Very interesting answer from Leica on 35mm 1.4

[Eoin]

Advertisement (gone after registration)

Tim,

 

The serial no is 3937515. I'm in Dublin (city center). It would be no trouble to post it to you.

[castelletta]

So is there a focus shift - yes? Is it significant? Based on the copy I tested, I would say no.

 

Cheers,

 

Sean

 

We're talking about very different lenses.

 

The focus shift in my lens (35 cron) is huge, very huge, more huge on biggest distances, but very very noticeable also in shortest ones. Like Tim at f/4 (for exaple) I focus on the target and compose with target in the center and the taget is OOF, I focus on the target and place the target far from the center and the target is crisp and in focus.

 

I definitely see SIGNIFICANT focus shift on my lens...and well documented, consistently.

 

My lens is # 40029XX

[Advertisement]

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

[jonoslack]

Hi Everyone!

What a thread - I'm coming to it late, because not having either of the 35 Asph lenses it didn't seem my place to comment. However, I've just received a 2/3 year old 35 'cron Asph, and I'm certainly going to give it a good 'going over'.

One thing does seem to me to be screamingly obvious, and that is that this happens with some lenses and not with others, and that the problem is more relevant at distance than close up. Sean's test, relevant though it is, simply goes to show that his lens doesn't have a significant problem close up - but there are quite enough serious people who have serious issues.

I think that continuing to discuss it in the 'does it / doesn't it' have a problem vein is not very constructive. To do this you either have to postulate that those with 'bad' lenses are totally neurotic, or that those with 'good' lenses are blind as a bat - knowing some of the people concerned, neither thesis is remotely feasible. So it must be an issue.

So - questions do arise -

1. Is it just happening with newer lenses? All the 'okay' lenses I've seen mentioned have serial numbers beginning with 3 rather than 4 - which makes them a couple of years old (correct me if I'm wrong). It seems that those with issues tend to be newer (is that right?).

2. Is it just happening with coded lenses?

3. Is it just an issue with the 35mm lenses (I know it happens to some extent with many lenses)

[guidomo]

Hi everyone,

 

This is my first post to this forum after having only "absorbed" for the past few weeks. Let me say thanks to everyone participating - it is a placing amazingly rich of information as well as of refreshingly nice manners.

 

Let me be quick in reporting the behaviour of my 35/1.4A on my M8 (the oddities in fact being to reason for finding this forum in the first place):

 

- based on both normal-use and focus tests it back-focusses

- badly

- I notice it most in 2-3m shots but it is apparent also when shooting the focus chart at 1m. The backfocus then is about 3cm at f1.4 and increasingly more when stopping down. Between 2.0 and 5.6 the point I focus on is out-of-focus.

- I have a current (coded) black model, S/N 40069xx.

 

Any more infos I can help with, let me know.

 

Guido

[guidomo]

According to

 

http://www.leica-camera-user.com/digitalforum/20185-front-backfocus-mit-35mm-und-m8.html#post214155

 

this issue may be addressed by the May Issue of LFI. Does anyone have access to that?

[castelletta]

- I have a current (coded) black model, S/N 40069xx.

 

Guido

 

Eheh....interesting coincidences are now emerging more clearly...come on Leica people

[jonoslack]

Advertisement (gone after registration)

Right - this is a 35 'cron aspheric f2 serial 3945***

so, it's 2 or 3 years old.

I've put together two tests, one at around 30 feet:

In this case the point of focus was the handle of the basket

 

and another at around 4 ft:

here the point of focus was the word 'find' near the bottom of the text

 

Initial response seems to me to suggest that the focus point does, indeed, move backwards as you stop down, so that in the close up shot at around 5.6 it's noticeably out of focus, by f8 the dof has begun to correct the situation, and by f11 it's pretty much okay.

 

the distance shot may be showing the same thing, but it's not terribly obvious, and not very important.

 

So - this is an older, uncoded lens, and although it does show the issue, it doesn't seem to be critical.

Do I dare send it in for coding!

[mjh]

Let me be quick in reporting the behaviour of my 35/1.4A on my M8 (the oddities in fact being to reason for finding this forum in the first place):

 

- based on both normal-use and focus tests it back-focusses

- badly

- I notice it most in 2-3m shots but it is apparent also when shooting the focus chart at 1m. The backfocus then is about 3cm at f1.4 and increasingly more when stopping down. Between 2.0 and 5.6 the point I focus on is out-of-focus.

Since at f1.4 there is no focus shift (by definition), the backfocus at f1.4 must be due to other causes – I suppose that either the lens or the rangefinder needs to be recalibrated. With a focus-shift-induced backfocus on top of an initial backfocus, the results must be worse than with focus shift alone (as if that wasn’t bad enough).

