Focus Shift on Some Lenses

[nscali]

Advertisement (gone after registration)

Tim,

How right you are!

If I redo a focus test with the subject off centre (trying desperately not to alter distance)

the focus does seem to be spot on as you stop down.

Are we safe in concluding that this is just a fact of life with this lens and perhaps are expectations are greater than others?

Again, after a minor hex adjustment the issue, while still there, has improved slightly,

perhaps to the point where I may not have noticed it if my rangefinder was at that point to begin with.

Nicky

[billh]

This might be my WILD guess:

 

Actually the problems are mostly on our lenses not on M8...

 

We never realize the problem with our lenses because when we use film, we never create a big print or use loop to check.

 

 

Jerry

 

 

Jerry, I always use a loupe, 8x to 20x, to check my negatives, and indeed I have made many big prints from M negatives. Possibly some lenses are off (my APO135 f3.4 was off, on film, and I sent jpegs to Leica and they replaced it), but the fact most being reported had no problems with film, makes me think the M8 is the culprit. I have one friend, now on his 4th M8, and this one finally focuses perfectly with all of his lenses.

 

Bill

[tashley]

Jerry, I always use a loupe, 8x to 20x, to check my negatives, and indeed I have made many big prints from M negatives. Possibly some lenses are off (my APO135 f3.4 was off, on film, and I sent jpegs to Leica and they replaced it), but the fact most being reported had no problems with film, makes me think the M8 is the culprit. I have one friend, now on his 4th M8, and this one finally focuses perfectly with all of his lenses.

 

Bill

 

Please don't tell me we have to buy FOUR! The pain (the pleasure...)

 

But seriously, what you post is fascinating if unverifiable with a sample of one: it implies that there is a magic sweet spot of M8 construction and RF adjustment that makes all lenses focus. Personally, I doubt it: if I had several Leica 35mm lenses with issues and a CV35 that focusses as well as my other glass (28 cron excepted) then to me that indicates a lens issue, and I'm guessing that tiny tolerances in shims are the answer.

 

But that really is a guess, cos none of the reviewers is pursuing this.

 

:-(

 

T

[tashley]

Tim,

How right you are!

If I redo a focus test with the subject off centre (trying desperately not to alter distance)

the focus does seem to be spot on as you stop down.

Are we safe in concluding that this is just a fact of life with this lens and perhaps are expectations are greater than others?

Again, after a minor hex adjustment the issue, while still there, has improved slightly,

perhaps to the point where I may not have noticed it if my rangefinder was at that point to begin with.

Nicky

 

I posted a 35 lux shot (and I can't find it now) of a scene where the image had foreground front left and background far right, and it was all in focus apart from a bit in the middle. That's when I learned this lesson.

 

As Hamlet said, "something is rotten in the state of"... (it said Denmark but hey, Old Bill never had a Leica...

 

t

[jonoslack]

Jerry, I always use a loupe, 8x to 20x, to check my negatives, and indeed I have made many big prints from M negatives. Possibly some lenses are off (my APO135 f3.4 was off, on film, and I sent jpegs to Leica and they replaced it), but the fact most being reported had no problems with film, makes me think the M8 is the culprit. I have one friend, now on his 4th M8, and this one finally focuses perfectly with all of his lenses.

 

Bill

Hi Bill

How does a 20x loupe on a negative compare with 100% on a 23" monitor?

 

Although I'm aware that out of sync rangefinders can cause grief, it seems really unlikely that different lenses can have different characteristics on one body (like backfocus etc.) and then all the same on another body?

 

I've done some cross - checking with my two bodies, and if they are behaving differently with lenses, then I can't pin it down. They seem exactly the same to me.

[t0tor0]

I have a VC Nokton 40mm which has a terrible back focus. So I did some research on the internet and found this.

 

http://www.focustestchart.com/chart.html

 

Was suppose to be for his D70 but I think can be used on any cameras.

 

I realize that this off focus issue also depends on the range. I have a 75mm lux and it focus nicely beyond 1m anything less it front focuses.

