« Calling All Geeks! Part II | Main | Ramesh Raskar and Femto-Photography »

Wednesday, 15 August 2012


I'm just curious ... is "idiosyncratic, unfair, and unscientific" a deliberate echo of Austen's "partial, prejudiced, and ignorant"? thanks -

I've done some similar comparison testing, and, like you, found that the image quality of my Olympus 14-42 (version II in my case) stands up quite well to the primes I have available. At 14mm and most apertures it seems to beat the Panasonic 14mm pancake in both sharpness and contrast (especially) in the corners and to be indistinguishable in the center. At 20mm I found the prime to be slightly better at the corners at most common apertures, but couldn't see differences in the center. I've only looked at these results on the computer (at 100%), so I'm not sure how meaningful they would be in my largest prints, which are 15x20.

Good to see Ctein's qualitative observations compare to the quantitative measurements done by Lens Rentals for the same set of lens in these articles from May 2012.


And for variance measurements on populations of full frame zooms (this is more general ... the moral is "it depends on the lens model")


This one will probably apply next week


And I'm pleasantly surprised to see the original version of the Olympus 14-42 kit zoom do so well at the wide end. Time to stop using lusting after a prime as an excuse and get out and use the kit lens I have on my cheap E-PL1.

Another alternative that I've found to be sharper on the OM-D than my Olympus 14-42mm IIR or Panasonic's 14mm prime is Olympus's older 9-18mm ED ultra-wide angle zoom. I bought a used 9/10 copy from LensRentals for $379. This ultra-wide zoom is somewhat larger and requires 4/3 to M43 adapter and 72mm filters but works very well. Back then, Olympus even included a lens hood.

Thanks for the tests.

One thing I'll say in favour of the 20mm; it's the only lens I've ever come across (I've never afforded Leica glass) that's perfectly capable wide open.

Didn't you find that the 45-200 is fine as long as you don't use the very long end? I find that that last bit is mushy whatever you stop down to and let's face it around the 400 equivalent even stabilized means you can't afford to play around with f/11 and the like.

I've always envied you micro 4/3 guys for that 45mm. It also seems I may have another reason fairly soon. But at this point I'm committed to the Sonys an Nikons I already have.


This whole thing is of massive interest to me...I shot a job out in San Francisco last year, where it turned out I just needed to use the Nikon 35mm 1.8 on a Nikon APS-C camera, and I can't tell you how much better of a shooting situation it was with a camera balancing like a camera I would have shot 30 years ago.

It's almost like I don't even care if the sharpness is the same on a prime as a zoom. I'm amazed at the quality of even cheap Nikon zooms now, it's that I want the small lens/bigger body shooting experience. I've realized over the years, I am NOT a "breaking situation" shooter: I don't walk around trying to capture decisive moments, I create moments for my clients through posing, sets, lighting, etc. Hence, no need for zooms. Sad to say, I'm struggling because many of my peers are shooting weddings to stay alive in medium and small markets (where my shooting specialty is no longer needed), and I am just not that "chase and snap" guy! Much to my personal detriment....

I wish that Nikon would get on the prime band wagon, like Canon has done, and make reasonably priced primes in the f2-2.8 range (like Canon's new 28 and 24mm 2.8's). I could easily buy a 16 and 24mm 2.8 that covered APS-C , and with my 35 1.8 and 60mm 2.8 macro, it could cover everything I do, and since I am NOT a "run-and-gunner", I can continue hauling my little show places and work out of a hard case, changing lenses like I do now...

I understand that this column isn't concerned with "relative size, weight, convenience, maximum aperture, or cost." But, still, I have to point out what's so obvious from the photo: the 20mm f1.7 makes my camera pocketable. In fact, without the availability of that lens or the corresponding Olympus pancake, I wouldn't have gotten a Micro 4/3 camera at all.

I wish you could evaluate the Lumix 25mm f/1.4 as well, not that I'd be willing to give mine up for long enough. :-)

I'm also curious how the Lumix 14mm stacks up.

Sadly, for other owners of the lenses you tested, the results are probably not really transferrable to their units. Despite makers quality control testing-which is probably unique to each maker- there is still variability among the individual units. I think in general, the better makers have less range of variance than the "second tier" makers, so your chances of good performance tend to be better with the better makers. Still, are "comparable" lenses from different makers, costing several hundred dollars different noticably or measurably different in performance? A cheaper unit may meet or exceed the more expensive lens in performance-its the luck of the draw. Not to mention camera differences.... Only testing a specific lens/camera combination can tell. And most people don't have the time, interest or technique to do good testing.