[wattsy]

Do I dare send it in for coding!

 

I have the same dilemma. I have a recent 35 'cron (serial number 4001xxx) that must date just before they were sold with codes. Whilst it does focus shift a little, it is well within the bounds of DOF and acceptable. I would like to send it to MK for coding but I'm not sure I want to risk it coming back and behaving like Roberto's lens.

[sean_reid]

OK, I just did tests at about 20 feet and there's no detectable focus shift at all from this lens at that distance. So let me clarify my post above lest there be any misunderstanding. Based on what I'm seeing from this lens, the inherent focus shift in the 35/1.4 Aspherical is very slight and is most visible at close focus distances where DOF is more limited. In other words, the design of the lens does not seem to create significant focus shift at all.

 

I realize that people are reporting focus shift problems with other examples of the lens. I don't know why that would be the case but if the sample I'm testing shows only very minor focus shift then significant focus shift is not inherent in the lens design. The focus shift that Leica was describing is minor and it was what I was seeing in the three foot tests. It would be interesting to know why some individual examples of the lens might not be performing at the intended level.

 

Tim, perhaps you want to do a survey of people who own this lens (asking them about focus shift, age of lens, etc.) to try to narrow down what might be common traits (among a larger sample).

 

Cheers,

 

Sean

[tashley]

OK, I just did tests at about 20 feet and there's no detectable focus shift at all from this lens at that distance. So let me clarify my post above lest there be any misunderstanding. Based on what I'm seeing from this lens, the inherent focus shift in the 35/1.4 Aspherical is very slight and is most visible at close focus distances where DOF is more limited. In other words, the design of the lens does not seem to create significant focus shift at all.

 

I realize that people are reporting focus shift problems with other examples of the lens. I don't know why that would be the case but if the sample I'm testing shows only very minor focus shift then significant focus shift is not inherent in the lens design. The focus shift that Leica was describing is minor and it was what I was seeing in the three foot tests. It would be interesting to know why some individual examples of the lens might not be performing at the intended level.

 

Tim, perhaps you want to do a survey of people who own this lens (asking them about focus shift, age of lens, etc.) to try to narrow down what might be common traits (among a larger sample).

 

Cheers,

 

Sean

 

 

I have a new, currently emerging theory based on some ideas I got from you Sean, and from what other very recent posts here have implied, and on comparisons with my new 35 Skopar, which is well-behaved. I also dropped by my dealer this morning and tried a brand new 35 cron off his shelf, coded 4*******.

 

Stand by for further reports but I want to test this out very carefully before I shoot. (Shoot my mouth off rather than with my camera I mean!)

 

Best to all

 

Tim

[Guest guy_mancuso]

Hell you guys are making me run a test on my 35 lux. Mine is several years old and went in for coding and seems pretty darn well behaved. Guess i will have to retest it and see what i come up with

[tashley]

Now as we all know, all new theories are subject to group sanity testing so forgive me if I'm missing something obvious.

 

Background: Amidst some emerging evidence that 4***** series (i.e. recent) lenses are more likely to exhibit 'the behaviour' I thought I would test my 35 cron against my new 35 Skopar at a distance of 20 feet. The result was very startling: even wide open, the Skopar at f2.5 was much sharper on center than the Cron at f2, and the same was true at f2.8 and throughout the aperture range. So I thought I should try Sean's focus-bracket test, and for variety I used my black body this time (Chrome first time).

 

I did this by focussing the Skopar once only, since it seems to agree with my RF on both bodies (like my other lenses). But with the cron, I focussed with the RF and shot F2 thru 16, then nudged the focus very slightly closer, repeat, nudge closer, repeat, nudge closer, repeat. That's four sets with the cron, and by the fourth there was a very clear double image in the rangefinder (the second and third sets showed minor mis-registration when viewed through the RF).

 

At 100% on screen, the Skopar had pretty good focus in its set as focussed on the RF. But the Cron was pretty fuzzy in comparison. It was not until set 4 on the cron, the most 'pulled forward', that it beat the Skopar wide open. And then it did exactly what Sean has observed: it progressively shifted focus in the mid-aperture range but even at its worst, it was perfectly usably sharp. F4 on centre was not quite as sharp as f2 but still pretty good.

 

So then I did the Doh! test. I pointed the Cron at infinity (hills and trees miles away with buildings about a mile away) and shot one with the lens set hard at infinity, then three more pulled closer. Infinity focus was blurry, and the sharpest was a toss up between three and four, in both of which the RF was clearly showing doubles.

 

So my new theory is that something in many of the 4****** crons is causing them to focus slightly past infinity. But if you bracket focus to get sharp wide open, then whilst there is focus shift, I would happily live with it since I don't think it would show in all but the very largest of prints if at all.

