« OT: There's Ugly, There's Butt-Ugly, Then There's... | Main | Happy Thanksgiving »

Wednesday, 23 November 2011


Feed You can follow this conversation by subscribing to the comment feed for this post.

Have you tried the Olympus wide-angle converter for the 14-42 kit lens? As I recall, it gets you to 11mm equivalent. My informal testing indicates that the performance is as good as the kit lens alone. Certainly works for me, although you don't get the f/2.0 speed.

Well, I still love this lens, which just goes to show what I already knew, namely that I am a still a rubbish photographer after 60 years of photography. Guess I should stop aspiring to improve after all this time, and settle for an easier life.

Pleiades? Or Ursa Minor?

I had noticed some smearing in the corners in early sample images.
I have a 4/3 9-18 for comparison., which is sharp from edge to edge.

However the m4/3 9.18 has the same problems, so my impression is that digital correction has problems with superwides and does some 'pixel stressing' at the edges, when correcting.

In other words it might be a format/distance to flange problem with wides in mirrorless, where APS is even worse (see NEX 16 mm).

Thanks for the review - you've probably saved a good many people a nice chunk of change (me included)!

Chalk it up as another puzzling move by Olympus though - they clearly know how to do good lenses (45/1.8) and even good WA lenses (ZD 12-60). So why are they trying to get away with mediocrity in an $800 product?

Thank you thank you thank you. Our family budget thanks you. :) Now I can reduce my lens lust to just the 45/1.8. I have the Olympus 14-24 and I agree that it produces excellent results for a kit lens. Between it and the 20/1.7 I find it hard to justify additional lens purchases.

Handheld at 1/2 sec, wow.

I was almost wooed by this lens, but bought a NEX 3 with 24mm Equiv f2.8 for half the price (yes, the lens came with the camera for 399). Given what I am seeing here, that may provide users of mirrorless systems a viable alternative, albeit non stabilized.

Another Support-Oly-no-matter-what piece?

How could an essentially critical review be a "Support-Oly-no-matter-what piece"? Did you even read the review?


I suspect that the achilles heel of m4/3 fast wides is the short flange distance combined with the need for a certain amount of telecentricty of the optical path. I am guessing that this results in the significant degree of barrelling the m4/3 wide primes display uncorrected. I further guess that the From what I've seen in other reviews, all thin-camera or in-raw-processor correction of this barreling results in the relatively poor edge performance that is the hallmark of all the m4/3 wides (i.e., the Panny 12mm and Oly 17mm).

Eh, Ctein,

I use the 9-18 and indeed some smearing occurs due to the fact that it has some horrible CA wide open and even stopped down a fair bit. Now I use a Pana.....and correct that in Silky Pics....maybe you could give that a try. Without CA correction the 9-18 stinks big time....but mannually correcting CA improves matter a lot, as to make it one of my favorite lenses.

Greetings, Ed

Dear Tom,

I am always dubious about converters, but if you want to lend me yours for a couple of weeks, I'll give it a test.


Dear Mike,

I never understand how readers can possibly interpret a less-than-favorable review of a product as an attack upon the product-owner.

Not my problem, guy. Really.


Dear Giles,

Hmmm, so if I understand you, the smearing might not be a problem with the original photo but with how the RAW is getting processed? OK, worth checking out. Can anyone recommend to me a way to look at the RAW files, unmassaged? On a MAC, thank you, and for free (for this purpose a trail version of a program or crippleware download will do just fine)?

I'd note, by the way, that if this is an unavoidable side-effect of software correcting some other lens aberration like geometric distortion, still amounts to the same thing-- part and parcel of what the lens can deliver.


Dear Scottag,

Have you ever heard the term "drive-by comment?" Just wondrin'....

pax / Ctein
-- Ctein's Online Gallery http://ctein.com
-- Digital Restorations http://photo-repair.com

"Pleiades? Or Ursa Minor?"

Pleiades. No doubt in my mind plus Ctein has a degree in physics and does astrophotography. Not a mistake he'd make.