I did a similar test with some of my lenses. I pit the 14-42, 45-200 and 14-150 against each other, and got pretty much the same results: as long as you're not pixel peeping, they are all more or less equal, specially if you stop them down by 1 full-stop.

That was when I realized I can use the 14-150 whenever there is enough light. I use primes when I need more light (20mm, 50mm) or when I want a small package (14mm, 20mm), but not because of the better IQ.

Of course less DOF and better bokeh is the specialty of the primes, but at least for me most of my shots don't need it too much.

(I guess it's time I give the 45mm a shot).

When I first tried the Olympus 45/1.8 against a well-regarded Nikkor 50/1.8 I was surprised at how big the difference in quality was at the edges. The 45/1.8 isn't easy to beat, particularly since performance is good at every aperture.

The 20/1.7 on the other hand is a good lens, produces sharp images and all, but its main appeal for me is the small size and f1.7 aperture. I don't doubt that a better lens could be produced if the size and aperture constraints would be relaxed.

I'm not quite sure what you were out to prove or find out.
Stopped down, most lenses are equal in quality, the third party lenses being no exception even though they offer more (?) bang for the buck.

To answer your first question whether or not to chose a prime you have to explain what the "I could use either one of them" means.

If you really do have the time to swap lenses, the job requires the best possible shot at 24mm and only there, then I'll grab a 24mm prime, get great quality stopped down and leave the possibility open to use a faster aperture.

Not really a question is it? I mean, why even bother with the bigger lens?

Now, for professionals the question is whether or not they have the time to do so. Probably not, thats why there is a prime on the second body.

For amateurs? Doesn't matter, does it?

In my opinion, only the fun is important and this can come from a 50mm f1.8 as well as the 24-70 f2.8 that costs 10 times as much.

So for amateurs the answer is always prime because you do have the time to change the lens.
Since fun is more important however, one really can't generalize.

Ctein how did you arrive at your results? By the way your $20 print continues to amaze the people I have shared it with.

Funny, I just did the same test with similar lenses. Olympus 14-42, Olympus 45mm, and Summicrom 40mm. I was most interested to see how my Summi held up to the 45mm. My results are similar to yours, Ctein - at the tele end of things, the zoom was not impressive, the 45mm was excellent from 2.8 thru 5.6 and very good at f/2 and f/8, with the edges only slightly less than the center. The sad news was that the 40mm Summicron, although pretty even across the field, only matched the 45mm at 2.8. At 5.6 it bested the zoom at the edge but not in the center.

I ran the Panasonic 20mm and 14mm lenses through the same tests, and they were as good as the 45mm in the center (at f/4-5.6), but not as good at the edges. My results ended in a range of apertures for each lens that meets my standards: 20mm: f2-11; 45mm: f/2-8; 14mm: f/4-8; 40mm Summi:f/2.8-5.6; 14-42mm: 5.6 only. Sample variations of course could affect any of these results. And my Summicron has suffered my abuse for over 35 years, hopefully mostly soothed by Sherry Krauter's CLA!

Ctein - I think the 25mm Panasonic prime would have been nice to have. Probably similar results to the 45mm I would guess.

Always the individual, always interesting . I love reading Ctein. May I add that I'm okay with the image quality of my Olympus 14-42mm kit lens at it's shorter focal lengths.

Hey Mike, Ctein,

Not to pile more work on you Ctein, but I live not far from you in San Jose, so it'd easy for me to swing up to Daly City and lend you my 9-18mm m.Zuiko if you want to compare it against the 12mm. I also have the Panasonic 14mm and the second version of the Olympus 14-42mm if there's any interest in trying those as well.

Although we can bolt expensive "legacy" primes to the M4/3 cameras I like to try out common manual focus lenses from days gone by. I have several adapters and the old Nikon 50 f2 and 75-150 e both impress with contrast and sharpness. Nothing scientific yet but I plan to do a serious test.

BTW yes my my 14-42II works just great is the one most likely to be attached to my camera for everyday use. The adapted lenses are used for macro, telephoto and DOF control via big apertures.

Dear Jack,

Sigh, no. And me an English major! It definitely wasn't a deliberate echo; I'm not even sure it was an unconscious echo, because I've not read that bit of Jane Austen.

I hang my head in shame.


Dear Bruce,

I pixel-peeped a couple of hundred comparison test photographs. Yes, it was tedious going. I actually did all this photography six months ago; I just couldn't bring myself to sit down and wade through all the results. Damn Jeff and his lens collection and me and my insatiable curiosity.


