Equipment

An Update and Comparison of the Sony FE 70-200mm f/2.8 GM OSS

By
Published January 6, 2017
Tags: , , ,

About a month ago, Roger posted his MTF bench results for the Nikon 70-200mm f/2.8E FL ED AF-S VR (TL;DR: it’s optically superb), and down in the comments there were some requests to compare it not only to the Canon 70-200mm f/2.8L IS II in the article, but also to the Sony FE 70-200mm f/2.8 GM OSS. There were MTF charts linked:

These are not very good looking results, especially for a $2600 lens, but Roger wasn’t sure if they were correct results. In his comments after these charts he said:

Let’s kind of keep this here for right now – I haven’t published it because I’m still a little uncertain about the results. Sony has suggested that a change in cover glass thickness might improve the results some. Not dramatically, but some, off-axis. This lens also has to focusing motors and we have to focus it electronically via a camera to test. I’m not absolutely certain that ‘setting it at infinity focus on the camera’ and ‘manually focusing on a an object at infinity’ are absolutely the same. So take these lab results with a grain of salt. On the other hand, they do seem to agree with what we see in real world results.

I try hard not to put out anything until I’m just absolutely certain our results are correct. We’re doing some stuff here that’s pretty cutting edge, honestly. No one does 4 rotation MTFs, for example. I’m pretty certain these are good results, but not absolutely certain. So I’ll post them in this discussion on my own site but I’d rather not see them reproduced all over the internet yet.

Someone suggested we try all three 70-200s on their respective camera bodies and shoot the same detailed scene with each, then share the files. But that would involve three different cameras and wouldn’t really be an apples to apples comparison, would it? Roger suggested he have one of our photo techs shoot all three lenses on a Sony camera with adapters, admitting that adapters add another variable, but there would be some good practical implications.

So I did just that. I took all three lenses, the Canon 70-200mm f/2.8L IS II, the Nikon 70-200mm f/2.8E FL ED AF-S VR, and the Sony FE 70-200mm f/2.8 GM OSS, and I shot our test chart with all of them on the Sony a7R II. For Canon, I used a Metabones T Smart Adapter IV, and for Nikon, I used a Novoflex Nikon to Sony E adapter. I manually focused all three lenses to get the most consistent results, and here’s what I got:

Full Resolution Examples are Available Here

If you view the charts at 100%, you’ll see that they’re consistent with the MTF charts Roger generated for all three lenses. The Sony just isn’t that great, and that’s really disappointing considering the price tag and how long customers have had to wait for that lens to be available. But if you need f/2.8 and working autofocus, it’s really the best option out there for Sony mirrorless cameras.

Author: Joey Miller

I’m Joey. I love cameras, especially old film cameras, and I can’t remember the last day I didn’t take a photo. Digital cameras are great, and they keep me employed, but I also still like processing my own film. I’m stuck somewhere in the middle. I shoot every single day, no matter what.

Tweet
Posted in Equipment
← 
Newer Comments  →
  • Hakann Vatansever

    Roger means all the lenses in their stock is in similar condition.He cant say it loud but i think sony again screwed up with production quality….do not be early owners of this lens.period.

  • Hakann Vatansever

    85gm has no scratching issue.its just the noise of af motor used.nothing is being scratched…check LensRental’s deep investigation about this phenomena.

  • Jekabs

    Look at the corners on the right side of that image. There is something very wrong with test or the lens.

  • Zé De Boni

    Guy! Do you realize what you just said? The optical path of a DSLR lens with the mirror up position while the image is being captured is exactly the same as on a mirrorless with the proper adapter. There is no optical element on the way (just the same amount of air!) and the only thing that may cause a different result is the internal reflection, hardly a problem if the adapter has good quality.
    There is absolutely no corner degradation when you use a DSLR lens on a mirrorless camera. In my own practical use and tests I got better results from my Minolta HS 300mm f/2.8 (handheld and on a tripod) on the A7RII than on the A99, despite the more demanding 42 MP sensor. That, I suspect, was due to the IBIS, which I recommend even when mounted on a tripod. My interpretation is that there is a subtle loss of sharpness because of the shutter action on the A99, like it is the case on the A7R.

  • Mr. T – I don’t disagree the bad copy of the Sony, but this was middle of the pack from a fairly large set we tested on the bench. On the other hand, as we’ve mentioned, these were all early copies.

  • We picked an average one of the samples we have, which are all from very early shipments. As I’ve mentioned, though, I still wonder if later lenses are going to be better than these early ones.

  • m hart

    This is a bit distressing if you’re on a B&H wait list for this lens. Have been very pleased with the GM 85 and the 24-70. Was looking forward to the 70-200 GM. Do I bail?

  • Sator Photo

    I suspect that there may be a major methodological flaw in the test here in that the Nikon and Canon lenses are mounted via adapters to a Sony a7RII.