I've done a lot of debating with myself recently and have almost sold myself on moving from a compact and aps-c DSLR to a Mft. The Olympus 12,17,45mm lenses would make the perfect prime lens set for me. This is the lens that is holding me back, I've read similar reviews online and now that I've read this here on a trusted site, I'll be sticking to my wonderful Tokina 116. The $800 plus another $1000 for the setup is just to much to make the jump.

Maybe if you had reviewed the lenses the other way around, first the 12 and then the 45, your review of the 12 wouldn't have had these conclussions. I was very happy and proud of my DZ 11-22, 14-54 and 50-200 until I got de 50 Macro and the14-35 f2, these two are so excellent and sharp that I'm not using the others anymore, and planning to get a couple of additional SHG from Olympus.


M. Guarini

Ctein's observation of this lens vs. the 45 f/1.8 isn't surprising, if you think about lens technology. Companies have been making lenses in the 40-50 mm range at f/1.8 and faster with excellent quality and reasonable prices for many years now, while it is only recently that they've even attempted a 12 mm, much less f/2. It's a much harder lens to make, and the fact that it even exists is pretty remarkable.

I have it on the e-p3 together with the 25 1.4 and the 45 1.8 and the 12mm stopped down to 4 makes crisp pictures!!


The price all but had me scared off of this lens already, and this review puts the final nail in the coffin. I have the (m4/3) 9-18, I love it, and I very rarely find myself wishing for extra light when using it. (The IS works great for that lens as well, turning out sharp photos at a quarter-second with great regularity, which is part of the reason I don't need more aperture so very much.) So, $800 will remain in my account, waiting for an Oly m4/3 body with an upgraded sensor.

I have the 12mm 2.0 25mm 1.4 45mm 1.8 together with the E-p3. Have had it since it came out have been shooting it alot. More than my 5DII lately. I think the 12mm is well worth the $$. And no i'm not praising Oly because service wise they are not so good to put it sofly.

Some samples of the 3 lenses above most have all exif. and are full size.


Dear Rob,

Pleiades; I was facing due east. But it does look kinda Ursa-ish, dunnit?


Dear Alan,

I'm hoping to get pointed at software that will let me figure out where the problem is.

Understand, I have no problem with software correction of aberrations. I don't care if the silicon doing the correction is bound up in silicon dioxide in a lens barrel or buried in my computer. I'm totally fine with it. All that matters is what comes out the end. If I see an unacceptable amount of image degradation, doesn't matter exactly how it is arrived at. Similarly, if the image looks great, I don't care what combo of hardware and software was involved.


Dear Ed,

It's not a lateral chromatic aberration problem. I routinely correct for LCA (if the RAW converter hasn't already done it for me). This is smearing that's visible in the individual color channels, it's not a simple misregistration.


Dear Marcelo,

In point of fact, I got the 12mm lens before the 45mm. I wrote the reviews in the other order because the 12mm review took a lot more work.

It made me kind of nervous, truth be told, about the 45mm I was getting. I had to cancel my order with B&H because the lens was on too-long backorder, and I bought it off of eBay. After getting a bad 12mm, I was feeling kind of gun-shy (never bought a bum lens before, really, truly). The seller was very good about holding my hand, sending me a couple of sample image to reassure me before I bid. If you ever find yourself doing business with "alans49" you'll get treated well. He's an A+ in my book.

pax / Ctein
-- Ctein's Online Gallery http://ctein.com
-- Digital Restorations http://photo-repair.com

There's always the Noktor/SLR Magic F1.6. A bargain if you can forget autofocus.


By your reply I take that you have been shooting Jpeg only with the 12mm?

It doesn't make such difference though. If you check the LensTip review you'll see that before (in camera) correction the lens has a huge amount of barrel distortion.
That is probably unavoidable if one wants to build a lens that is both small and super wide.

Now I know from experience that when one corrects a Fisheye in PS to make it linear, one has to throw away part of the image because of 'stressed pixels' at the edges. This smearing happens too to the m.4/3 9-18, so I take that it is unavoidable in m4/3.