Dear m3photo,

My 45-200 seems to be entirely decent at the long end. Overall contrast is down a bit, but that's fixable in Photoshop. It definitely isn't mushy. In fact, this photograph in my portfolio, which is a 17" x 22" print, was made with that lens at 200 mm:


Artistically, it depends on a crisp rendition, which speaks well of the lens.

I agree with you about keeping the thing steady. Bloody difficult, even on a tripod. One big reason why, even with my love of telephoto lenses, I don't have any interest in the 300 mm. I am curious to find out if the OMD does a better job of stabilizing that lens than the in-lens or in-body stabilization of the Olympus Pen.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 

Dear rnewman,

Except for the 85 mm Rokinon and the 45-200 zoom Panasonic, I have data from multiple samples of the other four lenses. Mine are typically good performers, neither “cherry” units nor defective.

I have more complicated and subtle reasons for thinking that the 85 and the 45-200 are typical samples.

In any case, it's unlikely that people own exactly the combinations of lenses I do, so I agree that the comparisons aren't transferable. You will note that I did not present a standalone objective analysis of each lens. That was by intent. I didn't subject them to that kind of scrutiny. Or, where I have, as in the case of the 12 mm and 45 mm, I've written entire standalone reviews of those.

I agree that most people will not take the trouble to do this kind of testing. But for those who are inclined to, these two columns are encouragement that it will probably not be wasted effort. I'm definitely going to be changing some of my lens carrying habits based on the results I got.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 

Hopefully you will include the new 12-35mm f2.8 X lens as well. It is spectacular. http://frugalfilmmakers.com/2012/08/10/no-need-to-resist-the-panasonic-lumix-x-vario-12-35mm-f2-8-asph/

Dear michael w,

“ I'm not quite sure what you were out to prove or find out. ”

Really, I don't think I could've been any clearer in my second paragraph. Try rereading it again. Maybe twice.

"... you have to explain what the "I could use either one of them" means."

And I did. Again, reread.

"Stopped down, most lenses are equal in quality,..."

Wrong, simply wrong. Seriously. In many cases the differences between the lenses are subtle; and others they are not, as where I wrote this:

"The 14–42mm is mushy and smeary in the corners; even stopping it all away down to ƒ/11, where you'd expect diffraction to completely dominate, doesn't clear that up entirely."

Tell me, did you actually read the article or did you just see the title and write a rebuttal?

There are many, many kinds of optical defects that are not improved by stopping the lens down. Even for the ones that are, there can still be substantial differences.

As for your style of photography and method for choosing the lenses you do, it wouldn't work at all for me. Whether it works for someone else is up to them.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 

Another internet myth busted! Well, probably not because they are seldom influenced by reason or evidence. However I concur. A lot of zooms fall within the "good enough" zone for most general photography.

But thanks for actually saying so.

Dear Craig,

Ah, good point. Thanks for bringing this up!

I am comparing corrected RAW's, because that's what is relevant to the kind of photography I do.

Someone running such tests for themselves should definitely make the photographs using the methods they most commonly use for their regular photography.

pax / Ctein

I'm just not empirically inclined where my lens selection is concerned (and I'm ok with that, and I'll bet Ctein is ok with it (as it pertains to me), too). I've got an E-PL1 with the two lens kit, and supplemented it with the Panasonic 20. The only aspect of 'performance' that dictates for me to use the 20 over the wide zoom is frankly, the zoom is a pig at AF locking indoors. I don't generally shoot wide open, but f1.7 sure helps the camera acquire focus. I do like the FOV of the 20, and generally don't use the wide zoom at all. For situations where it would be used, the 20 is easy enough for me to move into a different position. So my 'big kit' is the 20 and 40-150 kit lens.


But you missed out a bokeh analysis. Or do we just assume zooms underperform on that aspect?

Zooms vs primes was but one reason I made an unusual switch a couple weeks ago. I gave up an entire Fuji X-Pro 1 and 3 lens outfit (with excellent image quality) for an Olympus OM-D and a superlative 12-60 f/2.8-4 lens. Why? I was dubious about the forthcoming zooms for the X-Pro 1 and, living as I do in a wet climate area, decided a weather sealed system was a good idea. BUT, the overriding reason was my desire for one-lens convenience over constantly changing in the field in harsh conditions. I had to regain some carrying weight to do it, but I think I have very close to equal image quality over a 24 to 120mm equivalent range where virtually all my photography is focused on.