    DSLR lenses (with the mirror box in the optical pathway) are not designed to work on a mirrorless mount with an ultra short flange distance because there is nothing in the lens formula to reduce the angle of incidence of light in the corners. So you get corner performance degradation when you adapt DSLR lenses to a mirrorless mount.

    I would suggest that Roger measures MTF performance of a DLSR lens first at its native flange distance, and then adapted to a mirrorless flange distance. Example Sigma 35mm f/1.4 with and without the new Sigma FE mount adapter. I would hazard a guess that the adapter will cause significant performance degradation. This is why Sony doesn’t merely shorten the flange distance of A mount lenses when designing E mount lenses, because doing that degrades performance too much. So instead, they design entire new optical formulae from scratch for the mirrorless mount.

    A mirrorless lens is just not your grandfather’s DSLR lens with a shortened flange distance!!!!

    A better option when shooting test charts would be use the a7R for the Sony lens and a D810 for the Nikon, so that both are shot using a Sony 36MP sensor but with the lenses mounted natively on the optical system they were designed for.

  • I think that was their plan, and then reality got in the way…

  • Yes, if absolutely nothing else can be agreed upon, I think it’s safe to say that Sony has some QC issues. Their 35mm f/1.4 was (is?) all over the place wrt corner sharpness, a new 50mm f/1.4 turned up on eBay (because Sony said the cost to fix the focusing “unit” was more than the cost of a new lens), and their 85mm f/1.4 had (has?) issues with noise, and something scratching against the barrel during focus.

    My 3 main cameras are Sony mirrorless, so while these issues are frustrating, I’d rather Sony addressed these issues publicly (for once) and would start standing behind their expensive, premium gear.

    Pretending like issues don’t exist only breeds mistrust and a sense of unease about putting down thousands of dollars for a new lens.

    For instance, did Sony address the issues with its 35mm f/1.4, or is there still a 1 in 3 chance your $1,500 lens is going to have soft corners?

  • Chris

    The center sharpness of the Sony at 200mm is noticeably soft, before you even compare to the other lenses. Unacceptable really. I suspect my 24-240 is sharper on center at 200mm than this lens. And if it’s slightly decentered, you wouldn’t expect center sharpness to improve much with a properly entered copy. So at least at 200mm, the test charts seem to support the MTF results.

  • Chris

    Based on Sony’s track record, I would bet it’s decentered. This issue has become so common with Sony lenses it’s disturbing.

  • Chris

    I’m no expert in lens design but if the flange distance is too short, the lens designers can move the rear elements out from the sensor plane to effectively duplicate a typical DSLR flange distance regardless of the actual mount. So I’m not convinced flange distance is a factor.

  • m r

    I suppose several 70-200 gm were tested? Dumb question I am sure. Little disappointed (mainly because of price/performance aspect rather than absolute performance).

  • David Kilpatrick

    Although I have four decades plus as a lens reviewer, I’ve never had access to the number of samples or the sophisticated methods used by Lens Rentals. I would only raise the following points: first, the cover glass issue, which is real (even between Sony models): secondly, the focus in these lenses generally is only maintained and locked as long as the lens is under power and it seems the lens once focused may be removed from a body and placed on a test rig, which would leave the focus unpowered and perhaps able to shift slightly; thirdly, the OSS even when inactive requires power to maintain the centering of its moving components, which do not ‘park’ perfectly optically centered when powered off; fourthly, this lens uses floating/compensated elements or groups and depends on an exact mount to sensor register collimation for performance corner to corner. If you were for example to focus this lens on infinity, then mount it on a manual rig and sensor-focus on a closer target leaving the lens set to infinity, the outer field resolution would be degraded (it would only be correct if the lens was refocused on an infinity target). I don’t know any way of testing this lens except on an A7RII ‘blueprinted’ for mount collimation and sensor SSI carriage adjustment. There are simply too many variables, too many moving parts, and too many factors which depend on feedback while powered up.

  • C_S

    i have bought both gm zoom lenses and both have issues and are back for adjustment or replacement. the 24-70 was heavy decentered my loaner lens ist much better and incredible sharp far better than my canon 24-70 2,8 II. the 70-200 has another strange problem i have never seen before. my dealer suggest that it is maybe related to a faulty IBIS for me it is simply also a decenter issue. but both lenses are very close to primes when stopped down a little and in every respect far better that my canon L glass. but for me it is clear that sony has some serious quality control issue.

  • Zé De Boni

    I wonder how this article was published at LensRentals, where we usually find the best descriptions of lens performances, and the only place where I’ve seen tests of sample variations, which for me tells a lot from a product!
    At first look I was shocked by the uneven lighting of the images. In a glance I questioned what the heck is that “negative vigneting”? One second later, looking at all images from the 3 lenses, I realized that the lens tester could not perform the simple task of lighting a repro subject. So far, if he cannot achieve the simplest task, what to expect from all the other technical issues needed for a reliable experiment?
    I could comment other aspects of this post but as the procedures are not trustful, that would be a waste of energy and time.
    I will just rather disregard and forget what is published here.
    Sadly, this experience will make me think twice before following a link to LensRentals. It is a shame that they did not notice what I am pointing out!