This lens however is fast. When introducing it Olympus presented it as a street shooter, not as a landscape lens. So you might reconsider the intention. In mirrorless, with short distance to flange, one can't have everything.
Or else, as said above, one should use older 4/3 telecentric lenses for landscape, as I do with the old 9-18: no smearing at the edges. But the lens with adapter is way bigger.
You won't have this problem with teles, only with lenses whose focal is shorter than the m4/3 register, some 20mm.

i disagree with your conclusions , this is a vastly superior lens , but QC copy to copy is a bad as canon L glass....sadly ,it seems

i pixel peeped my copy when if first came out
and center and corners were great,

also the mf is in 13 stepped zones , which snap at intervals it is the best one will be in focus exactly the oof in stepped intervals in front or back

it is the best implementation of focus by wire in any modern af lens , but
thats not sayin much

my copy was made in japan ... were your 2 made in china IM NOT TRYING TO BE FUNNY HERE

Wow - thanks for the review Ctein. I was playing around with the idea of getting an EP-3 and the 12mm lens before Olympus vanishes in a puff of smoke. Now I'm having second thoughts, considering the price of the lens, but there aren't many other options for 24mm equivalent and fast aperture for a large sensor pocket camera.

Perhaps this is a reflection of the difficulty involved in getting decent image quality from short focal lengths on relatively large sensors? I'm guessing the 45mm 1.8 is technically not that difficult to make razor sharp - the designers can ride the back of 60+ years of 35mm standard lens development - but 12mm f/2 with flat field, minimal edge aberrations, and pocket size sounds very challenging!

"This lens is so stable that I would swear the image stabilization system in the Olympus EP-1 had been designed around it."

I don't quite get this. Isn't the IS in-body, and aren't lenses supposed to be stable? Specially prime lenses.

That snap zone manual focusing is quite a turn-off.

Ctein, thanks for this review.
I have the 12 and my copy is doing fine. Not great in the corners, but then I don't do landscapes nor architecture so I suppose it's enough. What really strikes me with this lens is the micro-contrast, which gives images made with it a natural, realistic look even at f/2.
The geared focus ring is a great idea, but I noticed its inaccuracy too. I've learned to compensate and get zone focusing right in most cases. It is an annoying flaw, though.

My experience with the lens is closer to Jeffrey's. (It was with me for just about a week, though, together with E-P3.) There is evidence of softness in the corners but nothing that would really bother me. (OTOH, I noticed slight CA in only one photo. If I wasn't looking at 100% because of your photos, I doubt I would have noticed it.)

Besides, this MTF chart shows that is an expected behaviour. In comparison, the SHG 7-14mm lens is better but it's not F2.0 and is much bigger and heavier.

But then, I may be an undemanding person.

Lighten up guys, surely Scottag just had a sarcasm attack. Didn't you Scottag? Scottag? Are you there?

Good review, straight and to the point. I hate having to read between the lines trying to decipher the true meaning of phrases like "well adapted to portraits due to the dreamy way it renders details at the edge of the frame". It's much better when you just say what you mean.

Gilles wright wrote here : http://theonlinephotographer.typepad.com/the_online_photographer/2011/11/olympus-12mm-review.html?cid=6a00df351e888f883401539373c40a970b#comment-6a00df351e888f883401539373c40a970b

That it may well be a limit with the format, exactly what I thought. Also the lens may be too heavily corrected through software, which can add to the problem.

BTW, Ctein, have you tried looking at the files in Olympus Viewer 2? (Wouldn't recommend it for regular work, though.)

I agree with Tim I have the same set up and have found the lens to be outstanding, I also own the K5 and various pro lenses and it has been left behind in proeference to the Olympus.

I would suggest folk try it in real conditions and shoot what you normally do to gauge whether it fits your expectation, whilst this review has said it is poor I have read a number completely contradicting this. It is the same for most things one mans honey is anothers poison.

Looking at various pixel peeping web sites e.g. SLRGear as well as some 'seat of the pants' reviews I had come to the conclusion that the Panny 14mm f/2.5 was a worthy addition in place of this much vaunted and expensive alternative. As a bonus the panny is a pancake.

The trio of 14mm f/2.5, 20mm f/1.7 and 45mm f/1.8 takes some beating in terms of bang-for-buck and sheer jacket pocket, no camera bag, portability.

