135mm Lens on M8?

[lct]

Advertisement (gone after registration)

I don't know this adapter sorry. I would try a Voigtlander (Adapters: LTM to M) if you don't want to send your lens in right now but my feeling is the lens needs calibration.

[maccaco]

I don't know this adapter sorry. I would try a Voigtlander (Adapters: LTM to M) if you don't want to send your lens in right now but my feeling is the lens needs calibration.

 

I will seek a Voigtlander adapter ... but I think you're right with the issue of calibration. Thank you very much for your help

[Advertisement]

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

[lct]

...I had to shoot the APO-Telyt mostly stopped down... The point is: the M8/ M9 rangefinder is perfectly capable of reliably shooting fast, long lenses wide open.…

Stop down is not f/3.5 and M8 is not M9 with respect and friendship as well.

At f/3.5, any 135mm lens is simply out of the accuracy range of the M8 w/o magnifier.

Edit: Not the case of the Elmarit 135/2.8 the goggles of which have a 1.5x magnifier.

Edited August 8, 2011 by lct

[maccaco]

You know of a professional (in Europe) to calibrate RF lenses?

Regards

[lct]

Canon lenses i don't know sorry.

[menos I M6]

Stop down is not f/3.5 and M8 is not M9 with respect and friendship as well.

At f/3.5, any 135mm lens is simply out of the accuracy range of the M8 w/o magnifier.

Edit: Not the case of the Elmarit 135/2.8 the goggles of which have a 1.5x magnifier.

 

Reread my post please - both samples shown, have been shot wide open with either the 135/3.4 APO or the more critical to focus 90/2 pre ASPH on a M8.2

 

Your claim, that the Leica M RF cannot be used, to focus such lenses reliably, but the 135 Elmarit, as of it's magnification goggles is … wrong (respectfully).

 

Here is a short list of things to check for the OP:

 

1)

- take the lens off the camera

- take any adapters off the LTM mount

- focus the lens to it's mechanical infinity stop

- measure with a calibrated caliper precisely the dimension along the lens axis from the LTM mounting flange to the now fully extended RF coupling device.

 

The measurement should be precisely 7.5 mm.

The important word here is precisely (with a small tolerance of ±0.03 or less.

With this measurement being as "slightly off", as just 0.1 mm (so watch out for your measurement technique and measurement equipment !), the RF focus patch in the viewfinder will be off coincidence by as much as a few millimeter, leading to misfocussing across the whole range.

 

2)

You have verified, that the RF coupling at infinity is at a correct measurement from LTM flange to coupling device of 7.5mm precisely.

 

Now to the next potential cause of problem:

- take your LTM - M adapter and measure the thickness of the flange - it should be 1mm precisely

- if it is, screw it on the lens mount (does the RF coupling device line up perfectly free at the position of the RF cam mill out?)

- now repeat the above measurement, precisely measuring the extension of the RF coupling device from the M-mount flange

 

This measurement should now be precisely be 6.50 mm, not 0.01 more or less.

If it is not, find an LTM-M adaptor, that is.

 

My experience with diverse cheaply made adapters is, that they are absolutely hit and miss. Many have the correct flange thickness of t=1.00mm, but instead have their LTM thread cut wrong, so either they do not line up with the coupling device or worse yet, do end up with a flange to RF coupling of a different measurement than 6.50mm from M mount flange.

 

regarding adapters:

I have good experience with Voigtlander adaptors. I have bad experience with several Asian made, cheap adapters (read: sold on eBay for 40 − 50 USD, actual sales price in Chinese camera shops: 8 − 10 USD)

Recently, I have handpicked one (!!!) out of a batch of about twenty (!!!) such adaptors at diverse local camera shops in Shanghai, which actually had not its RF cam cutout positioned at the wrong place, or it's locking mill out machined at the wrong position (ending up between two different frame lines, when fully locked), or it's pre milled 6-bit coding recesses milled at a slightly wrong position, leading to spotty 6-bit recognition after properly painting the code.

 

So, you have now verified, that your lens has it's lens mount and RF coupling properly adjusted for infinity for a Leica M camera body (we are far from being finished yet).

