Image courtesy: Photozone.
Photozone tested the 16-50mm E-mount lens:
The Sony E 16-50mm f/3.5-5.6 OSS broke a couple of records but unfortunately not in the positive sense. Its uncorrected distortion and vignetting figures are nothing short of insane. So Sony obviously doesn’t want the lens to be used without enabled aberration compensation.
And the two Sigma ART NEX lenses also got tested:
– Sigma 30mm f/2.8 DN A MFT at DxOmark (Click here): “In fact, it’s really only an average performer, but at $199, it’s sensitively priced and suitably complements some of the more affordable MFT camera models.”
– Sigma 60mm f/2.8 DN A MFT at DxOmark (Click here): “As for the optical quality, the 60mmputs in a solid a performance. And, at $239, it‘s not inaccessibly priced.“
52 Responses
Sigma 30mm f/2.8 DN A MFT …. Micro Four Thirds Lens version!
Test are not comparable across MFT and Nex System …
Photozone has tested the 19, 30 and 60 DN on NEX-7 so I don’t understand the comment about m43… ?
That short flange distance is the NEX achilles heel for wide angles ? Only the 10-18 seems to manage decent wide angle performance.
Dxo tested the 20mm and found that soft away from the centre too.
because sigma DN lenses were designed for aps-c sensor (NEX & sigma own DPs), on m43 sensors they produce softer images…
on dxo site status for NEX 30 & 19 is “preview”, not tested, only NEX sigma lens has been tested is 60/2.8
It is unfortunate for those of us using the 16mp NEX’s that the two most recent reviews of this lens I have seen have been on the NEX7 and its 24mp body.
So looking at these results: I was about to buy myself a nex camera 5r. The camera seems to be perfect for what I want and seems to be an outstanding sensor compared to say m43 cameras. However now looking at the results of all the latest lens test (I would use this lens quite a bit when travelling) I’m worried about all the issues mentioned in the article. Perhaps I’m better of buying a m43 camera with better lens and may still come out on top? At $600 max for a camera and lens kit what do you thibk I should do? Stick with nex or go to say a panasonic 4/3 camera?
Despite the test result, I still wanna buy NEX3 or 5 + 16-50mm. This lens comes almost for free with NEX body. What’s not to like about freebie? Another advantages are the size and 16mm coverage. I don’t expect any sort of miracle from a kit lens. For quality, manufacturers demanded us to buy their more expensive lenses. Remember, top lens from 43rd is expensive. And they also use in body correction. It’s necessary to keep the lens small.
Dan, For a budget of $600 your decision depends on which aspect of the camera is important for you. If your preference is pocketability then I would recommend Sony RX100. If your preference is image quality at resonably small size then Canon EOS 100D/SL1 with Canon EF-S 18-55mm f/3.5-5.6 IS STM Lens Review may be the right choice. This camera weighs only 404 grams with battery and SD card and the lens weighs 205 grams. The ebay price for the kit is now about US$600 with shipping included e.g. here:
http://www.ebay.com.au/itm/SALE-BRAND-NEW-CANON-EOS-100D-SL1-X7-DSLR-CAMERA-EF-S-18-55mm-IS-STM-LENS-KIT-/221223853769?pt=AU_Digital_Cameras&hash=item3381f81ec9
Here is the review of the lens at Photozone:
http://www.photozone.de/canon-eos/831-canon_1855_3556stmis
The lens gets near 4 stars for image quality vs 1.5 stars for the Sony zoom.
One of the problems with lens tests is that it doesn’t really test the lens in real world situations for extended periods of time. Yes, the 16-50 PZ needs to be corrected in camera or in software (CS6, for example). My copy, at least, yields nice images after the corrections. The Canon you bandy about is fairly close in size to the A55 which, arguably, is at least as good as the Canon. To make a decision on a camera based on a single lens is a little bit much, though if the lens is stuck on it (non interchangeable) that does make sense.
Here are three sample pictures taken with the NEX6 and the 16-50PZ (while waiting on the airport and in an airplane). All shot through less than clean windows.
http://500px.com/photo/40827674
http://500px.com/photo/40624242
http://500px.com/photo/40291624
Probably should have shot with a faster lens (on the aircraft pictures). I wanted to see what the PZ lens can do at slow shutter speeds and high ISO. The shots ended up underexposed to be able to be hand holdable (even with the image stabilization). Cleaned up in CS6. (Typical vacations pics).
All in all, I really like this lens. At least my copy of it.
The real problem with photozone is that they test the lenses on a Nex-7. The Nex-7 is somewhat “bitchie” and not so easy to handle.
