Stupid question about backfocus vs frontfocus

[rwfreund]

Advertisement (gone after registration)

Hmmmm...

If these focus issues are caused by error, then shouldn't the distribution of front-focus and back-focus be about equal?

If no is it because folks don' notice front focus as much unless they are looking for it or is it because there are systemic errors that favor one sort over the other, or is it because folks tend to approach focus by turning the focuc ring in one direction over the other and there is some hysteresis?

just curious

-bob

[Advertisement]

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

[lars_bergquist]

Backfocusing is caused by spherical aberration, i.e. rays from the periphery of a lens have a shorter focus than axial rays. Actual best focus then is a compromise. So when stopping down and cutting off those peripheral rays, focus moves further back along the optical axis. Obviously, the phenomenon is in principle always there, but it is more noticeable the more spherical you have, and the faster/wider the lens is; and fast lenses are more difficult to correct for spherical aberration (and the associated problem, coma).

 

Clearly, frontfocusing cannot have this cause! If it happens, it is normally caused by an error in the focusing mechanism. It may even be deliberate, in order to compensate for backfocus. It seems that Zeiss do this for certain fast lenses.

 

BTW there is a different term 'back focus' ('Schnittweite' in German) which means the free distance from the apex of the rear lens element to the film or sensor plane. This has to be quite long in SLR cameras in order to make room for the mirror, while that of the M cameras is very short. There have been early super wide lenses which nearly touched the shutter curtains.

 

The old man from the Age of the Hologon

[1234]

du Lars, what a bunch of baloney. It was a reasonable question. I have sent back three M8's for focus problems; two back focus and one front focus. It is strictly a mechanical problem, a nuts and bolts issue solely.

[rwfreund]

I refer to the use of the term "backfocus" and "frontfocus" as so frequently used in this forum to mean the situation where there is a lack of coincidence between the plane that appears to be in focus in the rangefinder and the plane that appears to be in focus as recorded by the sensor.

 

In the case of backfocus, the actual plane of focus is further from the camera than the rangefinder indicated plane, and in frontfocus, the other side around.

 

Yes, I agree that I am simplifying, and that the "plane of focus" is indeed a rather complicated shape (really more complicated than spherical, actually) which varies both with focus distance as well as with aperture.

 

But I know all that. What I am interested in knowing is why folks so much more frequently complain about backfocus problems than frontfocus problems.

 

cheers

-bob

 

Old man from the age of borax accelerators (Lars' copyright infringement, probably)

[Jamie Roberts]

{snipped}

But I know all that. What I am interested in knowing is why folks so much more frequently complain about backfocus problems than frontfocus problems.

{snipped}

 

Perhaps because a small number of M8s were sent out mis-aligned? That's what would make sense to me.

 

The other thing that makes sense, and that is under-reported, probably, is that a large number of very old lenses are now being used when they weren't before.

 

I bought what looked to be a perfect, mint minus 75 Lux (10 years old, though), but it appeared to front focus. I took it into the repair shop here in Toronto (Kindermann Canada) and sure enough, the technician said the lens was out of alignment (couldn't acheive infinity focus). Once fixed, all was well again.

 

(FWIW, I think Lars is describing "focus shift" with spherical lenses, not a backfocus error in this sense. But I may be wrong on that too).

[dlc43]

Another idea: Generally, when an object is in focus, what's actually in focus ranges within a zone where about one third of the zone in focus resides IN FRONT of the object, and about two thirds of the zone in focus resides BEHIND the object. Thus, focus errors are more forgiving if you as the operator mistakenly front focus: the object has a good chance of being captured in focus anyhow, by a factor of about two to one. In the case where the camera/lens combination back focuses, it is a less forgiving situation, and more notceable, as the object is simply less likely to be in focus. Thus, more dissatisfaction and complaints.

Does this make sense?

[jaa]

Advertisement (gone after registration)

It makes perfect sense to me.

