Sony ZEISS FE 24-70mm f/4 ZA on Sony A7R

Sony ZEISS FE 24-70mm f/4 ZA on Sony A7R

This is a beautifully constructed and well sized kit lens for the A7R, how does it perform?

Having heard stories on how large this lens, I was pleasantly surprised to see it is no larger than the 28-70mm kit lens. The build quality is excellent like the other ZEISS lenses with nice alloy metal which is high quality yet light and durable. It isn’t soft aluminum like on the NEX kit lenses. The hood is nice quality and flocked on the inside to reduce reflections, and snaps positively into place.

The zoom range is a wide standard zoom at 24-70mm. This solid lens tips the scales at 430 g, and has 12 elements in 10 groups with lots of special elements. It has 5 aspheric elements, 1 low dispersion element. It has an average for class minimum focus distance of 0.4 m with a max magnification of 0.2x. It uses 7 rounded blades for a circular aperture at large openings. The 67mm filter threads are standard on many f/4 lenses.

This lens is pricey for a fixed aperture f/4 professional quality lens, does it live up to the high standards?

SAM_1416 SAM_1417
SAM_1420 SAM_1419
SAM_1421 SAM_1422

Optical Performance

24 mm Corrected (ACR 8.4 RC Profile used in LR 5.3)

SEL2470z_24mm_A7r_files001
SEL2470z_24mm_A7r_files002
DSC01218_24 mm_0.4 sec at f - 8.0_YBR50_18_multi_cpp
SEL2470z_24mm_A7r_files003
MTF20 Chart maxes at vertical resolution of the sensor which is 4912 LW/PH for A7R. Anything above that is meaningless. The MTF50 chart is set at the value where the best lenses perform in the center, or about 3850 LW/PH on the A7R.
Note: 4k video is 2160 LW/PH
Note: HD video is 1080 LW/PH

34 mm Corrected (ACR 8.4 RC Profile used in LR 5.3)

SEL2470z_34mm_A7r_files001
SEL2470z_34mm_A7r_files002
DSC01249_34 mm_0.4 sec at f - 8.0_YBR50_18_multi_cpp
SEL2470z_34mm_A7r_files003
MTF20 Chart maxes at vertical resolution of the sensor which is 4912 LW/PH for A7R. Anything above that is meaningless. The MTF50 chart is set at the value where the best lenses perform in the center, or about 3850 LW/PH on the A7R.
Note: 4k video is 2160 LW/PH
Note: HD video is 1080 LW/PH

54mm Corrected (ACR 8.4 RC Profile used in LR 5.3)

SEL2470z_54mm_A7r_files001
SEL2470z_54mm_A7r_files002
DSC01258_54 mm_0.4 sec at f - 8.0_YBR51_18_multi_cpp
SEL2470z_54mm_A7r_files003
MTF20 Chart maxes at vertical resolution of the sensor which is 4912 LW/PH for A7R. Anything above that is meaningless. The MTF50 chart is set at the value where the best lenses perform in the center, or about 3850 LW/PH on the A7R.
Note: 4k video is 2160 LW/PH
Note: HD video is 1080 LW/PH

70mm Corrected (ACR 8.4 RC Profile used in LR 5.3)

SEL2470z_70mm_A7r_files001
SEL2470z_70mm_A7r_files002
DSC01279_70 mm_0.5 sec at f - 8.0_YBR50_18_multi_cpp
SEL2470z_70mm_A7r_files003
MTF20 Chart maxes at vertical resolution of the sensor which is 4912 LW/PH for A7R. Anything above that is meaningless. The MTF50 chart is set at the value where the best lenses perform in the center, or about 3850 LW/PH on the A7R.
Note: 4k video is 2160 LW/PH
Note: HD video is 1080 LW/PH

Comments on the Results

I have decided to publish corrected results when possible going forward, as the manufacturer designed this lens to be used corrected. I wouldn’t take a lens element group out of a lens and say it had bad distortion: the software correction is part of the lens. I use whatever Lightroom profile is available (in this case ACR 8.4 RC copied over to Lightroom 5.3). In general pincushion correction will hurt central sharpness and barrel correction will hurt edge sharpness.

Looking at the resolution, this lens resolves excellent in the center at 24mm, but by 70mm it is only good or very good. Resolution in the center is pretty flat at any aperture. Some of the reduction in resolution is from the pincushion correction, but most is typical for a kit lens how they tend to drop towards 70mm.

Most of the frame (partway) is pretty average performance. It is good to very good at most settings.

The corners… I will be honest. The corners at 24mm are poor and never improve. This is not due to field curvature, there is almost no field curvature at 24mm. This is because the corners are soft. At 35-50mm performance is pretty flat, but field curvature spoils the corners by 70mm (field curvature visible at even 50mm, which is why stepping down helps). So the field curvature is corrected out in my graphs (best focus for corners), in real world images your corners at 50-70mm will not be as good unless you step down a lot or focus partway out to reduce the effects. The dip at f/5.6 in the corner is because it is hard to always get best corner focus with a lens with excessive field curvature as is present at 70mm on this lens.

Distortion is negligible at any setting (corrected). Uncorrected it is extreme, but this is not how the lens was designed to be used.

Lateral CA is negligible at any setting (corrected).

Flare resistance: Good, a strong point of this lens.

Field curvature: Noticeable at 70mm, which is part of the corner performance problem. Be sure to focus on your subject as opposed to focus-recompose.