The latest E-Mount tests also seems to have some “agenda”…
I don’t know but it feels like E-Mount has to be tested bad @ photozone…
the 18-55mm OSS is superior in terms of IQ and ease of use compared to the 16-50mm, i like some features of the 5R but the AF on m43 is nice too. after i bought the 5R i sold the 16-50 and opted to keep the 18-55 instead so maybe have a look at using that lens?
And that’s the same shit lens on the lunar.
As a m43 shooter, I was a bit envious of the 16-50mm NEX kit zoom. A nice range starting at ~24mm equivalent, amazingly small size for APS-C, cheap price (as a kit). Well, after the photozone review, I’m no longer that impressed.
7.5% distorsion was not unexpected (Panasonic’s own pancake zoom PZ 14-42 is over 6% too at the wide end). However, the brutal vigneting (up to 5 stops!) shows how much that lens is underdesigned. They might as well put a nex mount on a m43 kit zoom!
At some point, regarding lenses size, it’s just physics. I think that for a “dark” and cheap portable zoom, the m43 propositions make more sense right now. There are the usual caveats regarding sensor size, but it’s less of an issue for a zoom that will mostly be used outdoor (at least that’s the way I use a kit zoom: in good light).
http://camerasize.com/compact/#472.335,472.374,33.360,ha,t
Heavy vignetting with p1650 is due to the sensor, not the lens. It will be less issue on newer sensors. Sensor in new RX100mkII is supposedly BSI, this suggests it will hit APS-C domain in a year or two.
Most nex lens releases for Nex will almost always be tested using the nex7 because lens flaws are much more obvious on 24mp than on 16mp.
If it’s only average on 24mp it’s usually pretty awesome on 16mp.
Therefore µ43 lenses should be testet on a 18 mpix sensor.
It is really stupid that they test Sony lenses on 24MP, since there is only ONE such camera in the NEX line.
Lens tests have a long shelf life, as lenses have long life compared with camera bodies these days. It makes sense to test on the highest resolution body, as the test will be usable for longer. These test results will be still used in 5 years time and at that time 24 Mpixels or better will be on even the nmost basic NEX cameras.
Lens performance is a very subjective topic, the test results for the same lens can be quite different across various review sites. For example, the SLRGear review finds the 16-50 to be better than most kit lens. Checkout what Photozone gives to other kit lens. 🙂
http://slrgear.com/reviews/showproduct.php/product/1552/cat/83
“The Sony E 16-50mm f/3.5-5.6 PZ OSS is a compact and inexpensive kit zoom lens that produces decent results, although it suffers from heavy barrel distortion at wide angles and produces images that are only super-sharp at the center. That said, it does better than most kit lenses, to the point that you may not feel the need to immediately rip it off and replace it with something better.”
On the 16mpx nex6 this lens works very good, again very good! On the nex 7 with its 24mpx sensor even the Zeiss 24 2.8 wasnt good enough for me. Thats why I sold nex7 and Zeiss. Nex6 with 16-50 good all round combo. Genarally I use it with correction ( even in Lightroom you can correct its raw automatically) For travel and other general porposes and video this lens with 16mpx sensor is recommended. Small, sharp, good focal range, short minimum focus distance, acceptable bokeh, stabilization
NEX6, RX1, Nikond600, Canon1Ds mk3
Really ??
You very probably had a bad copy of the 24 !!
For me, is nothing to compare with the 16-50 in term of sharpness and optical quality !!
And, as I have touit lenses, the 50 mm and as I have tested lot of emount lenses, I find the 24mm to be the best e-mount lens sio far, and the most versatile with its 16cm minimal focus distance.
You are right. The 24 Zeiss is good lens and is clearly better than 16-50 at 24. But on 24 mpx sensor there are several problem with it : ca, side and corner softness . 24 mpx is to much on apsc at this technical level of NEX system. I wanted to express that the 16 mpx sensor is better choice to date than the 24 mpx on NEX and the 16-50 good and useful KIT LENS 🙂
Dxo mark scores :
Zeiss 24mm f. 1.8 : 22
Sel 16-50mm : 12
……..
The Zeiss 24mm is not good enough, but the 16-50 works ok? Come on.
The PZ1650 is a lens about compromise. From the initial reviews I was very skeptical but eventually I got one because of the compact size. In real-world use on a 16mpx NEX camera it is as good or better than the old kit (which is STILL the kit on the NEX-7).
I like a lot of the Photozone.de reviews but this one shows a really weird bias. In the vignetting section he demonstrates that once the distortion is corrected for those -5ev corners are cut off–and yet he still judges the lens for those dark corners!