[blakley]

The terminology here is quite confusing. I've seen "focus shift" used to refer both to

 

1. the phenomenon which causes focus to recede further from the point at which the rangefinder indicates correct focus as the lens is stopped down

 

and

 

2. the phenomenon in which a non-apochromatic lens correctly focused for one wavelength (say visible green) does not focus another wavelength (say infrared) at the same plane

 

I've seen "backfocus" used to refer to

 

3. phenomenon 1 above

 

and

 

4. Lars' description earlier in the thread; i.e. the phenomenon in which a correctly adjusted lens, for some of its focusing range, focuses on a point further away than the point at which the rangefinder indicates correct focus - whereas at other distances (e.g. closest focus or infinity) the point at which the rangefinder indicates correct focus is in fact in focus on the imaging plane.

 

and

 

5. the phenomenon in which a mechanically maladjusted lens focuses on a point further away than the point at which the rangefinder indicates correct focus

 

I've only seen "front-focus" used to refer to

 

6. the phenomenon in which a mechanically maladjusted lens focuses on a point closer than the point at which the rangefinder indicates correct focus.

 

I sometimes observe "apparent" backfocus (sense 5) and frontfocus (sense 6) in my photos, but I got out a yardstick and tested all my lenses yesterday, and I discovered that there is another meaning for backfocus and frontfocus, namely:

 

7. the phenomenon in which a photographer's eyes are not good enough to use the rangefinder to place the point of focus where he wants it.

 

This is the phenomenon I'm actually seeing in all my pictures, except that my 50/1.2 "ASPH" Noctilux does exhibit actual front-focus at the near end of its focal range. I'm guessing it was adjusted this way deliberately in order to offset "focus-shift 1 = backfocus 3" at smaller apertures.

[Jamie Roberts]

{snipped}

I sometimes observe "apparent" backfocus (sense 5) and frontfocus (sense 6) in my photos, but I got out a yardstick and tested all my lenses yesterday, and I discovered that there is another meaning for backfocus and frontfocus, namely:

 

7. the phenomenon in which a photographer's eyes are not good enough to use the rangefinder to place the point of focus where he wants it.

 

This is the phenomenon I'm actually seeing in all my pictures, except that my 50/1.2 "ASPH" Noctilux does exhibit actual front-focus at the near end of its focal range. I'm guessing it was adjusted this way deliberately in order to offset "focus-shift 1 = backfocus 3" at smaller apertures.

 

Bill--do you have the 1.25x magnifier? I wouldn't be without it for the longer lenses! (edited to make sense! Thanks Jack!)

[blakley]

Bill--do you have the 1.25x magnifier? I wouldn't be without it for the longer lenses! (edited to make sense! Thanks Jack!)

 

I'm "Bob"; I do have the 1.25 magnifier, but I'm deteriorating faster than my equipment is improving :-)

[lars_bergquist]

-------------But I know all that. What I am interested in knowing is why folks so much more frequently complain about backfocus problems than frontfocus problems.

 

cheers

-bob

 

Old man from the age of borax accelerators (Lars' copyright infringement, probably)

 

Bob, the reason is that mechanical error or misadjustment (deliberate or not) can cause both front and back focusing, but this is fairly uncommon. Aperture difference however can cause backfocusing only, and is a fact of optical life – and thus not only common, but omnipresent, though not always very noticeable. So of necessity, backfocusing must be the more common complaint.

 

And BTW, you are welcome to any noncommercial copyright infringement!

 

The old man from the Age of ©

[lars_bergquist]

I would want to point out – a propos Jamie's posting, and others – that one or two aspherical surfaces are not enough to get rid of spherical aberration and thus aperture error. Especially as every correction must needs be a compromise between many incompaible demands. So the problem will probably be with us until we have all-aspherical lenses!

 

The 75mm Summicron and the current 50mm Summilux use floating elements, actually in order to counteract the image degradation at close focusing distance that spherical aberration causes, and which is of course most problematical in fast lenses. So they are less affected than other current lenses. To see what spherical does to a non-aspherical fast wide angle lens, try out a pre-aspherical 35mm Summilux. It is a horror wide open.

 

I tend to agree with the recent LFI article that the changeover from a film emulsion with appreciable depth to an essentially two-dimensional sensor exacerbated a problem that we had formerly lived with without worrying too much. Near-instant checking of the digital image did the rest by bringing the problem to the shooters' notice.

 

The old man from the Age of Scale Focusing