Differential Focus Errors

[pixelman]

Advertisement (gone after registration)

Hi,

 

I'm a new M9 owner (long time M film owner, and I owned an M8 for a weekend - long story). I seem to have a curious focusing situation. Tests below were done in bright afternoon sunlight (rare for Vancouver this year...) using a Lens Align test target. Numbers below are the optimum focus position on the lens align scale. All exposures were ISO 160 at 1/500th or faster with the lenses wide open:

 

Leica 35mm 'cron : +4

Leica 50mm 'cron: +14

Voigt 75mm f2.5: +14

 

Other factors: My 50mm 'cron could use some tightening in the mount (there is a small amount of play in the mount). The 75mm Voigt is a recent ebay purchase and I don't know the history.

 

Is it common to see differential focusing results like these?

 

I recently noticed some film shots with the 75mm that looked less than sharp - is it possible/easy to get a Voigt lens adjusted for focus position?

 

I would really like to NOT send back my M9 (just got it on Friday...).

 

Any thoughts from this group as to what else to test/try, think about?

[giordano]

What results do you get using the same test setup with your film M body/ies?

[250swb]

If you only got the camera on Friday I would say use it and see what happens.

 

I'm not sure setting out to find fault with focus tests is the right way to go. Its a way to confirm your suspicions if you feel something is wrong with a lens in real world shooting after consistent failures, but a test chart isn't the real world where the inherent focusing characteristics of a lens need to be understood. There is also the very very small margin of error to consider with a digital sensor, which unlike film has no depth (the thickness of the emulsion) to mask user errors. These things become second nature, and a higher number of focus failures from say using a longer lens are common, as are focus errors caused by focusing at minimum distance wide open, everybody will have more errors even on perfectly calibrated cameras.

 

Steve

[Xmas]

Hi

 

If is not unknown.

 

Repeat the tests wide open and at decreasing apertures, tabulate a matrix.

 

I think the CV lens can be adjusted easily, the CV 40mm f1.4 can, but it wont help much if you have lots of focus shift, note it should be ok for shift.

 

Leica (Solms) charge a lot to set up lenses to the tighter tolerances needed and some will still show focus shift. This would be less apparent with a film M, partly cause most people would not bother.

 

Noel

P.S. we live in the dark ages too

[jaapv]

These findings are quite normal for lenses that have been built in the film era. A digital sensor needs a more narrow tolerance span in focussing than film. I doubt that your findings will have any impact on real-life images, but if they do, send the lenses in to Leica for "adjustment to M8/M9 standard" as it is officially called.

The Voigtlander lens, well, you will have to use a third-party service for that one.

[plasticman]

This would be less apparent with a film M, partly cause most people would not bother.

 

I think this is such bs about film and focussing accuracy. I can see the acute accuracy of focussing in my film images, just as in my digital images. The main difference is the difficulty of setting-up tests with a film camera, where one lacks immediate visual feedback that can be adjusted for user error.

 

Also the in-focus to out-of-focus transitions are gentler with film (that's where the film thickness comes into play), and therefore the margin of error is slightly increased.

 

It's certainly possible that the RF is misaligned even on a new camera. I'd recommend trying to find a local Leica expert - preferably with a collimator - that could at least isolate lens problems. However, when my brand new M8 needed to go back to Solms to re-align the rangefinder vertically, it came back absolutely perfectly aligned and has given me zero problems since - so the trip may be worth it.

[jaapv]

Advertisement (gone after registration)

Sorry, Mani it is not bs. Film has a physical thickness, which means it has an uncertain plane of focus. a sensor has a virtual thickness which is determined by th Nyquist frequency, which is much less. Film cannot be perfectly flat so parts of the plane of focus will be inside the emulsion and others outside, a sensor is as perfectly flat as is possible.

This has an impact on the transition of the sharpness into unsharpness as well, making a sensor more defined (or harsh as film lovers call it) thus making the sharpness more noticable on a sensor.

I thought you revelled in these differences instead of denying them. It is the main thing that determines the "look"of film as opposed to digital.

I don't think Leica would set up a testing rig specifically to test on a digital level spending an amazing number of Euros if these were not basic facts.

[plasticman]

Sorry, Mani it is not bs. Film has a physical thickness, which means it has an uncertain plane of focus. a sensor has a virtual thickness which is determined by th Nyquist frequency, which is much less. Film cannot be perfectly flat so parts of the plane of focus will be inside the emulsion and others outside, a sensor is as perfectly flat as is possible.

This has an impact on the transition of the sharpness into unsharpness as well, making a sensor more defined (or harsh as film lovers call it) thus making the sharpness more noticable on a sensor.

I thought you revelled in these differences instead of denying them. It is the main thing that determines the "look"of film as opposed to digital.

I don't think Leica would set up a testing rig specifically to test on a digital level spending an amazing number of Euros if these were not basic facts.

 

I meant it's bs that film users "wouldn't bother".

 

I can perfectly well see that the in-focus to oof transitions are much more 'luscious' in film images, but the focus plane is very easily identified when using lenses wide open and close up even on film. Also the very different characteristics of different lenses: judging the exact focus plane in a Noctilux portrait vs a DR Summicron, for instance, is a challenge of differing magnitude.

