Focus calibration for IIIf/IIIg and LTM lenses?

[ironringer]

Advertisement (gone after registration)

Please excuse my "basic questions" as I continue to learn about these usable classics.

Examining (on a large PC screen) the scans of a roll of ISO100 colour print film shot with my IIIf and Summaron 3.5 cm, and not seeing any really sharp scans, I conclude that (a) if you can't see the grain of the film in the scans, and ( if the scans (of properly exposed, carefully focused, and stable-platform pictures) do not appear sharp at least in center, then it is possible that © the scanner was not well-focused (quite possible in a "public" photo lab), and/or (d) the rangefinder-indicated focus may not match the actual focus plane of the lens (i.e. the original picture was not in sharp focus). This assumes that the lens itself is clean and sharp, and that was verified by earlier tests at all apertures.

I also understand from replies to my "suggest a 9 cm lens?" thread, that longer lenses, with smaller depth of field, should be used at smaller apertures to have a better chance of getting in-focus pictures.

So ... my first question is, how are rangefinders in screwmount cameras "adjusted" to closely match the real focus plane of a lens, through its distance range? Is this adjustment a process that most users can perform on a IIIf or IIIg?

And another question - should a rangefinder camera (ideally) be adjusted for each screwmount lens, or is a rangefinder adjustment for one lens (which lens is best?) quite suitable for all lenses, at least within the Leica LTM family?

I am beginning to think that one should perform focus tests on longer lenses, shooting a roll of film at distance-marked targets and at different rangefinder-indicated distances, then review results to prepare a compensation table for, for example, a 9 cm lens used at wide apertures.

Information and suggestions are requested.

[jaapv]

Except for the very first Leicas with individually calibrated lenses, Leica lenses and cameras have always been adjusted to a common standard to ensure interchangeability of the system Since they have built-in rangefinders, for sure.

[Advertisement]

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

[luigi bertolotti]

I also understand from replies to my "suggest a 9 cm lens?" thread, that longer lenses, with smaller depth of field, should be used at smaller apertures to have a better chance of getting in-focus pictures.

 

This, generally speaking, is true for any lens, for the basic principle of depth-of-focus, which increases stopping down the diaphragm; the longer the focal, the more critical the focus, expecially at short distances, so that increasing depth of focus closing the diaphragm is more useful to avoid errors (and to compensate for slight imprecisions of the rangefinder mechanism)

 

So ... my first question is, how are rangefinders in screwmount cameras "adjusted" to closely match the real focus plane of a lens, through its distance range? Is this adjustment a process that most users can perform on a IIIf or IIIg?

 

The adjustment of the Leica rangefinder is a delicate operation, which imho a customer is better not to engage into.

 

And another question - should a rangefinder camera (ideally) be adjusted for each screwmount lens, or is a rangefinder adjustment for one lens (which lens is best?) quite suitable for all lenses, at least within the Leica LTM family?

 

(see above answer from Jaap)

 

 

I am beginning to think that one should perform focus tests on longer lenses, shooting a roll of film at distance-marked targets and at different rangefinder-indicated distances, then review results to prepare a compensation table for, for example, a 9 cm lens used at wide apertures.

 

Focus test is a good operation to be performed by user... it helps to understand well how complex is an apparently easy concept as "subject at x meters is in focus"... ; also useful for it requires to setup a "test environment" not complicated, but in which some details MUST be correctly tuned (position of the camera, of the target, alignement with lens axis etc...

BUT : good for verifying your own equipment , and, in case, have it adjusted by a lab... not for buliding complex tables to avoid an adjustement that is not too costly to make.

 

Information and suggestions are requested.

.

Edited September 12, 2014 by luigi bertolotti

[pico]

Ironringer: Begin with the simplest test to rule out the least difficult and expensive factors. Try this: First get an 8X loupe and examine some of the frames. Image sharpness should be apparent. It takes a while to learn.

 

Next, choose a frame to sacrifice. With the sharp edge of a pin, gently draw an X across the emulsion (non shiny) side of a frame and scan it. The X should be sharp and clear corner to corner, and center. If it is not, then try to find a better scanner or have a roll scanned by a shop. If it is the scanner at fault, you just saved a lot of money on a Leica repair.

 

If you like, you can post a sample frame for us to look at, or mail a frame to me and I will do it for you from my network.

 

Then get back to us. We love screw-mount Leicas and enjoy others' success.

 

Best,

Pico

[250swb]

Please excuse my "basic questions" as I continue to learn about these usable classics.

Examining (on a large PC screen) the scans of a roll of ISO100 colour print film shot with my IIIf and Summaron 3.5 cm, and not seeing any really sharp scans, I conclude that (a) if you can't see the grain of the film in the scans, and ( if the scans (of properly exposed, carefully focused, and stable-platform pictures) do not appear sharp at least in center, then it is possible that © the scanner was not well-focused (quite possible in a "public" photo lab), and/or (d) the rangefinder-indicated focus may not match the actual focus plane of the lens (i.e. the original picture was not in sharp focus). This assumes that the lens itself is clean and sharp, and that was verified by earlier tests at all apertures.

 

Have you answered the first question first? Which is, have the pictures been sharpened yet?

 

To get a sharp image from scanned film two things need to come into play, the image file has to be large enough to hold enough detail, and it needs sharpening. This is before judgements about scanners or lenses can be made.