How about zoom vs. zoom? I've been using the Olympus 14-35 f/2 for about two weeks, comparing it to the Olympus 14-54mm (the kit lens for the old E-1) at comparable focal lengths on my E-3. For edge sharpness the 14-35 wins hands down at all comparable focal lengths. Center sharpness - the two are very comparable at 35mm but the 14-35 wins at 14mm. And when I say wins I mean "OMG that's sharp!"

BTW, I've also compared the Olympus 50mm f/2 to the Olympus 14-54mm at 50mm and the center sharpness is just about equal; edge sharpness goes to the 50mm.

Finally, I've used the Olympus 300mm f/2.8 about three times photographing sandhill cranes, using on a sturdy tripod, and can only get sharp images at 1/1600 sec. shutter speeds. And that's with the anti-shock set on 5 sec. delay. On a windy day - forget it!

Did I read somewhere that the m43 cameras are applying corrections to raw files for same-maker lenses?

I don't know what body ctein was using, but if the above is true then certain lenses might have been benefiting from in-camera corrections?

Taking dozens of pictures with different lenses and comparing them in detail is a long, interesting, and probably very tedious exercise. I am constantly amazed that people will spend their time and energy to do these things, with the sole purpose of sharing the knowledge with others. Thank you very much.

I did the same type of comparison between the Oly 14-54 Mk II (w adapter) m14-42 IIR, and Panny x 14-42 at 14mm, 25mm and 42mm on the E-M5. Handheld in good light at my usual shooting aperture of 5.6, virtually identical. I took the 14-54 and the x14-42 on a walkaround shooting a variety do shots with each, and again, virtually identical, though CA behavior would let you ID which lens shot which. The gear is just really good these days.

Dear Robert,

Oh, I am far less altruistic than you give me credit for. I did this because **I** wanted to know, for my own photography.

pax / self-centered Ctein

Dear T N Args,

A good question!

You are correct that with some lenses today, lens designers are doing some correction for certain aberrations (like residual lateral chromatic aberration or distortion) in software rather than in hardware. There are purists who think this is a bad thing; so far as I am concerned, any tool for correcting aberrations that works well is a good one (sometimes it doesn't). Understand that designing a lens is always a trade-off, so leaving some aberration correction to software allows designers to correct other aberrations better in glass.

That said, I'm not looking at JPEGs, I'm looking at corrected RAW images processed by Adobe software. So the camera body doesn't come into play in correcting this stuff. ACR recognizes software lens corrections for all the lenses I own that need them, so far as I can tell.

I don't know if the relative lens rankings would change if I were looking at JPEG's. I don't care, since I don't make JPEG's.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 

Morning Ctein,

Not at all suprised to hear that the Rokinon 85 is doing well it's build by the Korean lens manufacturer SamYang. I own a SamYang FE 7.5 and that is a great performer as well, the Micro 4/3 FE du jour, I would say outperforming the Pana by far, especially in sharpness, fringing and chromatic aboration. I use it on a 12 Mp GF1 for shooting VR 360/180 sphere's in HDRi and that gives you a sharpness that is scary at times. You should try it out Ctein (for about 300 dollars you get a really nice FE).

Greets, Ed.

Here in Pentax World, where I live, high quality prime lenses are the standard, almost rising to the level of a fetish for many users. (If you think I'm exaggerating, check out some of the Pentax discussion forums.). The FA Limited series (full frame) are especially excellent. Nevertheless, some of my best photos have come from zoom lenses that are undeniably inferior to the primes, optically speaking. I take this to mean that most modern lenses can deliver IQ that is good enough for all but the most demanding purposes. Now which set of lenses are smaller and more fun to use? The primes, of course.

Dear Don,

You should assume nothing of the kind.

pax / Ctein

Dear friv,

Again, one should not assume. Consider that the 20mm f/1.7 Panasonic Lumix lens is also a much-lauded lens, but it does not perform better than Olympus' kit zoom at the same apertures. Yes, it gets you good performance two stops faster, but that was not the question under investigation.

If, by the 25mm Panasonic prime, you're referring to the Leica DG Summilux 25mm f/1.4, it happens I did do a series of test photos against the Lumix 20mm f/1.7 when I bought my OMD. I could not see any significantly better image quality from the Leica lens.

My test was bit cruder than my usual standards, so I might very well have missed some small improvement, but if I did it was not large enough to justify replacing my 20mm.

pax / Ctein

This review is very true. I wish we had more like this on the net.

The comments to this entry are closed.