Can anyone recommend to me a way to look at the RAW files, unmassaged? On a MAC, thank you, and for free...

Raw Photo Processor (RPP).

First, if we want to have Olympus stuff, we have to pay for some guys over the Caiman islands, therefore the lens price being so high. The japanese mafia won't let you have a decent lens hood for free not even a miserable 1$ pouch, so forget about the price.
I have the 12mm, the pany 20mm and the 45mm f1.8 and I have to make some justice to the 12mm. While I agree that the IQ of the 12mm is exceeded by the 20mm and the 45mm, the 12mm is still good, plus it's a wide angle lens! Normal and short tele lens always performed better unless we are talking about highly priced lens such as the Leica branded. Compared to them the 12mm is really cheap.
Second, I don't know if you tested the lens with a camera hood, but with this one you really need it or otherwise definition and contrast are heavy affected, also, the lens is useless closed up over f8. Even f8 produces just acceptable pictures. This is the main reason why the dof scale is not that useful, you can´t work it's hiperfocal capabilities.
Also, It maybe be just me but I see small details with this lens like fine spider web strings, that I don't remember to see on the scene, neither in the final results of pictures taken from, for example, the panasonic 20mm.
Finally, I have seen that you don´t care much about color rendition and maybe that's why you don't like this lens, because from all the m4/3's I have used, this one produces the best colors by far.
I guess my English is not that good so please forgive me.

simply put....my 12mm from Olympus is absolutely superb. I own every lens they have made as well as Panasonic and the 12mm is just wonderful.

It's not a matter of how many degrees one has or what their field is...I have been in this industry for over 50 years and my experience with many systems most of which I have owned tells me that olympus did a great job with the 12mm as well as the 45mm.

Ctein wrote:

"I am always dubious about converters, but if you want to lend me yours for a couple of weeks, I'll give it a test."

Ctein - not practical for me to ship it to wherever you live, as I'm about to leave on a long trip. It's only $100 ($80 at B&H right now). Perhaps you could find a dealer who would loan you one. These folks tested it more than I did, with fairly positive results:
My needs are less critical than yours, but for 13x19 max prints it does the job for me. I can't justify spending more for the few times I need wider than 28mm equivalent. I do share your suspicions about converters.

Thank You Ctein. :-)
Your artical was Illuminating as to why the lens is flawed for you as a photographer and equally illuminating as to why the lens has merit and usefulness in available light situations.
Richard in Michigan

Dear Tim,

I couldn't learn anything from the photographs on your Flickr site because I couldn't view them with anything more than one quarter resolution. At 25% scale, none of the problems I'm talking about are particularly visible, so I can't tell if your sample of the lens behaves better than mine. Also, lacking technical information for your photographs (in particular, aperture), I can't make any meaningful comparison.

Do you have another site where you have images posted at full resolution? I'd be happy to take a look at some. I am not above considering the possibility that I've looked at three substandard samples of this lens (pace Christina). While I doubt it (See my comment to Giles, following), it's not impossible. Not that that would make me feel especially kindly disposed towards this lens, but it would explain the “nope, it's wonderful” versus the “yeah, what you said” comments.


Dear Giles,

I always make my photographs in RAW. Not sure how you got the impression that I use JPEG.

I grabbed RAW Photo Processor and confirmed the substantial barrel distortion. I compared the barrel-distorted and ACR-corrected conversions in Photoshop, and it does indeed look like the smearing that I'm seeing in the perimeter is a consequence of correcting for the barrel distortion. So, in my book, this qualifies as a lens design problem: a poorly-corrected aberration that compromises final image quality. Considering the amount of distortion that is being corrected for, I'm impressed that ACR does as good a job as it does

Personally, I'd rather have the lousy manual focus ring eliminated and a couple of more elements put in the lens to reduce the barrel distortion to an acceptable level. This level of barrel distortion is not fundamentally inherent in a lens. A better optical design can reduce it.