 

Next is, verifying, that your camera body is properly adjusted regarding it's RF setting at infinity (if this is off, the whole thing is meaningless, and you can stop and ship the whole thing over to a knowledgable technician. If you are still with me, read on…

 

Set the camera on a tripod, point the lens to a verified, usable infinity target with clear view (the moon is great, a distant radio tower is great as well (make it at least a few kilometers away).

Turn the lens to it's infinity stop and look, if the RF patch coincides perfectly at infinity.

 

If it does, you're good, to continue. If it does not, you have the choice, to ship everything to a specialist or continue, to adjust your cameras rangefinder precisely after protocol (I will not go into that, there are many threads about this already and common sense dictates, to not fiddle with the camera's RF, when you are not knowing, what you are doing).

 

Now, that you know, that your lens' focus mount, RF coupling and your camera body's RF setting all fit perfectly at infinity, you can continue…

 

As the camera is already on a tripod, set to infinity and pointed at the horizon, why not check, if the lens' optical cell is actually collimated at infinity?

- take a remote release (or self timer), set the camera to RAW, base ISO, the lens to it's widest aperture, meter the horizon, set the camera to manual exposure and shoot a series of exposures:

 

- all exposures with the lens set to it's infinity stop (coinciding perfectly with your RF patch)

- shoot the lens at each aperture, correcting manually the shutter speed by a stop everytime

- load the pictures to your computer, and inspect each shot

 

Does the exposure @ f3.5 already give a reasonably sharp image at infinity? Or does it look, as the lens doesn't really focus at infinity (lens' shortcomings at this wide aperture aside).

 

The lens should be reasonably sharp already at f3.5 @ infinity.

If it is not, there are two possible causes:

- the optical cell of the lens is set at a wrong distance to the lens mount

- the lens' focal length of 135mm is not 135mm, but is deviating, resulting in focus being off.

 

These Canon LTM design lenses usually have two shims in their construction, to calibrate both variable.

 

Your lens is not sharp at infinity wide open (best case) - have the lens checked at a technician, he will adjust the optical cell's position regarding lens mount and possibly the focal length, if needed (I assume, your cameras RF is checked and verified as spot on).

 

Your lens is sharp at infinity wide open (best case) - proceed to check it's close up performance.

 

Now let's have a look at the lens' close focus performance:

- put camera + lens back on a tripod with remote release attached and same points, as in the infinity shooting test - you will take pictures now ;-)

- set up a second tripod/ light stand/ etc… with a close focus target (anything from a ruler, over a printed lens test target to a LensAlign Pro device does it here).

- set up the distance between focal plane and target's focus point to 1.5m

- shoot a series of test shots wide open (and stopped down, if you like), always refocussing the lens between shots (with or without magnifier - I will not go into depth here, why magnifiers suck as a focus improvement, as they introduce other variables of influencing focus negatively)

- measure the front or back focus, the lens gives you with this setup (inspecting the raw files on computer screen)

 

Now: the focus wide open should be spot on your target, if it is not, then:

 

- the camera's RF close up setting could be off

- the lens' focus mount internal RF cam is milled not properly

- the lens' optical cell is positioned at the wrong distance to flange AND it's focal length deviates from 135mm (having given you a sharp image wide open at infinity before)

 

So:

- check your camera first (technician or how does it perform with the same test with your other lenses?)

- if the camera's RF is out of the question, have a technician adjust the lens' shims for correct distance and focal length

 

For your interest, I am doing above procedures in one or the other way with EVERY lens and camera body, I buy - new or used.

With a RF camera system, you have to match your bodies and lenses and handpick them.

This is more critical, than with SLR lenses, where just the infinity collimation to the lens mount is critical (with old manual focus lenses with infinity stop, with newer lenses without mechanical infinity stop, this isn't even a critical tolerance anymore )

 

Nobody cares, if a SLR 135mm lens is in fact 135mm or 138mm - you get what you see!

This is the reason, why Leica engraves the precise focal length onto the finished lens barrel.

This is the reason, why all RF lens manufacturers are marking (mostly internally) one way or the other the optical cell and lens mount.

