of camera equipment on the sample gallerys and on my 27" monitor they look no better than an iphone, unless you pixel peep ofcourse

soft, colourless and no soul

nothing magical about this combo , my opinion ofcourse

ofcourse people insist on using a specialist sports camera to do work they should be doing with an a7rii or a99ii......

I see many artisans shooting portraits with the a9... why on earth would you want to do this the A7rii and even the A7R mk1 does a better job...

@barrym1966
maybe what your seeing is what LensRentals saw when they tested this lens.
They concluded that it was not a great lens by comparison to the Nikon and Canon versions.
That conclusion came after testing 10 different samples.
https://www.lensrentals.com/blog/2017/01/an-update-and-comparison-of-the-sony-fe-70-200mm-f2-8-gm-oss/

yeh im not alone, I am actually comparing it to my A99ii and SAL70200G2 the results I am getting are head and shoulders better, and they really shouldnt be, should they??

Totally agree the photos are really desaturated. Did Nikon/Canon pay for this?

@ barrym - You really can't compare images through DPR sample gallery online to your own experience in LR etc. They will always be miles apart. Need to download RAWs and compare on the same platform / viewer.

Not that hard... just zoom into 100% and check what the resolution looks like. Do this for a number of images and number of areas covering different lighting conditions to isolate for external factors.

I compared this to some of my own shots which were made at higher ISO, slower shutter speeds, and with athletes (i.e. moving subjects, indoors, compared to these which were shot in bright daylight and no movement). I can split hairs in some of my samples... whereas these are just soft.

If the IQ was close, you would be right... but in this case it's just so far apart that it's easy to see the difference. I don't want to hate, but this is the IQ I expect from a kit lens

There are many other factors tied to "sharp" and the impression of "sharp and clear" than sheer resolution, like quality and contrast of the light, the scene/subject itself, contrast of the lens, and also the balance between high and low frequency contrast for the lens. You also have focus presision.

So I would be very careful to draw any conclusion if I did not have pictures of the same scene, taken with identical lightning and with the same exposure parameters.

Of course your 70-200m and Nikon d850 get you sharper images ....

LessMirrored19: not at all... the pixels on the A9 are larger so it's more "forgiving" to a not so sharp lens. So if a lens is already soft at pixel level on a 24mp sensor, it will be far worse on a D850 or A7Rii or so.

Magnar W: yes you are correct, but these images were shot in just about the most ideal conditions possible (bright daylight, non-moving objects). As an indoor sports photographer I know what bad light looks like. So I'm giving the edge to the Sony here, and still it doesn't look very promising.
Also in terms of focus precision... mirrorless doesn't require focus finetuning due to on sensor AF, so that cannot be the reason either.

The point is... this is not a Nikon vs Sony debate. My point is I know for a fact a 70-200/2.8 wide open CAN and SHOULD do better than this (regardless of brand). They'll probably get it right on the 2nd version.

A consistent test is available at https://www.lensrentals.com/blog/2017/01/an-update-and-comparison-of-the-sony-fe-70-200mm-f2-8-gm-oss/

You know, you're right. Now that I've had a chance to actually take a look through the gallery, not one of the f/2.8 shots is anywhere close to what I would call "sharp". Going by the writeup I thought that the lens varied from "sharp" to "super sharp" as you stopped down, but the f/2.8 shots are soft. They just are.

Kind of disappointing for a lens that bears Sony's highest designation of quality.

No, of course not. But had Sony focused on producing a system around *one* mount, instead of scatter-shooting I-don't-know-how-many variants of bodies and mounts at the wall in the hope of *something* sticking, we might have had it way earlier than is actually the case. And then I might still have been a Sony customer...

Well. You could have got an Sony 70-200 f/2.8 G. They have made it since it was Minolta, and it's a great lens.

The Sony A900 was a groundbreakingly wonderful camera that took the other companies years to equal. But Sony basically abandoned the A-mount for this other craziness. I'd give them two years before they completely forget about the E-mount and go onto something else.

@SmilerGrogan - exactly, you get my point

@BlueBomberTurbo - and there’s the A-mount, the FE-mount, the Alpha DSLR bodies, the DSLT bodies, the ML bodies and so in and so on. It’s good Sony now finally seems to be building a single system. It’s not so good for those that - originally - bought into systems that are now abandoned.

I have to agree. I'm just an "enthusiast" but I'm sure even some pros want to FULLY exploit the amazing capabilities of Sony's APS-C mirrorless tech

Get over it, all three FF manufacturers don't support their APS-C line. Neither Sony, Canon or Nikon. Switch systems or upgrade, because you will be very unhappy over the next few years.
Canon EF, Nikon F, Sony A-mount still have some APS-C lenses out of times FF didn't really exist, but high end APS-C lenses are a niche product

With all of the third party lenses, I'm pretty dang happy with the selection.

It does seem to be the most well rounded lens in it's class. The bokeh quality of the near to near-mid field, and transition areas from in to out of focus are definitely more prime like than the other 70-200's out there. The out of focus foliage is really quite good!

