Back Focus???

[wilfredo]

Advertisement (gone after registration)

This shot was taken with a 50mm lens on an M8. I focused on the black dots in the middle of the eyes. The back hair seems to be in crisper focus than the eyes (see the spider web). Is this a back focus issue?

[Advertisement]

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

[wilfredo]

BTW This was shot @f/2

[wilfredo]

Here's another shot @f/2 but this time it seems the focus is better (right eye). Is it possible that an M8 can back focus sometimes but not all the time? I'm using a loaner from Leica. I was really careful focusing the first shot, and I even had it on a tripod, but perhaps I did not focus as well as I thought, but it sure did look crisp in the viewfinder with everything properly aligned.

[JWW]

50 F2 should have a wide enough DOF that focus should be fairly easy. I recommend taking a picture of a tape rule on the floor and see how far off the clearest focus is off from the aimpoint.

[lct]

50 F2 should have a wide enough DOF that focus should be fairly easy...

Not for close-ups. DoF = 4 centimeters at 1 meter & f/2.

[wda]

If the model is small, your result is within normal operating error limits. DOF is extremely limited at f/2.

[christer]

Advertisement (gone after registration)

It is quite difficult to establish if your lens/camera combination back or front focuses.

 

First of all you need to shoot with the lens fully open, camera on a tripod, and TMY or better film in the camera and with a camera to subject distance of no more than 150 cm. If you do not know why these requirements have to be met, then it may be that you yourself have a problem rather than the camera/lens combination. ;-))

 

Make several shots and refocus between each. Look at 16x enlargements with a 6x loupe and then you will know.

 

The best test object is a sheet of white paper with parallel horizontal lines 1,4 cm apart. There should be a break in the lines, roughly in the middle. Number the lines in the break, starting with 0 roughly half way down the page and number both upwards and downwards starting with the 0 in both cases. The numbers should be about 10mm high. If you use the in-camera light meter, then you should overexpose by two stops if you take the reading from the white sheet of paper.

 

The paper should be kept at an angle of 45°. That's why the distance between the lines is 1,4 cm. Looking at the final print you will be able to determine which number is sharpest. The number corresponds to the number of centimeters of front or back focus - provided you focused on the 0.

 

Do not worry too much about one or two centimeters back or front focus. Other factors will be more relevant for degrading technical picture quality.

 

I have no experience with digital, but the principle should be the same.

[Jamie Roberts]

Here's another shot @f/2 but this time it seems the focus is better (right eye). Is it possible that an M8 can back focus sometimes but not all the time? I'm using a loaner from Leica. I was really careful focusing the first shot, and I even had it on a tripod, but perhaps I did not focus as well as I thought, but it sure did look crisp in the viewfinder with everything properly aligned.

 

Wilfredo, if it's the same distance away, it's totally possible that your eyes misinterpreted the contrast on one shot but not the other... the dots look pretty small to me.

 

If they were at different distances (and you cropped the files to show something) then yes, it's totally possible to have back or front focus close up and ok ness by infinity, with a whole range of stuff in-between

 

I'd check depth of focus with a ruler (or book spines arranged "closer to farther" or something) to see.

[wilfredo]

Wilfredo, if it's the same distance away, it's totally possible that your eyes misinterpreted the contrast on one shot but not the other... the dots look pretty small to me.

 

If they were at different distances (and you cropped the files to show something) then yes, it's totally possible to have back or front focus close up and ok ness by infinity, with a whole range of stuff in-between

 

I'd check depth of focus with a ruler (or book spines arranged "closer to farther" or something) to see.

 

Thank you all for the responses and tips.

 

Jamie,

 

I will give your suggestion a shot (no punt intended). This M8 is a loaner from Leica, it's got issues (*) and I'm waiting for a replacement M8. I'll do the testing again as you and others have suggested once I get my replacement camera.

 

* Sometimes the auto metering underexposes by as much as 2 stops and the other day it died on me, but then started working again after I put the battery in the charger to see if was dead, and it wasn't and for some reason that got it working again. So it could be that this "Loaner" camera itself back focuses.

[Guest mc_k]

... Is it possible that an M8 can back focus sometimes but not all the time?...

 

it's probably just variation (user error), but a rangefinder or lens with backlash can have intermittent problems.

[adan]

One thing I've noticed in all M's going back to my first M4-2 nine years ago (and as recently as last night) is that I have more focus issues shooting verticals. I'm sure it is something extraneous to the focus mechanism itself - how the camera lies against my face, eye centering, or other such factors.

 

So much so that I always and by habit turn the camera horizontally to focus and then back to vertical for the shot.

 

There is also the problem of focus-and-recompose. If you focus on something (which is thus in the center of the image where the RF patch is) and then re-aim (i.e. tilt the camera for verticals or swing it sideways for horizontals) to put that point offcenter in the image - it will also now be a different distance from the lens (and thus back-focused).

 

From Wilfredo's shots, looks like either of these effects might have been involved.

[christer]

.............

 

There is also the problem of focus-and-recompose. If you focus on something (which is thus in the center of the image where the RF patch is) and then re-aim (i.e. tilt the camera for verticals or swing it sideways for horizontals) to put that point offcenter in the image - it will also now be a different distance from the lens (and thus back-focused).

.........

 

I do not think it is that way.

 

Imagine you are standing right in front of a building with a door in the middle and three windows on each side. You are right opposite the door and focus on the door. The door and the six windows will be rendered sharp on the print (unless there are other problems).

 

You want to make sure the right most window is sharp on your print, but still want the door to be in the middle of the picture. So you focus on that window and reframe. Sorry, but that is the wrong approach. Everything will be a bit out of focus. That is because most modern lenses have a flat field of view, not a part circle field of view.