 

Just as digital images can be sharpened in the camera, or later in post processing, a scanned image should be sharpened at some stage in the process. And if it is a lab scan, and anything other than a small 'thumbnail' file, the image should not be already sharpened because the lab doesn't know what you are going to do with it, for instance what size you want to make it for a print, neither do they know your preferences for the amount of sharpening. So, have they perhaps only sharpened it a nominal amount, or have you sharpened it at all, just so people aren't trying to solve the wrong problem?

 

Steve

[pico]

Have you answered the first question first? Which is, have the pictures been sharpened yet?

 

Don't you think directly inspecting the film under magnification is the first thing to do?

[gyoung]

Advertisement (gone after registration)

Surely if the neg isn't sharp, or the scan isn't sharp then no amount of sharpening in post processing will help. Cart before horse IMHO

 

Gerry

[ironringer]

Thank you members for several good comments. If/when I find my light table, packed away over 20 years ago in some unknown box in the basement, I can better examine the negatives with my 8X loupe.

But to identify the "scan quality/focus" parameter, today I commissioned test scans of a few negatives at a proper photo lab. As the 100%-crop pictures (of a small area at the edge of image) below show (for some reason they do not appear at the same physical size), the initial scan at right (by a Western Canadian drugstore that has a good photo department and lab in-house), was not as well focused as today's better scan at left (by Montreal's Camtec, with a Noritsu scanner). As I suspected, if you don't see the film grain, the scan is not in focus. Same as the "scratch an X on the negative, then scan" test.

Other factors that made my images less-sharp off-center are still active:

- older lenses like my 1953 Summaron have less resolution off center

- at f4, edge resolution and contrast are limited compared to performance at smaller apertures

- possible difference between rangefinder-indicated focus and actual focus plane

I assume that the Summaron's depth-of-field scale is accurate, so that when I focused at about 50 feet, at f4, trees at infinity should (in theory) be in focus.

While it would be difficult to perform focus tests (targets, etc.) for a wide-angle 3.5 cm lens (large depth of field covers up errors), I may try a few test pictures with my next roll, to try to identify any differences between indicated and actual focus. Of course those focus tests would be easier (to see differences) with a 9 cm lens.

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!

[250swb]

This is the drug store scan on the right, if I read your description correctly, with a touch more contrast and sharpened a little bit. Given working on a tiny JPEG isn't ideal the result, to me, looks pretty close to the 'pro lab' scan? This may indicate the pro lab scan has been sharpened already, hence you thinking it is a 'better' scan. It's just an idea, but give the horse a carrot.

 

Steve

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!

Edited September 13, 2014 by 250swb

[giordano]

- possible difference between rangefinder-indicated focus and actual focus plane

 

Always a possibility, but the rangefinders on screw-mount Leicas are very rugged, so exclude other factors first. (Also, rangefinder adjustment on these cameras is complicated by the fact that you can't open the back to fit a focusing screen so you can see the image from the lens the way you can on an M.)

 

I assume that the Summaron's depth-of-field scale is accurate, so that when I focused at about 50 feet, at f4, trees at infinity should (in theory) be in focus.

 

The scale is accurate, but it's calibrated for the traditional 1/30 mm circle of confusion - i.e. on the assumption that you'll be making postcard-size prints from the sorts of film available to Leica users in the 1930s. With modern emulsions and larger prints, sharpness at the extremes of the indicated DOF range may not be satisfactory. Many people compensate for this, where sharpness is important, by using an aperture two stops smaller than suggested by the DOF scale.

 

 

While it would be difficult to perform focus tests (targets, etc.) for a wide-angle 3.5 cm lens (large depth of field covers up errors), I may try a few test pictures with my next roll, to try to identify any differences between indicated and actual focus. Of course those focus tests would be easier (to see differences) with a 9 cm lens.

[ironringer]

Thank you Steve for the adjustment on the drugstore scan. My preference is based on the visible grain in the Montreal photo lab scan, that I don't believe can be realistically "created" by computer adjustment, so I believe it is better than the drugstore scan. There is more detail in the in-focus scan, and I don't think it is created by computer sharpening.

But I see how computer processing can be a useful tool to improve even a fuzzy scan.

[TomB_tx]

There really is no such thing as a "100% crop" on a film image, because it has no native size other than the 24x36 mm negative size. Once it is scanned, a 100% crop just crops to the resolution of the scan. Film to that crop never looks as sharp as digital. In film days the job of the photographer was to capture the image so as little crop as possible was necessary. If you enlarged an image to 48x72 inches, it was either being projected as a slide or a huge print; and either was looked at from some feet away. Your image would look fine when viewed that way.

I often shoot both M9 and film Leica, and mix the images projected for lectures. You can tell the difference, but most viewers don't notice when looking at the full image.

But I do crop more with digital, and allow for that difference in shooting. On film I use 50mm most, but with the M9 I use 35 more, as when needed I can crop to find the "50mm" view with little loss of quality.

[250swb]

There is more detail in the in-focus scan, and I don't think it is created by computer sharpening.

 

A scan is a computer generated image with all the normal adjustments of contrast, brightness, colour, and sharpening. All scanning software has a sharpening tool and it is used more or less depending on the skill of the operator or setup of the machine. And don't kid yourself all scans apart from very high end drum scans will have had some sort of computer sharpening done to them. Indeed it could be argued your lab scan has been over sharpened.

 

Steve