This is, why most of the time, Frankensteining a lenses optical cell into a different lens mount needs additional calibration, as both have to be matched.

 

My last project (still ongoing) is the perfect match of a rare, black Canon 85 1.8 LTM lens to Leica digital.

 

Out of nine Canon RF telephoto lenses (85 to 135mm), I bought or inspected over the last half year, none had matched the Leica RF precisely, most of them severely front or back focussing.

From the ones, I then continued, to disassemble and calibrate, none had showed any internal or external trackmarks of disassembly, repair or calibration, leading me to believe, they left the factory like this back then.

 

I have read on numerous occasions, that Canon had a slightly different approach regarding film bending and best used flange to film distance, possibly leading to a sightly different standard of lens adjustments, than Leica.

 

Also understand, that back in the 50s, there was no Digital M9 and no anal retentive approach regarding 1:1 pixel sharpness on 30" screens and the internet hype about sharpness, "focus shift" and alike, as it is now the norm.

 

Serious photographers back then simply picked their lenses, found about issues, using them, either ditched them, picking a different sample or simply learned workarounds, to use them (stop down/ lean forward, backward a certain amount at certain distances and apertures, etc …).

 

This could very well be an endless post, going into all the detail, one could wish.

but let me end with one very important last sentence:

 

A knowledgable, experienced user of any Leica M (M3 through M9) is very well capable of reliably focus a 135/3.5 lens wide open, when properly calibrated.

[zombii]

Advertisement (gone after registration)

Lots of good info Dirk. I'm not as exacting as you are but I know you're completely right about needing a good match between lens and body for ultra fast lenses. (A Summilux 50 is a fast lens but not ultra fast. Noctilux, Summilux 75, Canon 85/1.8 or Nikkor 85/1.5, Canon 100/2 and Elmarit 135's are all ultra fast.) A lens such as a Summilux 50 will mask a lot of inaccuracy due to its relatively large DOF. It fooled me until I bought a Summilux 75. I think a lot of the criticism of the Noctilux wide open is a result of unmatched lens to body.

 

I do think that magnifiers serve their purpose, especially for those of us with out of spec vision. After getting your lens matched to your body, though, you really have to practice. I wanted to think it wasn't going to be hard but I found out it sure isn't easy. Nevertheless, you improve if you stay with it.

[lct]

Menos, we've discussed this so many times here and on RFF about M8 and R-D1. The effective base length (EBL) of a rangefinder is based upon the formula b' = (e * f^2) / (k * z) where b' is the effective base length, e the visual acuity, f the focal length, k the aperture and z the circle of confusion. To focus a 135mm lens at f/3.5, the M8 would need an EBL of 69mm whereas it has only 47mm as is. No wonder why high hit rates are out of reach w/o a magnifier.

[lct]

Little chart of mine to summarize.

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!

[jaapv]

Ummm.. CoC on the M8 is 0.023 and it is wise to halve CoCs on a sensor...

Otoh the formula does not take vernier vision into account, which helps on the other side, so your table is probably correct

Edited August 9, 2011 by jaapv

[lct]

0.0300 for the M9 and 0.0226mm for the M8 (0.03 : 1.33).

[SJP]

Yep. That is how it works.

[menos I M6]

Oh god, there we are again … :-(

 

Anyway, thanks for the chart. I will print it out big and check next time, before being bothered, to pull out the camera, to shoot with long or fast lenses, as the EBL and CoC do not coincide with the moon phase and star pattern.

 

It is on this forum, that a strange aura exists, that tends, to overcomplicate things, making people regularly stuck inside endless threads calculating, theorizing, arguing and scientificating about theoretical technical potentialibilities of Leica M hardware, that makes my head spin and myself wondering, (or not) why some here must have very, very light skin, low shutter counts, little processed film in the archive and more keepers of test charts than actual photographs to print.

 

Seriously.

Get out and shoot.

Use your 135mm lenses with confidence wide open or not on your M cameras and enjoy …

 

… all else being nothing …

[lct]

It is very simple to me. Better use a magnifier with crop cams for anything below f/5.6.