Here’s the skinny people: The lens is compromised. It needs digital correction (like many of the mFT lenses). The correction is not baked into the raw file so you CAN see the image without it. With the intended correction applied it makes a great kit lens, especially stopped down. It is amazingly small, fast, stabilized, and silent. It is about the perfect snapshot lens. Like all kit lenses it will be bested by even a moderate prime (the Sigma line). If digital corrections bothers you stick with the SEL1855 kit, it is almost as good but much bigger because it handles all the corrections optically.
“Like all kit lenses it will be bested by even a moderate prime”
That was my point, wasn’t it?
I own the 16-50pz for my nex 5n. The vignetting and distortion are truly astounding especially at wide angles. Lightroom has a profile that does quite a decent job of removing both when processing from raw.(I haven’t tried the in-camera JPG correction)
As ever, it’s a compromise. For me, the tiny size of the lens outweighs the minor inconvenience of correction in PP. The 5n/16-50 is truly pocketable enough for my needs where the 5n with 18-55 was just slightly too big.
i use the 1650 on ma nex 6, and i love it. the distortion is pretty bad, but thats what lens correction is for, photoshop does it automatically for raws, and with jpgs its not even an issue.
and the sharpness is also pretty awesome, not only for this size, but in general. i guess the nex7 sensor just sucks in the corner area, cus the 1650 is not the only lens that scores low in the corners….
they just forgot something important in the 16-50 review:
the given “16mm uncorrected” are in fact not really 16mm but rather 14-15mm, according to many users… in fact, this focal lengh is used to give a 16mm after correction.
So of course you get strong vigneting/distortion at wide position: because the uncorrected wide angle position of the lens is just beyond the official specifications…
One thing is true: this lens is NOT designed to be used without correction. But what is important is only the final result… the photo you get at the end… This lens is not design to care about uncorrected result.
+1
+1
+1 +1 +1
+1. I use this lens on NEX6. It’s a good, light combination that I always bring them with me.
I don’t think Photozone “forgot” anything, it is one of the better review sites. What readers forget though, is that he is also reviewing the best lenses out there, so it is all relative. Of course the 16-50 would like crap compared to the 135mm STF.
im sorta glad i sold the 16-50 pretty much after a week with it, i don’t really like the “zoom by wire” thing as it was a bit too sensitive for my liking and it was in the wrong direction (i got one of the first copies with my NEX 6 within a month of release). i now still use the 18-55 with my NEX 6 and NEX 5R 😀 i use to want the 16-50 so bad but after playing around i prefer the 18-55 even if its bigger
All right. Terrific.
But, concerning the FF nex you’re talking about since one month? “Stay tuned on SAR”, “follow me on twitter, fbk”, “something big coming soon?” Something new?
for my intended uses , photozone tests are useless since they focus too much on corner and extreme edge performance rather than center resolution and contrast level at wide aperture range.
so I shun them altogether and I personally dislike the guy running the site.
Lol, and what they should focus at? The center?
well, I meant Photozone.de put too much stress on extreme edge resolution,which is kind of less critical point to real life image quality(center resolution is much more important most of times).
some of us value real life practical image quality more than minor difference in extreme edge resolution.
what about its flare resistance ? what about light transmission? what about its coma characteristics?
photozone ignores all of these important aspects of practical image quality.
so I much prefer SLRGEAR , Lenstip and DXO over photozone.
Photozone (or SLRgear for that matter) give you some data and it is up to you to judge its relevance. The mroe reviews you read, the more complete the information you have. Although, I personally completely dislike DXO and LensTip.
Center performance is generally where any lens is good. Stopped down, plenty of lenses will deliver amazing quality. Example, the 1650 when stopped down has center resolution probably near that of the 135 STF, a rather amazing lens. Thus, corners are very important, because that is generally one of the big differentiators of a good lens and it is of practical importance (you can’t make up resolution in post). Yes, Photozone didn’t speak of coma or flare, sometimes he does, but for this lens I doubt it makes much sense. If you really are worried about those specs, you wouldn’t be buying a tiny cheap zoom.
I bought the Sony NEX 3N with 16-50 zoom as kit. I find none of the issues with the lens discussed. When I placed the 16-50 on my NEX7, then I have some noticeable problems, but then these problems can be corrected with software. So YMMV!
I guess it depends on what you can or willing to notice or accept.
