[need advice] 35/2 Asph: focus shift, how much is too much?

[Steve McGarrett]

Advertisement (gone after registration)

Hello everybody,

 

I know that Summicron 35 Asph. (v5), does focus shift a bit if used on digital, while not as much as Summilux Asph pre-FLE or others, older design.

 

I've used one on film for years without any problem back in the 90s, sold it many years ago, and recently I've bought another one, used, for my M digital.

 

I'm not looking for perfection, nor doing endless testing on focus chart to split hairs, but this one seems to show too much focus shift even in real pictures: please look at these samples (I have many many others, and, by the way, when I engage the EVF I can clearly see the difference between the sharp, magnified image in EVF and the yet-not-coincident one in rangefinder patch at f/4, so I can exclude a focusing error from my part)

 

(I use Dropbox to upload full-res samples)

 

First sample, f/2, focus on the first plant (a couple of meters away from the camera), seems to me spot-on (and a pretty good wide open performance)

 

https://www.dropbox.com/s/6kq47e913lugejp/216%20f2%20plant.jpg?dl=0

 

Second one, f/4, focus on the same plant... but sharpness has shifted to the plant behind the first one, and to the background: the plant in the foreground is clearly less sharp than in the wide open shot.

 

https://www.dropbox.com/s/0qej8xtdvxn7042/215%20f4%20plant.jpg?dl=0

 

Third one, f/4, focus on the little shovel's handle... now first plant seems to be in focus as it should:

 

https://www.dropbox.com/s/2v9kjqn4o2gbwxx/214%20f4%20handle.jpg?dl=0

 

Should I return the lens, or any Summicron 35 v5 behaves more or less that way when used on M240? (today is the last day I can return it, by the way)

 

Thanks in advance, any advice will be appreciated.

 

S.

 

edit: fixed link URLs

Edited September 28, 2016 by Steve McGarrett

[michaelwj]

Since you have only today, I'll offer my 2c.

 

It seems to worry you, would it worry you in use? Would you have a nagging sense that it's not quite right?

 

You are not happy with it, so return it and try another. I have the feeling you would always wonder "what if..."

 

If the other next lens is the same, then that's probably just how it is, but at least you know and can make a better decision. Then you can decide if it is the lens for you or not.

 

*I assume everything was shot on a tripod?

[Advertisement]

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

[Steve McGarrett]

michaelwj, on 28 Sept 2016 - 14:34, said:michaelwj, on 28 Sept 2016 - 14:34, said:

Since you have only today, I'll offer my 2c.

 

It seems to worry you, would it worry you in use? Would you have a nagging sense that it's not quite right?

 

You are not happy with it, so return it and try another. I have the feeling you would always wonder "what if..."

 

If the other next lens is the same, then that's probably just how it is, but at least you know and can make a better decision. Then you can decide if it is the lens for you or not.

 

Yeah, fact is this is the second one already (first one was so badly decentered I return immediately, without any further testing). I start to think I'm a bit unlucky with cron 35s and/or that Leica's QC has worsened a bit in the last 10 or 15 years 

 

This one seems to be a solid performer when in focus, very nice lens... probably if I had a SL and not a M I will be perfectly happy with it.

 

*I assume everything was shot on a tripod?

 

Not the sequence above, but others ones, yes, and trust me, it's exactly the same thing, short or long distance, hand-held or tripod: spot-on at f/2, still good at f/2.8, a bit blurred at f/4 and f/5.6, good again at f/8 (but barely sharper than f/2 shot). Focusing with EVF at every aperture, instead, lens is a lot better at f/4 than at f/2, like one should expect (and like you can see in the third sample I've uploaded above).

 

Thank you very much for your advice, Michael.

Edited September 28, 2016 by Steve McGarrett

[edwardkaraa]

I'm not sure about the cron, but Zeiss calibrates the similar ZM 35/2 at f/2.8 in order to mitigate the shift issue. You get front focus wide open and perfect focus at all other apertures. Your lens according to your description is calibrated for wide open (I didn't look at the photos). I would suggest you have it adjusted for f/2.8.

[lct]

My 35/2 asph from 2010 has no focus shift on M240, M8.2, Sony A7s mod and Fuji X-E2. Seems like your lens and/or body needs some calibration. Probably your lens if it's been made in the film days.

