21mm Elmarit-M ASPH Poor wide open

[Tim B]

Advertisement (gone after registration)

I wonder if anyone would like to comment on my recent experience?

 

I have just bought a secondhand 21mm Elmarit-M ASPH, which appears in mint condition, boxed, etc. It appears scarcely to have been used. It must be a few years old - the box is white.

 

Wide open it produces a very poor quality image - it is very soft at the edges, rather like an old-fashioned soft-focus lens. One stop down it is perfect, indeed brilliant and remains so all the way to f/11 if not f/16. Is this the characteristic of this lens, or have I got a dud? I doubt that it is really usable at f/2.8 and I didn't buy an f/4 lens.

 

I will try to post some examples later, if possible, but at the moment time is short.

 

Thanks for any comments.

 

Tim

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[marknorton]

Certainly not my experience of this fine lens. Can you post an example?

[Eoin]

I have this lens also and while I tend not to use it wide open but on the occasions I do the results are sharp in the DOF range across the whole image.

[xrogers]

At high magnification the corners are visibly not as sharp as the center wide open, but definitely not "soft focus". For typical photography (certainly 8x10, perhaps even 11x14 prints), this lens wide open appears wickedly sharp across the entire frame. Stop it down and you can go even larger.

 

--clyde

[Tim B]

Many thanks for your comments - it seems that I may have a problem.

 

Here are some rather hastily prepared examples.

 

The lens is not coded, there is no IR filter and the camers is set for lens recognition OFF.

 

First is the full image at f/4 with the crop area for the next two examples marked in the top right corner

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Next the f/2.8 100% crop

 

 

Next the same area from the f/4 image.

 

 

I hope that the quality issue I'm conerned about shows up on the reduced resolution image.

 

Thanks for any further comments.

 

Tim

[xrogers]

The example doesn't look obviously bad to me. It certainly shows reduced definition at the edges (as it should wide open), but not poor definition---for instance, I can still easily count the leaves on the tiny branches poking out of the larger leafy masses.

 

I expect a properly printed 8x10 would look very sharp---perhaps print it up and see?

 

--clyde

[Guest guy_mancuso]

Advertisement (gone after registration)

Try a wall or something , really hard to use a tree as evidence , they move. Seriously use a building or something that has no chance of any doubt. I know it was a calm day and not trying to give you a hard time but i have tested many many lenses and anything that can move I don't even look at. You may very well have a issue but why doubt a test right out of the shoot, also make sure your lens is focusing okay also, i would test that first before testing a lens at 2.8.

[robertwright]

I have the non-aspherical 21, and it does show reduced performance at 2.8, it really crisps up at f4. I might wager that the asph version is better, but it may not be all that different. what you have looks a little like my 21.

[Tim B]

Thanks, guys.

 

I know it wasn't a very good test, Guy, although it is the difference between the two shots that is quite marked, isn't it? They were taken as close together as it takes to turn the aperture ring from 2.8 to 4, so leaf movement would show up on both shots - even more on smaller apertures as the shutter speed slows down, although it was a very still morning. I noticed the problem first on a shot I took in the dealer's shop, but couldn't examine the image until I got home.

 

But it is a good point and as soon as I get the chance I'll try a brick wall - my usual target for lens tests.

 

What struck me about the problem was that it didn't look like normal edge of frame decline in sharpness, more like there was a "halo" around object edges, especially high contrast transitions. I've actually not seen anything quite like it on a Leica lens before, but then I've never had a 21 mm before.

 

Bear with me, guys, while I make some further tests later on today and thanks for your help.

 

Tim

[adan]

Let me quote from E. Puts' Leica Lens Compendium RE the 21 ASPH.

 

"At full aperture...very fine detail is rendered with crisp outlines over a larger part of the center (until image height 9mm). In the outer zones performance gently drops and becomes very soft in the extreme corners....Stopped down to f/4 the field improves visibly...at 5.6 extremely fine detail is recorded over the whole picture area."

 

"Image height" means radial distance from the center of the frame, so the sharp area at f/2.8 should be a circle that just fits the narrower dimension of the M8 sensor. Anything outside that circle will begin to suffer. At f/4-5.6 the "field" ( the rest of the image) begins to catch up.

 

In short, his description of a "perfectly functioning" review sample lens from Leica matches your description almost exactly.

 

The 24 ASPH does better - sharp at f/2.8 over a 12mm circle - but obviously has the advantage/disadvantage of not having to capture as large a chunk of scenery.

 

BTW: the haloing or fuzzy-over-sharp quality that I see in your samples is likely a touch of coma. You can check that by shooting point light sources (decorative fairy lights or Xmas lights or the like). If they start to show a slightly triangular, butterfly, or comet shape towards the corners, that's coma (and not an unsual aberration in wide-angles).