 

Anyway, let's not make this yet another film vs digital thread. My point was:

 

1. the OP should have been able to identify focus errors even on film images (and I'm assuming he's looked at those images at sufficient magnification/enlargement)

2. the RF can be misaligned even on a new camera

3. sometimes a local expert can be cheaper/better than sending everything to Solms

[jaapv]

Ok - you assigned the bs differently than I did Yes, it might have been visible on film, but not neccesarily so, as both instant feedback and the higher precision of digital make the testing much more easy and direct. The OP found the problem, which was seemingly non-existent before in "tests" instead of photographs. And yes and yes In this case, however I think prima vista that the lenses may well be the culprits.

[Xmas]

Hi Plasticmam

 

I think the term 'most people' has been misunderstood, e.g.

 

I neither said you would not notice.

Nor the O/P

I never use the rangefinder, so I'm not going to notice...

 

Lots of people dont use the lenses wide open, e.g. cause their style requires depth of field. They are not going to notice.

 

Some people woulod not use a focus target, and might put down focus errrors to themselves.

 

Perhaps English is not your 1st Language?

 

Noel

[jaapv]

Possibly the OP would not have noticed, like most people as you say, but now that he started out by "testing" he will certainly notice any focus error in his shots, whether present or not. Such is human nature.

[plasticman]

Perhaps English is not your 1st Language?

 

What is this thing about "Perhaps English is not your 1st Language?" Can't we keep the personal attacks out of this damn forum? I'm so unbelievably tired of it. No-one seems able to take any criticism of anything they say without replying with some sort of ad hominen retaliation.

 

I've read the 'focus issues are not noticeable on film' thing innumerable times on this forum - so often that it's virtually presented as a truism in the digital section. In my opinion that's not the case, and I questioned it when you restated it with the additional "most people would not bother".

 

Obviously I shouldn't use terms like 'bs', but I seriously don't see the need to do the personal bitching.

[plasticman]

Possibly the OP would not have noticed, like most people as you say, but now that he started out by "testing" he will certainly notice any focus error in his shots, whether present or not. Such is human nature.

 

Whenever I buy a new lens, naturally the first thing I do is slot it onto the M8 and aim it at a ruler. It wouldn't occur to me to do the same with a film camera unless there really was no option.

[Xmas]

Hi Plasticman

 

I thought you were from The US, I see that Sweden is more likely, too late.

 

I said

 

'This would be less apparent with a film M, partly cause most people would not bother.'

 

I'm a stats person and >50% is 'most people'

 

I still suggest partly 'most people would not bother', e.g. ...

 

Whenever I buy a new lens, ... It wouldn't occur to me to do the same with a film camera unless there really was no option.

 

It confuses people when you dont read what is intended and react out of context, I did not treat it as a personal attack, I assumed my English was too oblique.

 

Noel

[plasticman]

I'm becoming sensitized because the general tone of the forum is deteriorating, in my opinion. I myself (regrettably) use terms like bs, but I try to avoid personal jabs (even I'm not above them, I'm afraid to say), but it seems to me that the general drift is towards vindictive reprisals over the most trivial disagreements. I'm sorry I mistook your intention.

 

With regard to the focus issues, as the OP stated, he'd already seen problems with his 75 on film - thereby my suggestion to contact a local repairman. Leica wouldn't help with a CV lens, in any case, and I'm on a personal mission to support local experts - when they eventually disappear, we'll have only Solms left to turn to - a less than optimal sitiuation.

[lct]

I would love to speak English as well as does Mani and avoiding paranoia in the same time.

[Xmas]

With regard to the focus issues, as the OP stated, he'd already seen problems with his 75 on film - thereby my suggestion to contact a local repairman. Leica wouldn't help with a CV lens, in any case, and I'm on a personal mission to support local experts - when they eventually disappear, we'll have only Solms left to turn to - a less than optimal sitiuation.

Hi

 

I'd not worry about the forum cyclic tolerance problems these occur like the economy or traffic jams on autoroute i.e. normal, statistics again.

Note we have lots of problems communication with US people, even verbally as we dont use the same English... My English is not that clear...

Yes the OP probably has three lenses that need adjusting, for the M9 & his film camera may also need adjusting, unless he is lucky.

I've not have any Leitz servicing since '77, learnt a lesson once.is enough. We used to have a Leitz service centre in central London that did over the counter jobs while you read glossy magazines in their showroom, all gone now.

Solms do charge a lot for a lens re calibration BTW.

For the Kiev and Contax II I use I just rebuild them at need, Canons I think about first, Leicas are more difficult, I could procure or make more tools. We do have three good repair centres in UK, and 'Red Dot' has hs own man.

 

Noel

[pixelman]

The (used, but new to me) 75mm lens doesn't look sharp on film. It has been a long time since my 50mm 'cron has been used with film (I've been in the DSLR world for sometime and recently started using my M film bodies which lead me to just get an M9). The 35mm 'cron looks good on film. I am going to try another lens or two that I have to get a bit more data and I'll report back to this group. I'm hopeful that I have a lens issue and not an M9 issue.

[ho_co]

Since the 75 doesn't look sharp on film, send it first for adjustment.

 

I was happy with my 75/1.4 on film (I wasn't too demanding, it seems), but with the M8 I discovered that it was backfocusing badly at middle distance. Close-up it was okay, and at infinity it was superb.

 

I sent it to Leica, and it came back greatly improved, at both medium and close distances.

[Advertisement]

Hallo,

Look here: Leica M

[pixelman]

One more data point: quick test of my Voigt 28 1.9 shows it seems to ficus ok (very small amount of back focus). I'm thinking the M9 doesn't need to go back, but I have some lens work/thinking to do.

 

I'm thinking about a new 50mm, but they are tough to find!

 

I'll try and take some more data as soon as the RRS M9 baseplate comes in and the sun comes out again...