[Steve McGarrett]

edwardkaraa, on 28 Sept 2016 - 14:49, said:

I'm not sure about the cron, but Zeiss calibrates the similar ZM 35/2 at f/2.8 in order to mitigate the shift issue. You get front focus wide open and perfect focus at all other apertures. Your lens according to your description is calibrated for wide open (I didn't look at the photos). I would suggest you have it adjusted for f/2.8.

 

Then I fear the lens will be unusable at all at f/2, if calibrated spot-on at f/2.8, since the focus shifting seems to me quite severe, looking at the pictures.

 

Especially if Lct says his Summicron does not shift at all (mine is 4.1xx.000, so it's around 2010 too...)

Edited September 28, 2016 by Steve McGarrett

[edwardkaraa]

Advertisement (gone after registration)

Then I fear the lens will be unusable at all at f/2, if calibrated spot-on at f/2.8, since the focus shifting seems to me quite severe, looking at the pictures.

 

Especially if Lct says his Summicron does not shift at all (mine is 4.1xx.000, so it's around 2010 too...)

From the reviews I have seen the cron does indeed suffer from some focus shift, but it shouldn't be as bad as you describe.

[lct]

[...] Lct says his Summicron does not shift at all (mine is 4.1xx.000, so it's around 2010 too...)

 

You might wish to take a look at my lens' behavior below. A good workshop should fix your issue.

http://www.l-camera-forum.com/topic/255654-new-summicron-asph-35mm-how-and-why-did-they-improve-the-current-one/page-4?do=findComment&comment=3004109

http://www.l-camera-forum.com/topic/255654-new-summicron-asph-35mm-how-and-why-did-they-improve-the-current-one/page-4?do=findComment&comment=3004366

[michaelwj]

Hi Steve,

What did you decide to do?

[Jeff S]

My sample mildly focus shifts at f4-f5.6, the same as Sean Reid's experience when he reviewed the lens years ago (I no longer subscribe).  But it's not a problem with my prints, which is what I care about, and in any event is easily accommodated when shooting once the issue is identified.

 

Only you can decide if it's a problem for you.

 

Jeff

[Steve McGarrett]

michaelwj, on 28 Sept 2016 - 23:26, said:

Hi Steve,

What did you decide to do?

 

My vendor proposed me to send the lens to Leica, and I accepted with the agreement that I can return the lens afterwards, if they don't fix the problem.

 

Btw, courier will pass on Monday, so before that I still could decide to keep the lens as is - but perhaps it's better to see if the problem can be solved by Leica.

 

Thank you very much to you all, guys, for your kind answers.

Edited September 29, 2016 by Steve McGarrett

[Frase]

I would give the lens another test using a tape measure or similar, I don't have the V5 but I do have a V4 and that has no focus shift maybe I'm just lucky but I've never had any problem with focus shift on any of my M lenses.

Good luck.

[Jeff S]

I would give the lens another test using a tape measure or similar

 

He did better than that....he checked magnified view in his EVF (which is what the camera actually 'sees') and compared it to the RF focused image. Assuming he used a tripod, this accurately demonstrated that there was a focus distinction.  And assuming this is the only lens he has exhibiting a difference, then the issue relates to the lens, not to the camera.

 

The only issue remaining is whether it's enough to bother him....or whether Leica can improve the lens.

 

In my own testing, I don't care unless it negatively impacts my print results....and it doesn't.

 

Jeff

Edited September 29, 2016 by Jeff S

[Steve McGarrett]

He did better than that....he checked magnified view in his EVF (which is what the camera actually 'sees') and compared it to the RF focused image. Assuming he used a tripod, this accurately demonstrated that there was a focus distinction.  And assuming this is the only lens he has exhibiting a difference, then the issue relates to the lens, not to the camera.

 

The only issue remaining is whether it's enough to bother him....or whether Leica can improve the lens.

 

 

Here's the point. We'll see...

[wlaidlaw]

Sample variation on both the 35/2 ASPH and pre-FLE 1.4 ASPH appears to be wide. I seem to have won the lottery with examples of both on which the aperture shift, even on the M240 is not a problem. My 1.4 is a chrome/brass bodied one which from lots of posts over the years, seems to be less problematical than the black alloy bodied model, maybe because it is easier to machine brass to very tight tolerances than aluminium alloy. During a visit to Solms in 2007, I discussed the problems of the 35/1.4 APSH, which were big news at the time. The technician who was showing me round admitted these lenses were a nightmare to assemble, as the tolerances were pretty much beyond human ability to get perfect and therefore, there was a big element of chance on the ultimate quality of the lens. My 2014 year 35/2 had to go straight back to factory when new, as its optical focus did not agree with its RF focus. It is a matter of seconds to do this now on the M240. When it came back it was perfect. I don't use the chrome/brass 35/1.4 a whole lot now, as it seems very heavy compared with the alloy 35/2. 