[Telyt2003]

(...) Comparing the 135 Elmarit with say a Noctilux, (...) or even the much faster focussing APO-Telyt, I would go out on a limb and would say, one could handle this lens wide open, when matched precisely to a certain camera body.

 

The last part is the most important one.

It will certainly not work on any body, which leaves the factory with standard tolerances.

The lens has to be collimated, it's focal length verified (most certainly correct, given the meticulous optical work @ Leica), it's RF coupling has to be checked for any possible inconsistencies and finally, the camera body has to be checked for precise RF coupling from close focus to infinity.

 

I have been going through all of this with (... a 85 mm ...) LTM lens, which was out of the factory (...) not spot on adjusted to Today's Leica M9 tolerance (...).

 

It is possible, to shoot such thin dof lenses with the M. One has to know the limits though and must make sure, all mechanics are up to spec.

(...)

 

(...)

The point is: the M8/ M9 rangefinder is perfectly capable of reliably shooting fast, long lenses wide open.

Magnifiers do not necessarily improve focussing (more on that later).

 

Especially do they not help at all, when a certain lens is not properly collimated to the Leica M standard, it's optical cell out of spec, it's RF coupling erroneous, it's LTM-M adapter out of spec, the bodies RF out of alignment, the users eyesight out of spec or simply a lack of experience with the particular lens/ camera combination and it's weak spots and demands.

 

This is not a rant or an offense, but I see this argument written in big letters on the internet a lot, more so, since many new users of the M system are not fully educated yet about the technical boundaries of the M and it's demands on mechanical precision on all involved components (+ the probably most important factor: "Mensch" with the fully manual camera system).

 

(...) non Leica LTM lens (...), which was not set up properly for the Leica M system out of the manufacturers factory, but misfocussed badly on calibrated M bodies.

(...) Prepare a well maintained vernier caliper, a tripod, cable release, a close focus target, a suitable infinity target with good atmospheric condition, several LTM-M adapters of different makes, a computer with card reader and some patience and you will be rewarded with a wonderful (... 135 mm ...) LTM lens.

(...)

 

(...)

For your interest, I am doing above procedures in one or the other way with EVERY lens and camera body, I buy - new or used.

With a RF camera system, you have to match your bodies and lenses and handpick them.

This is more critical, than with SLR lenses, (...)

 

Also understand, that back in the 50s, there was no Digital M9 and no (...) approach regarding 1:1 pixel sharpness on 30" screens and the internet hype about sharpness, "focus shift" and alike, as it is now the norm.

 

Serious photographers back then simply picked their lenses, found about issues, using them, either ditched them, picking a different sample or simply learned workarounds, to use them (stop down/ lean forward, backward a certain amount at certain distances and apertures, etc …).

(...)

A knowledgable, experienced user of any Leica M (M3 through M9) is very well capable of reliably focus a 135/3.5 lens wide open, when properly calibrated.

 

Dirk,

thank you very much for communicating your experiences with fitting fast AND long lenses to (digital) M-cameras!

I could not have said it much different (as cited above), or any better ...

... and it matches my own experiences quite exactly.

 

I cannot remember that the necessities of focusing widely opening and highly corrected (long) lenses on the (digital) M-cameras have been described very often that completely (and openly) .

 

My own approach to avoid (major) misfocusing of fast (not super-fast) lenses, so far, was less exact (than in your latest, very long post), and finally mainly consisted in avoiding to buy out-of-spec (or non-fitting) M(8)-cameras and M-lenses (as described here: M8 and Apo-Telyt 135/3,4 (and the correction in post 21), M8 und Backfokus, in German) by vigorous testing beforehand:

This testing normally takes hours, with checking all the apertures at close, medium and far (infinite) distances, both with my own (trusted) lenses (in parallel to a candidate lens), and my own (known, but often not trusted) cameras, and when in doubt with additional cameras of the shop (M8.2 and M9).

I know that such a strategy, at best, will work with fast lenses, such as the Summilux 50/1,4, Summicron 75/2, Summarit 90/2,5 (Summicron 90/2 ?), and the Apo-Telyt 135/3,4 (and Tele-Elmar 135/4). For me, it did. And I refused quite a few lenses and cameras.