[Tim B]

In case any of you come back to this thread .....

 

Today was rather wet in London, but I went out with the lens and took a few more test shots of (stationary) buildings, although in less than ideal conditions and I also took some photos at home with flash of a full bookcase. Results confirm my earlier feelings that this lens is indeed very soft at the edges and corners at f/2.8. However, they are usable in most circumstances and stopping down to f/4.0 removes all doubts.

 

It is considerably less sharp wide open that my new 28mm Elmarit-M ASPH, but compares well from f/4.0 on.

 

I went to two of my favourite London dealers to ask their opinion and advice and showed them my test shots. They were non-committal, but suggested I speak to Leica UK about it. One of them even phoned Nobby Clark for me there and then, but he was out. They were very helpful and encouraging. They seemed to think that my results were probably not too far from normal.

 

There was a suggestion that the fact that the lens was uncoded may have had an influence on the results, but apart from some obvious vignetting, I don't think that this can be a factor - or can it?

 

Anyway, thank you for taking the trouble to make helpful comments. When I take the lens to Leica UK for coding I'll ask them to check it over, just to put my mind at rest.

 

You know how it is with gear, if you don't have confidence in it there is always a niggling feeling that it won't give you the results you expect.

 

Tim

[luigi bertolotti]

I am interested for I am waiting for a 21 Asph (chrome) to come back from Solms for coding: I found it at my dealer, used but as new, box, warranty etc : I was there to take on my M8 (had to wait 2 months for a chrome one...) and decided to acquire the 21 without even a single trial shot, just looking at aspect and glass condition: I thought that my beloved Super Angulon 3,4 would result not so fit for M8 (compatible, but not metering, impossible to code) and I put my trust in the 21 Asph... Have I to be concerned by these results ? Of course, I will be able to answer only when I'll shot one after another my ASPH and my old SA (that I always considered an excellent lens on my M4). My impressions based on the posted pics are these:

 

1) I also red Put's tests: agree that on 24x36, for a 21 wide open a certain degree of softness at the borders is a must. I have also the first SM Super Angulon (f4) : it's delicious in aesthetics, and almost unusable at f4 : corners are supersoft, better to say "confused", and 2 stops darker than center.

2) but there is something special in the published pics: colors are strong green (foliage) against blue (sky) : I think that this is a specific combination in which the famed cyan drift takes on (green "contains" blue, sided on the spectrum): effectively, it seems to me the problem is mainly in the transition of colors (border of green leaves to blue sky). I hope that the coding can solve a good amount of this issue: after all, this is one of the reasons for the coding has been engineered, expecially for WAs.

3) So, I am very curious to see other examples You'll post regarding very different situations (buidlings etc...) with different colors: I would not be surprised (and hope so... my Asph shall arrive, some day...) if the problem is quite different or even absent, or maybe other kind of problem could appear (loss of light at corners, for instance... I would consider a certain degree of it as normal with a 21 wide open)

[woodyspedden]

I am interested for I am waiting for a 21 Asph (chrome) to come back from Solms for coding: I found it at my dealer, used but as new, box, warranty etc : I was there to take on my M8 (had to wait 2 months for a chrome one...) and decided to acquire the 21 without even a single trial shot, just looking at aspect and glass condition: I thought that my beloved Super Angulon 3,4 would result not so fit for M8 (compatible, but not metering, impossible to code) and I put my trust in the 21 Asph... Have I to be concerned by these results ? Of course, I will be able to answer only when I'll shot one after another my ASPH and my old SA (that I always considered an excellent lens on my M4). My impressions based on the posted pics are these:

 

1) I also red Put's tests: agree that on 24x36, for a 21 wide open a certain degree of softness at the borders is a must. I have also the first SM Super Angulon (f4) : it's delicious in aesthetics, and almost unusable at f4 : corners are supersoft, better to say "confused", and 2 stops darker than center.

2) but there is something special in the published pics: colors are strong green (foliage) against blue (sky) : I think that this is a specific combination in which the famed cyan drift takes on (green "contains" blue, sided on the spectrum): effectively, it seems to me the problem is mainly in the transition of colors (border of green leaves to blue sky). I hope that the coding can solve a good amount of this issue: after all, this is one of the reasons for the coding has been engineered, expecially for WAs.

3) So, I am very curious to see other examples You'll post regarding very different situations (buidlings etc...) with different colors: I would not be surprised (and hope so... my Asph shall arrive, some day...) if the problem is quite different or even absent, or maybe other kind of problem could appear (loss of light at corners, for instance... I would consider a certain degree of it as normal with a 21 wide open)

 

I have the 21 asph and find it very sharp in the center and gets a little soft at the corners but nothing untoward.