 

An also did realize that some lens are slightly front focus while some are back focus.

 

In most of my lenses, most of the time at wide open, the line is in the focus field. But just barely. Mostly front focus though which I am quite surprise as most people are complaining of back focus.

 

Just my contribution here to all. Hope it's useful.

[t0tor0]

Advertisement (gone after registration)

I realize that this off focus issue also depends on the range. I have a 75mm lux and it focus nicely beyond 1m anything less it front focuses.

 

Let me add on. This picture is taken with a 75 lux at about 0.75m wide open. Focus is on the eye.

 

 

Does it seems that the Field of Focus is diagonal or curved rather that parallel to the film/sensor plane? See the focus on the left. That part is about 15cm - 20cm from my cat's eye.

 

See that his paws are off focus? How did the focus jumped from the left from to the nose and eyes? And yet the window grills same distance to the camera as his nose is off focus?

 

Weird...

[russell]

Maybe the sensor is warped?

[Jamie Roberts]

{snipped}

Does it seems that the Field of Focus is diagonal or curved rather that parallel to the film/sensor plane? See the focus on the left. That part is about 15cm - 20cm from my cat's eye.

 

See that his paws are off focus? How did the focus jumped from the left from to the nose and eyes? And yet the window grills same distance to the camera as his nose is off focus?

 

Weird...

 

Uh, guys, before this gets out of hand, the focal field *is* curved. Or rather, the distance relative to the sensor creates a curve due to the angle you take something on...

 

The left part of the image that is "towards the camera" is actually the same distance (or similar) to the cats eye from the sensor's perspective. The paw is not. The window grills are not either, evidently. I'm assuming this is taken wide open.

 

This is why we wedding photographers have groups gather in a slight semi-circle. It's also why "focus and recompose" technique many AF shooters use does not work effectively (because you're changing the angle of the subject--and hence the distance).

 

There's more to focus than flat-plane stuff.

 

See here:

 

Why Focus-Recompose Sucks

 

and here:

 

Don't recompose!

 

Of course, at f4 and higher this means less and less, since DOF compensates for this somewhat. But if you're very close in, DOF is also decreased.

 

IOW, Tim is not having a focus and recompose / angle of focus problem (but it's a pretty common with very fast lenses)

 

And yeah, my 35 lux is as sharp as razor blades at f4. To me this is a lens adjustment issue Either that, or Leica has a ton of bad coded adapters, which seems a bit unlikely (but not of course, impossible).

[t0tor0]

Err... Jamie:

 

I didn't read anything on a curve DOF. In fact, it says that the DOF is a plane and hence focus and re-composing will result in back focus.

 

If DOF is a curve, then focus and re-compose will be perfect right?

 

hmm... so how many percent of this back focus issues on the m8 due to recomposing?

 

anyway, thanks for the link.

 

what about the idea that the sensor is curve a little? maybe they did it intentionally for light to hit the sensor more perpendicularly at the edges? and forgot about the focus issue? Remember the saying that a digital range finder cannot be full-frame because light has to hit the CCD perpendicularly. So maybe the sensor is warped to squeeze out a 1.33 crop rather than 1.5. If the RD-1 do not have any focusing problems with M-lenses, then this could be true...

 

i think i will just forgot about all this and go on taking pictures.

 

hahaha.

[Jamie Roberts]

The effective focal field (the shape of what's in focus, in effect) and depth of field are two different, but related things.

 

IIRC, the effective focal field, and the results you're seeing with your cat shot, are mainly the result of different points of the shot being further away than others relative to the film plane.

 

Having said that, some lenses exhibit more or less curvature of field anyway. Certainly wide open, my 35 Lux has a curved effective focal field.

 

I really don't think the sensor itself is curved

 

At F8 the focal field curve doesn't look much that way, due to DO (unless I'm close to the near focus limit).

 

Hope that makes sense. Shooting is a good idea (me? I need sleep!)