 

What you describe in addition, should be probably strongly recommended to fit any of the super-fast lenses (Noctilux 50, Summilux 75, Summicron 90, Elmarit 135 and your Canon LTMs) to any (digital) M-body.

However, for me, already all attempts to have the cameras calibrated did not work. Thus, I am currently not considering any super-fast lenses (albeit a Summicron 90/2, or the Canon 100/2 which you recommended, would still really tempt me).

 

I had no more choice after all attempts to have my (early) M8-cameras EXACTLY adjusted, or fitted exactly with my best fast (not super-fast) lenses - or to any used or even brand-new candidate lenses - had failed (again and again).

In the "early years of the M8" (2006 and 2007), users with focus issues (in my case: exactly documented backfocus which normally was of greater extent than the depth-of-field with all lenses widely opened) apparently were not taken very seriously (by - almost - nobody).

Things started to change only with the release of the M8.2 (and probably the latest M8 batches).

Finally, I decided to just take those of my fast lenses which I trusted in, and to find a digital M-body fitting as good as possible. And to re-learn the focussing - as done before with several M8s - with such a more manageable set-up (i.e., re-learn the still necessary minor focus compensations). At least, with this strategy, I have to apply much smaller (and more uniform) corrections to the focusing with all my critical (and selected) fast lenses - than before with any non-picked M8.

 

I still have one very good, modern lens sitting on my desk as a paper weight. Since the last attempt to have it serviced and calibrated for the M8/9, it focuses so badly that even a very experienced focus-compensator (like me) has a very low hit rate with it (Before this service, it was a very good match to my very exactly focusing Apo-Telyt 135/3,4!).

My Apo-Telyt 135/3,4 is very exactly matching my current late M8 (and different M9s tested so far), especially at f/3,4 (fully open). However, stopping it down even a little bit (f/4, and also at f/5,6) obligatorily requires the compensation of a tiny bit of focus shift ("Blendendifferenz"), because otherwise the focus could already be missed due to a small amount of backfokus. I am just mentioning this, because I normally would clearly miss the (critical) focus, at least at f/4, if not compensating this backfocus shift.

BTW, I mostly use the 1,4x magnifier, and with it my focusing becomes more accurate (at least in low light), albeit focusing is also possible without a magnifier. I am obviously looking through the magnifiers in the correct way (non-oblique, centred) due to vast experience with oculars.

 

Thanks again,

and with best regards,

Telyt2003

[maccaco]

I just received my Emar 135/4 and works fine on the M8 ... I focus pretty well! :D:D

[Telyt2003]

I just received my Emar 135/4 and works fine on the M8 ... I focus pretty well! :D

maccaco,

me too, I like the Elmar 135/4 (from the 1960s) very much on the M8.

It is nearly at its optimum from aperture 4 onwards (fully opened).

It is relatively light-weight and compact.

The rendering of the unsharp background is especially pleasing at the aperture of 5,6.

An example image:

http://www.l-camera-forum.com/leica-forum/nature-wildlife/191596-grab-frisbee.html

 

Since I am lacking a well adjusted long (rangefinder-coupled) focusing mount for the Elmar 135/4 , I am currently using only its head as a (very small) 'short focal length' addition to the Visoflex systems, for example with a short (uncoupled) focusing mount on the Visoflex 2/3.

 

The short focusing mount which was specifically offered for the Elmar 135/4 head (14071 (J)) apparently is quite rare.

Instead some -not all- versions of the (older) short focusing mounts for the head of the Hektor 135/4,5 (ZOOAN / 16495 () can also be used with the Elmar's head - as long as no internal 'rings' against stray light physically prevent screwing it in (in some versions of the ZOOAN there are less of these rings making it compatible with also the Elmar).

With both short focusing mounts (made for Visoflex 1), in addition the adapter OUBIO / 16466 (M) is needed to use them with the Visoflex2/3. The OUBIO also provides a tripod socket (and a mechanism to rotate from horizontal to vertical format): very practical, if a shoulder support is an option for stabilisation.

 

Best wishes,

Elmar/Telyt2003