 

You may want to also read Sean Reid's review of the super wides where he uses a 21 asph as a sort of reference by which to evaluate the other lenses in the test, particularly comparing the 21 asph to the 21mm focal length of the WATE. He shows that the 21 asph does get a little soft in the corners as does the Zeiss 15 and to lesser degree the magnificent little CV 15 4.5.

 

I think you may want to send your lens to Leica repair for a full evaluation and CLA. Just my humble opinion

 

Woody Spedden

[martinb]

That looks strange! I think even my 21mm Elmarit E60 would look sharper..

[Tim B]

Thanks again for your comments.

 

As a result of previous contributions and my conversations with the dealers I decided to keep the lens.

 

All further tests confirm that my original test shots were accurately reflecting the problem. The only question really is whether or not this is normal for this lens, or have I a dud? I don't think that posting further examples will be very enlightening (sorry, Luigi! - if when you get yours back it seems very different from mine I would be interested to hear).

 

I called Leica UK about having it coded and they say that they are out of stock of the coded mounts, have quite a queue, but hope to get a shipment in a couple of weeks. I had hoped that the coding might help the soft edge problem, but I suspect not.

 

Whilst waiting for the coded mount I have manually coded the lens, after quite a lot of hit and miss with the process (finally successfully using matt black modelling paint), and believe that this has eliminated the vignetting. I am still waiting for the 55mm IR filter.

 

The fact that from f/4 onwards the lens is so good encourages me to keep it, but when I take it to be coded I will certainly ask to have it tested (thanks for the suggestion, Woody). I doubt that there is any solution to the problem should it prove it be exceptional.

 

It would be interesting to try another example of the lens, but they are as rare as hens' teeth in London at the moment.

 

 

Tim

[carstenw]

The coding won't help the soft corners, since that is only about colour, but simply removing the old mount, putting a new one on, and then re-tuning it (which I believe they do automatically when changing the mount), might fix a minor problem, if one exists.

[j. borger]

Why would anybody want sharp corners at f2,8 with a 21mm lens....... for what scene:confused:

What is the sense of tests like this .. shooting a landsape at f2,8 to test a lens

[Artichoke]

Why would anybody want sharp corners at f2,8 with a 21mm lens....... for what scene:confused:

What is the sense of tests like this .. shooting a landsape at f2,8 to test a lens

I have to agree with you

I have been puzzled by this thread ...the laws of physics are hard to beat & I cannot fathom why some folks don't understand that virtually any lens improves when stopped down ...this is as true for Leicas as for Nikkors

I love my 21 Elmarit, which is the later version and coded

it is not as sharp at f2.8 in the corners, but at one stop down improves & by f4 is tack sharp ...even opened up is reasonably good ...I think I have taken only one photograph with it wide open

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

I think it did a good job here & produces lovely bokeh for a WA

I find this an outstanding lens on the M8

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[luigi bertolotti]

Why would anybody want sharp corners at f2,8 with a 21mm lens....... for what scene:confused:

What is the sense of tests like this .. shooting a landsape at f2,8 to test a lens

 

OK, you're right, no doubt that when I will try to get a fine lanscape PICTURE (not a test) with my 21, I surely won't use 2,8... but after all we are Leica fans in the hearth... we like to test and discuss of "extreme" situations... If you own a Ferrari, you mainly drive around the right way... but sometimes you shall afford the risk to try something exciting...

[Tim B]

Yes, I admit that I would rarely use this lens wide open for a landscape, and yes, Arthur and J. Borger, as a test mine was not a good one; I've done others since - same result. And yes, of course I know that stopping down a lens improves quality, I've been using lenses for 45 years. But that doesn't mean that I shouldn't expect usability wide open, especially from my Leica lenses.

 

The reason for the thread is simply that I was surprised by the edge and corner quality of a lens with such a good reputation and I wondered whether I had bought a bad example (£1500 is a lot of money for a secondhand lens). Even Leica can occasionally make mistakes! I simply wanted to know what other users of this lens thought about my situation. I'm grateful for the interest shown.

 

In fact it is a lens I bought as much for interiors as for landscapes, like your shot Arthur, and for architectural interiors (e.g. a cathedral visited on holiday) when sharp corners in low light would be useful.

 

And, as Luigi says, we have such high expectations of our lenses that we want to be sure we are getting what we pay for. My new 28mm Elmarit-M ASPH is much better in the corners wide open, although, of course, it is not such an extreme wideangle lens.

 

Thanks all,

 

Tim