I have a number of camera bodies, Nikon and Sony mostly, and I when I can I choose a combo for shooting different subjects. The 16-50 is not my choice for landscapes or even portraits. I can use it for that, but it is not ideal. On the other hand, for close ups of flowers or casual street photos, I like the 16-50 at lot. It has been optimized for NEX3N with its zoom lever and lens wheel. It works well to create a shallow depth of field in closeups, and the zoom makes it easier to work with in framing the picture. I imagine its optical characteristics were worked over in the software of the camera body (NEX3N) to minimize distortion and vignetting.
Precisely, horses for courses.
“I personally dislike the guy running the site”
So, you know him and dislike him personally?
I don’t know him but I do find the site useful
I never met him but I ‘ve exchanged emails and discussed many issues on his fora and I felt him kind of an anal person, not so nice or friendly at all.
I personally like lenstip reviews much better and they seemed much kinder and more friendly when I questioned them something about their testing method in emails.
But with that said , I think I love DXO , SLRGEAR and LENSTIP all of them are good in some ways and not good in some ways.
So it is all personal taste and feeling against the guy , I just dislike him and his site , that’s all.
I dislike photozone , I think I have right to dislike them , I never siad hate or anything.
Finally, I do not trust their tests because they usually test only one copy of most of lenses, and which is useless since there are so many level of copy variation issues as lensRentals once explained.
I read the explanation of the DXOmark testing when the site came online. It did not make any sense to me. I reread a large portion of it a few months ago. It still does not make any sense. I do not think I am alone in that conclusion. I also doubt that most people have taken the trouble to try to read and understand it. I think people accept DXOmark results simply because they publish an overall number which is kind of meaningless for just about everything, but people do like simplicity. I think the assessment of the lens at SLRgear is a lot more fair, detailed and objective with graphs of sharpness and softness over the whole frame at each f-stop and varying focal lengths. They said the 16-50 was better than most kit lenses and provide you with the test results to confirm that.
I’ve reread the test at photozone and compared it to the review of the 14-42 Panasonic Pancake and the Olympus 14-42 Kit.
It’s plain ridiculous…
Just look at the uncorrected distortion the Olympus 14-42 Kit has… Look at the vignetting uncorrected both lenses have. But in the end the corrected ones are taken into account in the final verdict.
You have one guess what’s happening with Sony…
(There are also the 2Mpix divergence, which can alter the outcome especially in the corner…)
Just look at other sites like lenstip or slrgear. They’re much less biased!
Panasonic PZ 14-42mm (pancake kit lens) has far less uncorrected vignetting than the NEX 16-50mm (-1.5 EVF vs -5(!) EVF) on the Photozone tests. Distorsion at the wide end is bad for both (6.8% on the Lumix vs 7.5% on the NEX) and one could explain that NEX is a little worse because it is wider (24mm equivalent vs 28mm). The resolution drop off center is also significantly worse on the NEX, at all focal lenghts. Of course, we cannot compare absolute MTF numbers accross systems with different pixel counts and sensor sizes.
Overall, both lenses have bad optical performance, the NEX being the worse (1.5 stars vs 2.75 stars for the Lumix). They have good justifications for those ratings. It doesn’t mean that one cannot use these lenses.
If it can make you feel better, note that m43 lenses usually get bad reviews on photozone too, even expensive ones (see Lumix 12-35mm – 3.25 stars). They strongly dislike digital correction.
Just one Sample:
Panasonic Lumix 7-14
http://www.photozone.de/olympus–four-thirds-lens-tests/530-pana_714_4?start=2
Sony Sel-1018
http://www.photozone.de/sony_nex/829-sony1018f4oss?start=1
Any questions? The Sel-1018 outperforms on a Sensor with higher resolution the Pana 7-14. Only the vignetting seems to be a problem but it again sounds like they compared the uncorrected with the corrected values.
Just look at the MTF-charts, this is just unbelievable! The distortions are HUGE at 7mm for the Panasonic lens!
Just for information:
http://www.dxomark.com/index.php/Lenses/Compare-Camera-Lenses/Compare-lenses/(lens1)/1084/(brand)/Sony/(camera1)/0/(lens2)/536/(brand2)/Panasonic/(camera2)/0
There it looks quite different to photozone!
“Thus we feel that Sony was somewhat too aggressive here. It would have made more sense to stick to a manageable range of -say- 20-50mm – just like Samsung did it with its really desirable pancake zoom lens.”
what a biased test. samsung 20-50mm doesn’t have optical stabilization -.-”
sure the 16-50mm isn’t anything spectacular in optic performance but for the size and weight and optical stabilization and overall performance it’s really amazing.
please show me a similar lens for apc-s sensors. oh wait… there isn’t any.