 

Now there are clearly a lot of variables. One of my bodies (the Chrome) I have tweaked the RF on though I suspect it is now back to where it started. The black has never been tweaked and works with my other glass happily, though it is true that my 50 lux on both bodies gives a minute increase in sharpness at infinity wide open if the focus is pulled a minute amount forward. However the 50 lux is impeccable in daily use and the 35 cron is definitively not so.

 

So all this time with the cron, I may have had a lens that was exacerbating its own mild focus shift by having a bigger problem, being that it backfocusses in general. I didn't notice before since I thought 'Oh well it's bound to be a bit soft wide open' and then I saw it against the Skopar and realised this was untrue.

 

Now, I am clearly not the only person having this issue. I also recently had to send back a new 28 Cron that was focussing so far past infinity that Stephen Hawking couldn't have found the place it focussed on. That was also a 4**** lens.

 

So can those of you with troublesome 4**** crons try my test and report back?

 

And can those of you with larger brains than me criticise my assumptions, methods or conclusions so I can narrow this down?

 

Thanks guys!

 

Tim

[thpeters]

Following this tread now for a couple of weeks, my question is?

 

Have somebody that problem with that lens before, when using the lens on a Leica film

camera?

 

Maybe somebody can compare that and put the results in this forum, to make it more clear, if there is no problems with using a film camera with this lens.

Then it must be the combination lens and M8.

 

I have wrote it earlier and for now keep my statement, the distance between lens and film and lens and digital sensor is not 100% the same,(my humble opinion), if the test shows there is a different between the 2 cameras using the same lens. I'll bet that the lens with the film camera is just fine.

 

Theo

[guidomo]

I may be wrong with this, but my understanding so far is that any lens will show a degree of backfocus when stopped down, just that the magnitude of this varies a lot and depends on the lens design (and from Sean's comment above I understand that the 35 Lux Asph has not a particularly strong tendency to do this - and I would not expect this to very between people's lenses). However, normally the inherent backfocus is compensated by a combination of the following two factors:

(1) DOF increases when stopping down

(2) adjustment is supposed to be such that at aperture fully open the focus point should be slightly IN FRONT of where you focus to, I understand a 2/3 w.r.t. 1/3 ratio is the target.

 

Is this in line with cumulative wisdom among people here or did I get it wrong?

 

Point (2) seems to be what is wrong with my lens, at full aperture I am "back" already, and thus run out quickly, with DOF increase not being able to compensate that. Only at f8 it has cought up.

 

Since at f1.4 there is no focus shift (by definition), the backfocus at f1.4 must be due to other causes – I suppose that either the lens or the rangefinder needs to be recalibrated. With a focus-shift-induced backfocus on top of an initial backfocus, the results must be worse than with focus shift alone (as if that wasn’t bad enough).

 

Leaves the question whether it's the lens or the M8 but after all the reports I've read I tend to point to the former (don't have other fast lenses to test yet unfortunately).

[tashley]

Guido, please let us know the first four digits of your serial number, black or chrome, coded/uncoded/retro-coded, and whether you have other lenses that appear to focus well at all apertures and distances to subject.

 

Otherwise your grasp on the wisdom seems about right to me!

 

 

Best

 

Tim

[sean_reid]

Tim,

 

If you tested another 35/1.4 Asph, I'll bet you'd find much the same thing that you observed with the Summicron (28?).

 

Cheers,

 

Sean

[tashley]

Tim,

 

If you tested another 35/1.4 Asph, I'll bet you'd find much the same thing that you observed with the Summicron (28?).

 

Cheers,

 

Sean

 

 

I actually am using a 35 cron, (I sent the two luxes back) but I tried a brand new 35 cron at my dealer today and it behaves exactly as does mine: it does not focus correctly as per the rangefinder wide open, but needs to be tugged a couple of notches closer to achieve sharp results.

 

Any theories?

 

If you trust that your M8s RF is correctly calibrated, then does your press pool 35 cron focus accurately wide open with the RF or was the 'best' series one of the bracketed ones?

 

Best

 

Tim

[guidomo]

Tim,

- I have a current (coded) black model, S/N 40069xx.

(similar to yours I believe.)

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[guidomo]

Tim and others with the backfocus problem,

 

Did you follow this thread where people have adjusted their M8's rangefinders?

http://www.leica-camera-user.com/digital-forum/13933-new-backfocus-thread-solution.html

 

Have you checked enough (fast) lenses to be able to remove the camera from the equation? I am not 100% sure I trust my M8's rangefinder (thanks also to Michael for his comment pointing this way).