4.18.2016

Testing new lenses depends more on how you intend to use them and less on "scientific" parameters. Do your own tests!

The lens on this camera is a 24-70mm f4.0 Vario Tessar T* FE OSS

The 24-70mm f4.0 Vario Tessar zoom lens was released into the market in 2013 at a time when the highest res, high performing mirrorless camera from Sony was the Nex-7. At the time of its release there was much excitement in the photography world for just about everything new. Reviewers immediately put the lens on the Nex-7 (cropped frame) and started writing reviews. Had they waited a month or two they could have tested the lens with the full frame A7.

What they found in early tests was nothing more than confirmation that the sensor in the Nex-7 could be....picky. Very picky. There are many lenses that did not work well with that sensor for whatever reason. It could have been the way the light from the lenses hit the edge pixels or even a design issue with the cover glass. At any rate the reviews of this lens were lukewarm, at best. Most said, "Sharp in the center and soft on the edges..."

These early reviews, like most first impressions, are the pervasive signposts that linger in our collective subconscious where this lens is concerned. 

When I was deciding which short, standard, zoom lens made the most sense for me I looked at the results that DXO got when using the lens with a modern, high resolution body, the A7R. As I recall the overall numerical score they got was around 25 which is quite good for a zoom lens in their rating system. I bought the lens without trepidation knowing that I could return it to my local dealer if I was not satisfied with its performance. 

But I guess my real point here is that everyone has some different idea of what constitutes a "high performance" lens. I want high center sharpness but I am less concerned with what's going on in the corners of the image frame. All lenses are compromises and while you can have a lens that is corrected across the frame you may find that the design requires that there not be a "peak" of sharpness in the center; instead the uniformity requires a slightly lower overall sharpness that is more homogenous. 

Which performance metrics matter most for you? If I were shooting architecture I might worry about corners that were less sharp than the center of a lens. If I shot flat copy work I would be looking at a whole different range of lenses. But am I saying that the Vario Tessar really does perform poorly?

Not at all. While I haven't had time to test it extensively I took advantage of a quiet Sunday evening to put the lens on the A7R2 and to shoot lots of stuff around the studio. I tried the lens wide open at all focal lengths as well as stopped down to f8.0. I used it with flash. I used it on a tripod with LED lights and I used it handheld at 12,000 ISO. I like the lens and find it to be a very good performer. 

I'd be happy to use it for portraits, even wide open. 

I think that what people were seeing in early days was a combination of mediocre focusing on the first round of A7 camera and a mental predisposition to be critical based on the weight of the web reviews as conducted with the older Nex7 cameras. My actual experience tells me that the lens is a good match for the way I shoot and will work well on either the full frame A7R2, or on the APS-C a6300.

I will say that I like the color rendition and the overall imaging characteristics of the lens very much. I also like the compactness and the lesser weight when compared to the f2.8 version. Everyone's mileage will vary but I think it's always a good idea to find out for yourself. 

13 comments:

Kirk Tuck said...

Have packed up a couple of lenses and the A7r2 and am heading out to a healthcare company to shoot a quick executive portrait. It will be my first excursion with this collection of gear. Wish me luck.

Peter Ziegler said...

Those less-than-stellar reviews kept me from buying that lens for a long time. It wasn't until I'd seen some examples from others on the net that I decided to go for it. I'm glad I did.

Unknown said...

I avoided buying the FE 24-70mm because of the poor reviews and negative chatter on the photography sites. I used an adapted Canon 24-70mm instead; a nice lens, but bulky with the Metabones adapter. I bought the FE 24-70mm several months ago second hand from a friend for $700 and I have been very pleased with it on my A7R2. I expected it to be a dog, but the IQ is much better than I had hoped for. I'm selling the Canon lens now as the IQ of the FE 24-70mm is very good and it is much lighter and compact for travel photography. I totally agree with you point about testing lens for the way you use them versus strictly going by reviews!

ODL Designs said...

Good Luck Kirk :)
It will be interesting to read your experiences with the new kit!

tnargs said...

Lab tests also showed this lens to be below expectations. So I think it is a little bit generous to put it down to 'early reviewers and a NEX 7 body with funny characteristics'.

The slrgear review on a A7r body summarised sharpness: "Despite carrying the Zeiss branding, which is typically indicative of high-end results, we felt that the Sony FE 24-70 Zeiss lens fell a little short of our expectations. At 24mm, the lens displays good sharpness right in the center of the frame, even wide open, but outwards and especially in the deep corners it's noticeably soft. Surprisingly, even stopping down doesn't improve the corner softness at 24mm, and by ƒ/16-ƒ/22, diffraction comes into play and reduces sharpness all around even more."

Sure, 'do your own tests for your own usage' is good advice, but don't make out that the lens is as good as the opposition and it merely 'got a bad rap'. The above bench-test's overall conclusion? "When it comes to RAW-file image quality, however, we found ourselves a little disappointed. With soft corners at essentially all focal lengths (and even stopped down), plus strong vignetting and distortion, the Sony FE 24-70mm ƒ/4 ZA OSS Carl Zeiss Vario-Tessar T* didn't meet our expectations."

I think that is more balanced. Let's not shoot the messenger in the rush to defend what is a below-par lens. Every lens is suitable for some purposes, but that doesn't excuse ordinariness with a Z.

Daniel Walker said...

I had the same experience. I only bought the FE 24 to 70 to give me an option with the 28 & 55. When I say the result I could not believe the results. I thought either the reviewers were wrong or I was better than I thought I was.

Kirk Tuck said...

tnargs, makes you wonder how DXO got it so wrong, huh?

Gregg Mack said...

Dang, I was hoping that you were going to tell us about the new FE 24-70mm GM lens. I have been very curious about that lens combined with the A7R2 camera....

Wolfgang Lonien said...

tnargs, have you heard about field curvature? Maybe you should first read http://www.thewanderinglensman.com/2016/04/is-that-lens-sharp-in-corners-no-are.html - and then think again about all those lens tests...

If you want or need a flat field, take a macro or a tilt-shift(able) lens. Most others are just fine for real world stuff like people, landscapes etc.

J Davidson said...

tnargs - My problem with most of these reviews is that they do not test enough lenses to apply their findings to all lenses of that type made. You can't test 0ne, two or even five lenses and make a recommendation with any confidence. For example, let's say Sony made 200,000 of these lenses (I have no idea how many were made), to have a confidence level of 95% with a margin error of 5%, you would have to test 377 lenses.

Kirk Tuck said...

Wolfgang, Thank you so much for posting the link to Dennis Mook's article. It's a really good read and pretty much exactly sums up my feelings about the modern concept of uninformed lens testing. Leica published some articles many years ago that also discussed the compromise between speed and flat field sharpness. Can't have it both ways? At least not for a reasonable price.

jw52tx said...

I have used this lens for several years with excellent results. As you have pointed out before there is often quite a difference between "real world use" and scientific tests. I have also found the Sony 28mm F2 to be much better than the reviews suggest. If only the 24-70 was a 24-105!

Joepet Macariola said...

I love your blog. I enjoyed reading the content of it.
I was amazed by it.
Keep it up. Love to see more post from you. Thank you for sharing.