Actually the Sony 70-200 GM has been shown to not be best in class...
Actually worst in class...
It is actually behind the Nikon and Canon offerings, as was widely discussed by Lens Rentals exhaustive test of 10 different samples of this lens all falling short of the Nikon and Canon equivalents.
https://www.lensrentals.com/blog/2017/01/an-update-and-comparison-of-the-sony-fe-70-200mm-f2-8-gm-oss/

Everythingis1: I agree the bokeh looks good on these images, but doesn't stand out to me. No better or worse than on other 70-200's in my experience.

@NexLupus If you're looking only at MTF charts from that testing mothodology, sure. But if you look at all other tests done by others, and also take into account all factors other than MTF, you'll see it's the best all around 70-200mm lens that exists.

I looked at the other tests, and they all showed the Nikon as being the best 70-200 currently available. Which tests were you referring to?

@Everythingis1
You realize that MTF charts speak directly to how well the lens performs in pretty much everything don't you?
Sharpness in the center
Sharpness across the frame
Light fall off
Micro contrast
and finally the characteristics of the Bokeh

But if you are talking about real world tests, I have yet to see any test that compares it favorably against the Nikon or Canon.
Don't get me wrong, it is a nice lens, but class leading in any category; NO

You're missing the little asterisk and the footnote that says 'in the class of 70-200mm f/2.8 lenses manufactured by Sony'.

@Everythingis1
Ok, let’s start with the first.
The author of that review could not believe the results on his first pass, and concluded that he must be doing something wrong as the lens did so poorly against the competition.
The link that I posted is the updated where he retested and actually shot a target. His conclusion like the original was, the lens does not compete against Nikon and to a lesser extent Canon. READ THE CONCLUSION.

Sorry to barge in again Everythingis1, but you haven't posted a link to anything that show the Sony as being in any way better than the Canon or Nikon. You said there were tests that showed the Sony was better. Well, is there, or isn't there?

"You haven't addressed anything I said, other than a post talking vaguely how you can tell about a lenses bokeh from it's MTF, which isn't even a valid talking point a bout bokeh quality in modern lenses.".

O.k, i'll bite. Why isn't that valid? Have the laws of physics changes recently?
Show me a lens with widely diverging sagital and meridonial lines that has great bokeh.

And please, post a link to site that show the Sony besting it's competitors (the DXO test doesn't include the current Nikkor).

If the tests and reviews state that a lens doesn't suffer from spherical aberration to a high degree, then the convergence of sagital and meridonial lines in the MTF chart suggests very good bokeh. It's a fairly logical assumption. That and the fact that bokeh looks fine on the Canon and Nikon. Oh, and the fact that you refuse to point to a single test that backs up your assertions.

@NexLupus
I guess you're probably right. It's just that the guy said...

"If you're looking only at MTF charts from that testing mothodology, sure. But if you look at all other tests done by others, and also take into account all factors other than MTF, you'll see it's the best all around 70-200mm lens that exists."

...and I was thinking that, well, with ALL THE OTHER TESTS to support that assertion, he'd be able to point to at least one!

From the e-photozine links you posted...

Nikon: Generally sharper. Far better chromatic aberration. "Lovely" bokeh.
Sony: Not quite as sharp. Awful chromatic aberration at the edges. "very pleasing" bokeh.

Wow! Well that just proves your point, doesn't it!!! Or not. Are there some "other factors" that make the Sony better, and if so, why not link to a test that illustrates them?

You have issues. I never said the Sony was worst in class. And there you go with "every single test proves" the Sony is better thing again, except that you continually refuse to link to single test that actually shows that. Look at it this way, the Sony scores 26 P-Mpix at DXO, while the OLD Nikon scores 27 (and the new one is considerably sharper).
Since you're the guy who's basing his arguments on "tests" that you can't link to, maybe you should stop with the personal insults.

Screwed up outlier? Are you talking about the test where they measured 10 copies (how many other sites even test more than one?) and found it to be lacking? The same test where Roger has this to say in the comments: 'We have now run MTF tests on several different sets of 70-200 GM MTF charts and the results stand. It's a good lens. It's not the best in class.'?

And yet again...

"every single other test shows show unanimously better results"

It's the last time I'm going to ask, PLEASE post links to all the other tests that show the Sony is the best in class.

@Everythingis1 I'm perfectly aware of what an outlier is. Roger's uncertainty, which he explained in the comments of the original article where the Sony MTFs appeared (the test of the Nikon 70-200), was to do with whether the lens was properly focusing at infinity. His more recent comments mention that they've now tested several batches on several charts as well as the optical bench and their results stand. How is that now an outlier? At this point they've likely tested more copies of this lens than every other site combined. How many other sites have even tested the lens at infinity? I can only think of one other site that even does testing at infinity at all.

I'd also point out that every other test does not say what you seem to think and multiple people have already pointed that out to you.