  • Guido

    These lenses are usually more used at the long end. If you’re primary looking for something around the short end you will use something like one of the many excellent 85mm. Much faster, much cheaper, less weight. But there are not so many 200mm primes on the market. At 200mm the Sony is clearly much worse than the Nikon and worse than the Canon. It’s not a worse lens at all, but regarding the highprice tag it’s very disappointing.

  • Magnar W. Fjørtoft

    What about testing the lenses at infinity? At a long distance? At test chard distance? And manybe even closer distances? How would this affect the test results?

    Also, how will autofocus precision (!) affect real world performance, using a camera like the Sony A7rII versus a DSLR with less precise phase focus?

    From what I can see, all 3 lenses are great, and any of them would for sure do well in the hands of a competent photographer. For real world use, I think aspects like autofocus speed and precision, distortion, vignetting, handling, etc., would be of more interest than slight differences in resolution wide open. What about bokeh, which is one of the priorities from Sony for this lens? People/customers would probably not be able to tell any difference between these lenses even when inspecting large format prints at a close distance!

  • Mr T

    Looking at all four corners and center.

    The Sony is looking strong at 70 mm in all corners and center ok. Canon worst of the three, corners bad compared to the others.

    200 mm Sony is soft in center and left side. But right side corners are clearly better than Canon. Very close with Nikon here. Nikon overall solid performance at 200 – all corners and center.

    My conclusion. Bad copy of Sony lens. Nikon clear winner.

  • Zinc

    Took an another look at those pictures, totally agree with you!

  • Brian Smith

    BTW, thanks to Joey for posting these and making the effort. Much appreciated!

  • Brian Smith

    Yes, I pointed this out as well. The Sony at 200 is either decentered or titled or the test is not parallel.

  • Brian Smith

    I have to agree with the previous posters. How did you conclude the results mirror the MTF results previously posted? At 70mm the Sony beats the Canon significantly and is very close to slightly better than the Nikon. At 200mm you can easily see the Sony is either titled/decentered or the test is not parallel. The right side of the Sony at 200mm is much sharper than the left. Based on the right side along it again matches the Nikon and beats the Canon.

    Ideally all three lenses would be tested for decentering/tilt and then an outside shot at infinity (with plenty of top/bottom, left/right detail) would be taken. This setup would rule out decentering issues and/or alignment issues.

    Perhaps you labeled the images wrong, but from what I am seeing the Sony beats the Canon and matches the Nikon. Other than the decentering at 200mm (which IS an issue) the Sony performed very well IMO which does NOT correlate with the MTF charts posted heretofore.

  • Sator Photo

    The other oddity about the Sony 70-200mm f/2.8 for FE mount is the way it is formulated with two aspherical elements. Most designers avoid aspherical elements in 70-200mm lenses because of its adverse effect on bokeh. The reasons for using aspherical elements might include improving on acutance and for reduction of lens size. Due to the need to make mirrorless lens designs as telecentric as possible to reduce the light angle incidence in the corners, fast mirrorless lenses for ultra-short flange distance mounts like the FE mount tend to have blow-outs in size, which is why they are usually made with slower maximum apertures, making size reduction a likely principle motivator driving Sony to the use of aspherical elements. Yet despite resorting to the use of aspherical elements, Sony have been thoroughly beaten by Nikon who have no asphericals in their new FL 70-200. Canon will soon follow in Nikon’s footsteps with a refresh of their 70-200 and it will probably truly leave the Sony behind in the dust. It highlights the fact that one of the key weaknesses of mirrorless mounts with ultra-short flange distances is that telecentric lens design results in blow outs in lens size thus imposing engineering constraints when designing lenses with ultra fast apertures. Sony might do better if they stuck to slow lenses for the FE mount and focused on fast lenses for the A mount.

  • Zinc
  • Zinc

    Huge difference at 200mm between Nikon and Sony. Sony should drop the price below 2000….

  • picbod

    I downloaded your files of the 70mm results for Canon and Sony and displayed them, untouched, at 100% in layers in Photoshop. Unfortunately for your conclusions, the comparison significantly favours the Sony lens. Huge difference actually, especially at the edges. So, I am perplexed how you could use your chart shots as evidence of poor Sony performance. Perhaps you mixed up the chart labels? And for the record, I’m a longtime professional Canon shooter who has used Canon 70-200 2.8 IS lenses on a daily basis since they were first introduced.

  • Adam Sanford

    [Full disclosure, I’m a Canon guy] Your 100% samples don’t present such a cut and dry indictment of the Sony vs. the Nikon to me. They don’t look nearly as far apart as your prior published MTFs do.

    But: focus by wire and not clearly sharper for $500 *more* sure seems a big whiff to me.

  • ZM

    That hurts

←  
Newer Comments  →
Follow on Feedly