As Kelvin also noted, making a lens of this focal length and aperture excellent is a challenge. Well, that's why I'm paying $800. Doesn't matter to me if there are other lenses out there that are even more expensive. This is still an expensive lens, in my book, and I expect a lot from it accordingly.


Dear cosinaphile,

The boxes say the lenses were made in Japan. I don't know another way to check this.


Dear toto,

That sentence could have been written better. Try this on for size: “Stabilization is so good when using this lens on the Olympus EP1 that I would swear that the image stabilization system in the camera had been designed with this particular lens in mind.”

In-body image stabilization exhibits very different degrees of effectiveness with different lenses. Some lenses stabilize really well, some nowhere as well. For example, relative to the 1/FL sec rule for exposure, image stabilization only gains me about a stop with the 45 mm lens. With the 12 mm lens, it's gaining me almost 3 stops.

Hope this clears the matter up.


Dear Ricardo,

Your English is entirely understandable. I think there may be a little confusion over my use of the word “soft”. When I talked about image detail being a little soft, I did not mean it was low in contrast (it can mean that). I meant that the detail wasn't really crisp. That is, the acutance was not very high. A lens hood won't change that; it would improve contrast by reducing flare and other scattered light, but that wasn't the problem I was seeing.


Dear Elliott,

It's not only not a matter of how many degrees one has or what their field is, it's also not a matter of how many years one claims in the business. I've got 45. So what? This is not about seniority.

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

It seems that most small wide-angle lenses in the m4/3 and NEX systems suffer optically partly due to the emphasis placed on miniaturization. Having a fast max aperture compounds the problem.

You can't have small, wide, fast, and flawless optics all at the same time. Pick any three. Layer on price constraints, and it becomes even harder.

I'm sorry no i don't but you are able to see full resolution now and you can see the exif data if you click on the camera it's been taken with.

And please keep in mind it's a 12mm 2.0 lens everyone!! Name a better one thats sharper throughout the picture?!?

I don't say your sample isn't working right i'm just saying something must be wrong somewhere because i think and many many others think this is the best wide fast m4/3 lens no doubt.

But no worries i'll still read your blog :)

Dear Tim,

First, an apology and correction-- when I went back to your Flickr site, this time I could bring up the original-size images. I don't know why the browser wasn't showing me that option before. But clearly my problem, not yours.

Second, if you want to show someone an example photo, point them at the specific photo, OK? There were hundreds of photos on your site. How am I to easily know which ones were made with the 12mm lens and at what apertures? I've got lots better things to do with my time than find appropriate samples by trial and error.

That said...

Third, I did find two or three appropriate comparison images. Frankly, the corner image quality sucks. Lots worse chromatic aberration, lots worse smearing than my lens. Worse, even than figure 3, right, which is from the defective lens I returned.

I don't know if that's due to you having a poor sample or if it's in how your JPEGs got processed for posting. All I can tell you is that those photos aren't anywhere close to "crisp."

If they're making you happy, well and good. YOU are the one who needs to be happy with YOUR lens, not ME. You already own the lens, so my review should be entirely irrelevant to you.

But I will tell you that the photos you pointed me to that are supposed to praise this lens only serve to confirm my modest damnation.

pax / Ctein

Dear e_dawg,

That's a good question, where the root of the design-concept problem lies. It could be in the miniaturization, but since the primary aberration turns out to be geometric distortion, it may have to do more with price points and creeping-featuritis. Geometric distortion's definitely not an inherent problem with the constraints of this system.

And, to some extent, it's a market position problem. If this had been a $400 lens, no one would be demanding as much from it. But for the marketing people that then becomes a call they have to make on how much the customer is going to demand at a certain price point. For my purposes, Olympus made the wrong design decisions for a lens at this price point. But it's entirely understandable how those decisions got made.

What I can't figure out is why the full-manual focus system has so many warts. I would think all three of the problems I described could be very easily fixed. Either that means there is something about designing those kind of focusing rings that I don't understand (entirely possible) or a designer at Olympus dropped the ball on that (also possible). We will probably never know.

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

I am concerned with reports like these. Are we looking at pixels or are we see the picture?

The comments to this entry are closed.



Blog powered by Typepad
Member since 06/2007