 

Wilson

[james.liam]

At >$3k a copy, I am dismayed that so many manage to leave the factory in such unusable condition. 'lct' seems also to have hit Lotto and has a copy with little shift at all. Others, particularly some notable reviewers, have not been as lucky. I have as yet to see tester's comments on the new iteration and whether this has been mitigated.

[wlaidlaw]

At >$3k a copy, I am dismayed that so many manage to leave the factory in such unusable condition. 'lct' seems also to have hit Lotto and has a copy with little shift at all. Others, particularly some notable reviewers, have not been as lucky. I have as yet to see tester's comments on the new iteration and whether this has been mitigated.

 

James, 

 

I agree. Since the M240 came out, as long as your M240 is correctly calibrated, it only takes seconds to check RF accuracy against optical focus accuracy at various distances. In recent years my 35 ASPH Summicron, as set out above had to go straight back (it was an ex-display one from the Montmartre boutique, so could not be exchanged). The same story with my 50/.95 Noctilux, where Leica offered an exchange but told me I would have to wait three months until the next batch, so I opted for adjustment. At least then you know your lens should be correct, whereas the exchange lens could easily have problems. I suggested that Leica must have employed Mr. Magoo (whur's that dang Wabbit), as their sign off lens checker. Recently with the SL, there seem to have been very few lens problems in comparison. Only one or two reports of the zoom locking solid, which was a screw on one of the sleeves coming undone. 

 

Wilson

[yeeper]

I did a bit of 35mm testing recently, including 2 samples of the 35mm Summicron ASPH.  Both suffered noticeable focus shift.  Interestingly, my Voigtlander 35mm Ultron f1.7 (ltm) and the ZM Biogon f2 had zero shift.  It was a test to convince me to buy the 35mm Summicron.  I bought the Biogon instead.  (My Ultron has great rendering, but suffers from a .9 close focus and crazy chromatic aberrations.)

[wlaidlaw]

I did a bit of 35mm testing recently, including 2 samples of the 35mm Summicron ASPH.  Both suffered noticeable focus shift.  Interestingly, my Voigtlander 35mm Ultron f1.7 (ltm) and the ZM Biogon f2 had zero shift.  It was a test to convince me to buy the 35mm Summicron.  I bought the Biogon instead.  (My Ultron has great rendering, but suffers from a .9 close focus and crazy chromatic aberrations.)

 

I replaced my f2 35 Biogon with the ASPH Summicron. Whereas the Biogon is a wonderful lens above f2.8, where I suspect it out performs the Summicron, wider open than f2.8, the Biogon shows that it may have been "stretched" beyond its limits. As many of my photos are early morning or evening, the f2 performance is important to me. Sean Reid tested 35mm RF lenses a few years ago and came to pretty much the same conclusion. I also don't think he was a fan of the super high contrast of the Biogon, in contrast to which, I like it. The Summicron is also quite a bit smaller than the Biogon, which is down to the differing philosophies in lens design between Leica and Zeiss. Leica try to do the corrections in as few steps as possible even if it means expensive aspherical elements and anomalous dispersion glass. Zeiss prefer to do their correction in smaller steps but more of them and generally not requiring exotic elements. Either Zeiss are awfully good at making and testing their lenses or their designs are more assembly error robust than Leica's designs. A "wrong un" seems a rare occurrence for the ZM lenses but almost par for the course for Leica. A bit of both factors I suspect. 

 

Wilson

[Advertisement]

Hallo,

Look here: Leica M

[Steve McGarrett]

I did a bit of 35mm testing recently, including 2 samples of the 35mm Summicron ASPH.  Both suffered noticeable focus shift.  Interestingly, my Voigtlander 35mm Ultron f1.7 (ltm) and the ZM Biogon f2 had zero shift.  It was a test to convince me to buy the 35mm Summicron.  I bought the Biogon instead.  (My Ultron has great rendering, but suffers from a .9 close focus and crazy chromatic aberrations.)

 

Was it comparable to my samples above, Yeeper?