Putting e=mount lenses in perspective

Started Jan 16, 2012 | Discussion thread
Ron Parr
Forum ProPosts: 13,701
Like?
Re: Putting e=mount lenses in perspective
In reply to jtyoung, Jan 18, 2012

jtyoung wrote:

Ron Parr wrote:

jtyoung wrote:

Ron Parr wrote:

The 16mm 2.8 has issues that are measurable and quantifiable, and these things have been measured and quantified by sites such as photozone.

I've seen this statement before and I think it's a bit disingenuous.

Really? You think that pointing out objective, quantifiable deficiencies in the lens is disingenuous. You must have a different understanding of that word than I have.

While I would assume that it was clear in my usage of the word from the next paragraph I would also point out that if the primary point of reference you are using is photozone.de, I should point out that the warning that test results cannot be compared across systems is plastered all over the site. So it would be disingenuous to suggest that the 16/2.8 can be quantitatively compared against similar lenses from other systems and found wanting.

The test results use units like pixels, which are clearly tied to the sensor used for test. If you change the sensor, the number of pixels of CA will change, but the optical defects in the lens won't.

If you want to believe that the other cameras/lens combinations they they tested are somehow making the tested lenses on other systems magically better in their overall rating I suppose you can believe that but I think any serious interpretation of the rankings will acknowledge that the tests are not biased in a way that makes lenses look worse on the NEX.

Also, you can restrict your consideration to lenses tested on the NEX. You STILL won't find a worse lens in optical quality. You will find worse NEX values if you start looking at rare lenses designed for other systems that were adapted for the NEX.

Why should we settle for the worst?

I would assume that you are familiar with the problem of bar chart tops. Setting that and my first point about comparing across systems aside, the idea of the 16/2.8 being the "worst" focuses on one metric, optical quality. Now there are those who would argue that it is the only criteria that matters with lenses but I would disagree. And photozone does as well as they have 3 or 4 metrics depending on the lens.

If it were me looking at photozone's metrics, I'd probably favor the price to performance metric myself. If there were a single metric that best encapsulates the "worth" of the lens, that would be it for me. And in that field, if we were to compare the Samsung to the Sony to the Olympus in the approximate range, Sony comes out in the middle (3) not too far behind the leader Olympus (3.5) but definitely ahead of the last in the pack, Samsung (2). So we're not settling for the worst. Just average.

Of the Sony lenses they tested on the NEX, it has the worst P/P.

Comparing across cameras (I thought you claimed we shouldn't do that?) puts the Sony ahead of the Samsung on P/P. I'll give you that. Reviewing the optical issues with the Samsung doesn't make this feel like a particularly proud accomplishment though.

Even at small sizes, on screen, the corner softness in the 16mm 2.8 is noticeable in many shots. You're right that for $250 people shouldn't expect miracles, but people could reasonably expect the only lens that Sony offers with a 24mm equivalent field of view to have deficiencies that are apparent at small viewing sizes even when the lens is stopped down.

Y'know, I gave that a bit of thought and went back and looked at the photos I'd shot with the 16. Out of all the photos I've shot with the 16 (and I've had the lens about a year), there were not a lot that I'd discarded due to the distortion. Lightroom does a reasonable job of eliminating the distortion where I didn't want it and understanding the characteristics of the lens, I'd kept the distortion to maintain the character of the photo in many cases.

And honestly, the distortion/corner softness in the applications I've used it in is not that far off from my Canon EF 24-105/f4L IS which at 5x the cost, I'd have expected better performance.

I wouldn't claim that the 24-105 is a bargain, but I'd point out that it's solving a much harder problem. It's a zoom that covers 24x36mm. APS-C coverage from a fixed focal length lens is an easier problem.

--
Ron Parr
Digital Photography FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/

Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow