Focusing on M9/M240

[xardas]

Advertisement (gone after registration)

I understand that manual focusing on the Leica cameras M is done by using this small square in the middle of the screen. My question is, can you move the square off-center?

 

I mean what if i want to focus on something that is not exactly in the center of the frame ? Focus and recomposing is out of the question especially if the subject is moving (like people on the street)

[TomB_tx]

Focus and recompose is hardly out of the question, as people have been doing this with the Leica M type rangefinder since the M3 of 1954. Because you see outside the framelines on the viewfinder it just takes an instant to shift and recompose. In the 1960s I even shot football and hockey with M2 and M4 Leica with good results because of the view outside the framelines. The viewfinder basically shows all that a 28mm lens would see, but with framelines within that view showing what the attached lens will actually record. I prefer it to an SLR for action, for those times a 135mm or shorter lens will do.

Yes, it takes getting used to; and depending on the lens recomposing isn't as theoretically exact as full-field focus, but I've aways been satisfied with the results.

[Jeff S]

Focus and recomposing is out of the question especially if the subject is moving (like people on the street)

 

A few people in the history of photography have gotten by using an M for people on the street.

 

Lots of techniques...pre-focus, zone focus, quick focus tracking, etc...are possible. Of course practice is key, and attention to lens choice, DOF and distance to subject, etc. helps as well. The RF is particularly effective for shorter focal length lenses.

 

An M isn't an ideal sports camera, but some people still use it effectively for that purpose.

 

Jeff

[A miller]

A few people in the history of photography have gotten by using an M for people on the street.

 

Lots of techniques...pre-focus, zone focus, quick focus tracking, etc...are possible. Of course practice is key, and attention to lens choice, DOF and distance to subject, etc. helps as well. The RF is particularly effective for shorter focal length lenses.

 

An M isn't an ideal sports camera, but some people still use it effectively for that purpose.

 

Jeff

 

On a sunny day with a focal length of 35mm or 28mm and hyperfocal zone focusing, i find the M to be outstanding for action sports.

 

Here is one of my kids playing flag football taken with my M9 and 35mm summilux asph fle

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!

[A miller]

One more...

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!

[Giorgio Festa]

One more...

 

Great pics!

[Stealth3kpl]

Advertisement (gone after registration)

I mean what if i want to focus on something that is not exactly in the center of the frame ? Focus and recomposing is out of the question especially if the subject is moving (like people on the street)

 

A little practice is all that's needed.

Pete

 

[Your Old Dog]

Your question is very basic so I will volunteer this. If you are using a 50mm lens wide open at say f1/4 and your subject is fairly close to you, you will have a hard time focusing on moving action. But, you can frequently go to f5.6 or f8 and take the same shot as "depth of field" is working for you in this instance.

[Lindolfi]

Another point is that the error in focus you get when focussing with the center and you want the object in focus to be at the edge of the frame, depends on

 

1) focal length

2) aperture

3) distance to object

 

Sorry for the length of the post. If you don't want to read the rest, the summary is that for 21, 35, 50 and 75 mm lenses you have to move your camera away from the object when focussing at closest range and full aperture by 157, 57, 27 and 11 mm respectively

 

I've calculated what happens with the focus error at several focal lengths [12, 21, 35, 50, 75, 90, 135] mm and apertures [5.6, 2.8, 2.0, 1.4, 1.4, 4.0, 3.4] and at various object distances [700, 700, 700, 700, 700, 800, 1600] mm and I've assumed you rotate the camera around a point 300 mm behind the camera (based on standing posture and turning your head together with camera). Camera is in landscape orientation.

 

It appears that the following focus errors occur: [437, 157, 57, 27, 11, 9, 10] mm

 

Those errors are only inside the DOF (with a circle of confusion of 0.03 mm) for the 12 mm, the 90 mm and the 135 mm lens. The other lenses create out of focus images of objects at the short edge of the frame on which you focussed in the center of the frame.

 

To get sharp images with the other lenses, you have to move the camera back by 157 mm for the 21 mm lens, 57 mm for the 35 mm lens, 27 mm for the 50 mm lens and 11 mm for the 75 mm lens. At further focus distances, the necessary amount of movement of the camera backward is less and the error quickly falls inside DOF so that it is not necessary anymore and you can just turn your head and camera to recompose. When de object that needs to be in focus moves to a position between the middle and the edge of the frame, the necessary correction backward is half of what I indicated.

 

As you can see the most problematic lenses are the 21 mm lens and the 35 mm lens. Moving back is a matter of learning. From my experience it can be done, so you end up inside the DOF of the lens, but it is a bit of a black art , nicely demonstrated by Pete in this thread.

 

With the M (typ 240) using the EVF, it remains a black art, since you cannot move the focus patch outside the center. This is usually no problem since M photographers have learned to do it with the rangefinder for a long time, but for instance with a Tilt-Shift lens it is really a problem and you resort to focus without zooming in the EVF or trial and error.

[k-hawinkler]

Bert,

 

Thanks. May I ask which algorithm you used?

[FlashGordonPhotography]

Bert, I'm not sure whether that information will hold true for all lenses in the focal lengths you mention. Lens design can alter a few things. One of them is the plane of focus. Many lenses focus on a curved plane while some (most noticably macro lenses) focus on a flat plane. And then theres the many versions of how curved that plane of focus may be. It's quite possible that lenses with the same focal length may have very different curvature of the plane of focus (90mm 2.0 vs 2.5 vs 4.0 for example). So you may need more or less adjustment than the plain DOF figures might indicate depending on the lens design.

 

Generally I gotten a good enough "feel' for the lenses I use often to be able to fairly consistantly make an accurate adjustment for focus and recompose. And after a couple of years I've actually gotten quite quick at it as well.

 

Gordon

[Lindolfi]

Bert,

 

Thanks. May I ask which algorithm you used?

 

Here is the figure to explain the principle

 

 

And here is the formula derived from geometry

 

http://www.photoplaza.nl/lindolfi/focusrecomposeerror.pdf

 

 

Gordon, you are right, focal plane curvature is a factor that influences the results. If we have the data, it can be added to the algorithm.

[Guest borge]

It's very easy: Focus on your subject and re-compose the picture as you want it.

This is quicker on AF cameras with selectable focus points as well. Use the center focus point to focus and recompose and click, rather than fiddling around with focus point selection buttons and stuff like that.

[Thomas Chen]

Here is the figure to explain the principle

 

http://www.photoplaza.nl/lindolfi/focusrecomposespineprinciple2.png

 

And here is the formula derived from geometry

 

http://www.photoplaza.nl/lindolfi/focusrecomposeerror.pdf

 

.

 

Bert,

 

May I ask you a question about focusing by LV?

 

In case of setting M 21/2.8 at f=16, the light goes through the aperture should be too dim to make the image clear enough for focusing on the rear LCD, no matter in 1X, 5X, and 10X .

 

Then that's the reason why we need the EVF as it features the brightness compensation?

[Lindolfi]

In case of setting M 21/2.8 at f=16, the light goes through the aperture should be too dim to make the image clear enough for focusing on the rear LCD, no matter in 1X, 5X, and 10X .

 

No, not true: LiveView (LV) on the LCD and through the EVF are both adjusted in brightness so that the aperture is not important, the same way the actual exposure is adjusted. On top of that a sensor on the camera is used to adjust the brightness of the LCD and the EVF depending on the environmental light. This second brightness adjustment can be set independently in the menu of the "Monitor" and "EVF".

 

There are several other advantages of the EVF over the LCD for focussing: better shielding of the environmental light and mechanical coupling of the camera to your head, making it more stable, while the increased resolution is also nice.

 

Then that's the reason why we need the EVF as it features the brightness compensation?

 

So no, not true either.

[Lindolfi]

Bert, I'm not sure whether that information will hold true for all lenses in the focal lengths you mention. Lens design can alter a few things. One of them is the plane of focus. Many lenses focus on a curved plane while some (most noticably macro lenses) focus on a flat plane.

Gordon

 

Just tested the Summicron 35/2.0 ASPH at a distance of 0.8 meter and at f/2.0. When focussing on the center of the image and after that moving the camera in 5 mm steps forward and backward (up to 50 mm in both ways), the edges of the image did not improve, but became worse, indicating that this lens has no field curvature, within even a critical measure of DOF (CoC = 0.015). So for that lens the recipe of backward movement I gave holds.

 

I'll test other lenses later. In principle other members can do the same. I used a glass frame with a sheet of paper with very fine printed text and placed the camera with the optical axis at right angles to the glass. Camera was mounted on a macro slide with distance scale, for moving the camera forward and backward.