Focus performance is a little on the slow side in indoor lighting. Outdoor it is quick and positive.

Measured Focal length at 24mm and 1:35 is about 26 mm, corrected.
Measured Focal length at 70mm and 1:35 is about 74 mm, corrected.

Pros and Cons

Effective Resolution 24mm

HD Video = 2 MP, 4k Video = 8 MP

Aperture Weighted Center Partway Corner
f/4
22.6 MP
31 MP
15.8 MP
9.1 MP
f/8
22.9 MP
30 MP
18.4 MP
7.5 MP

Effective Resolution 34mm

Aperture Weighted Center Partway Corner
f/4
24.4 MP
29.3 MP
21 MP
14.8 MP
f/8
23.4 MP
27.6 MP
20.8 MP
14.7 MP

Effective Resolution 54mm

Aperture Weighted Center Partway Corner
f/4
20.4 MP
23.9 MP
17.4 MP
15.6 MP
f/8
22.2 MP
24.5 MP
20.6 MP
17.6 MP

Effective Resolution 70mm

Aperture Weighted Center Partway Corner
f/4
15.7 MP
18.6 MP
14.2 MP
8.7 MP
f/8
19.3 MP
23.1 MP
17.4 MP
10.1 MP

Bottom Line

I am harsh on this lens because it is a $1200 US lens, and it doesn’t perform well over much of the range. This lens does not live up to being a professional quality lens. I was a bit neutral on the A mount 24-70mm f/2.8 lens, but it is a much better lens than this kit zoom. Honestly many APS-C cameras with 16-xx lenses will do better than this lens, the Fuji X 16-50mm (which is a very cheap kit lens on an APS-C camera) is better in the corners (about 3000 lw/ph) than this $1200 ZEISS which should be seeing an advantage due to the larger full frame sensor. This ZEISS will do better over the rest of the frame with the larger and higher resolution A7/R sensor.

The lens will also work well for 4k video (might notice a bit of softness in the corners), so it might be good to consider if video is your primary concern, but don’t think it will be any better than the other kit lens (28-70mm) in this regard.

I don’t have links to Amazon or BH for this lens as I don’t recommend buying it.

Gallery

Default light room settings from RC 5.3.

Gallery: http://viking79.zenfolio.com/p39285487 (download all originals by clicking on one, hovering image, menu in upper left, download, all originals)

The gallery images are © 2013 Eric Tastad, and may only be used for personal evaluation of the lenses. Click the play button and click “visit gallery” in the upper right to go to Zenfolio, to download full size images click on an image, hover over an image, and hover over menu, download, and click all originals.

9 Comments

  1. DBM says:

    Great test. Pretty much what I’m seeing from the lens.
    Except: I don’t have the same take home message. Its actually the cheapest 24-70 at f4 that there is. Similar to the Nikon 24-70 except at 70 (but better in the centre at 24)
    How much better could it be at this size weight and cost? It’s very good except at 70 (60 is a lot better) and the corners at 24. And the 24 is really useful; just make sure the extreme corners don’t matter too much in the image.
    Of course whether its worth it is a judgment which is very personal. I find it invaluable for multi day hiking; and can live with its limitations – and I suspect that its limitations are almost necessary for its size, weight and cost.
    When not hiking, it’s primes….

  2. DBM says:

    Oh: when I said similar to the Nikon etc in line 3 I meant in quality, not cost, of course…

  3. Bene says:

    Great review. Thanks a lot! It would be also very nice to see how this lens performs on the A7, compared to the kid lens. Do you plan to review the lens on this camera, too?

  4. viramati says:

    Interesting review and I really have to wonder if there is not significant differences in performance depending on the copy of the lens. My first copy was soft on the left and not particularly good in the corners but my 2nd copy shows significant improvements even in the corners at 24mm (at least at mid to long distances). I have been using the lens profile preset for the FE 28-70/3.5-5.6 in Lr5.3 which does a pretty good job of correction but of course hope for an update soon to get the correct profile. I also wonder if the recent firmware update where they say they have made alterations to the IQ has made a difference tot he jpeg engine when it comes to this lens. By the way I am using the A7 not the R which may help because of it’s slightly lower resolution

  5. E says:

    Will you test this lens on the A7? I have a hunch the corners might be better, similar to what was found on the NEX-5/NEX-7 tests.

  6. admin says:

    Probably not since I rented it. I ran out of time to run the data for A7 before returning the lens, but it would have been nice to have both. My hunch is the A7 would do at least as well, maybe slightly better, but I don’t think it is going to be a case where it is great on the A7 and terrible on the A7R. I did use the lens for photos with both cameras and don’t notice a big difference between the two, but I will have to go back and look and see if I have anything I can compare.

  7. E says:

    Thanks. For some reason I thought you had both cameras. Photozone’s numbers for the 18-55 kit lens show great centers and weak corners on the NEX-7. However, on the NEX-5, the centers were predictably not as good, but the corners in some cases were remarkably better. I thought a similar effect might happen between the A7 and A7R. I’m looking for a reason to buy this lens, but I’m having a very difficult time pulling the trigger.

  8. tecnoworld says:

    Great review, as always.

    Please provide a similar for the nx16-50 f2-2.8. I have a feeling that it’s a better lens than the sony.

  9. Matt says:

    Nice review. I’d be interested to see how this stacks up against the 28-70 kit lens which is much cheaper. DXOmark seems to think they’re fairly similar in terms of performance/IQ.