How to properly focus shallow DoF lenses?

[russell_brooks]

Ok, I was testing my noct and found something that might explain why my "hit" ration of in focus pictures is so low. I focus on the door knob across the room (roughly 3 meters) as the test. If I bring the focus from infinity to the knob I get just a bit less than 3 meters. If I bring the focus from 1 meter to the knob I get just a bit over 3 meters! Ok, normally no big deal, but the DoF at f1 with the noct is so small that even this small discrepancy is greater than the DoF.<br><br>I remember reading once about the way to focus a rangefinder is to not "hunt" for the focus but to bring it from one extreme and then stop when it is in focus. But what exterme were they talking about infinity or 1 meter? Thanks, Russell

[karl_georg_wolf]

Hello Russel,

 

<p>

 

when focussing close with lenses wide open, that´s f2, f 1.4 at 0.7 -

2 m, 3 m would be a bit out of bounds for this, set your r/f and then

move your head (with your eye at the camera) to an fro (that is:

towards your motif and back a bit). Stop moving where you think focus

is perfect and release.

 

<p>

 

I found, I got more precise results that way. Just try.

 

<p>

 

Good shooting

[michael_darnton1]

The advice, which I've seen before, to not hunt for focus is bad, I

think. Shoot past the point, in both directions, attempt to figure

where the middle is by going over in smaller and smaller increments,

and then hit the spot. Regard the last pass as your "not hunting"

pass, if you want, but by that time you already know a lot more than

just doing it the first time.

 

<p>

 

Another hint--a overshoot coming from the front going back takes you

a lot farther out of focus than the same overshoot with an equivalent

angular error on the focus ring if you're coming forward.

 

<p>

 

And third, check the infinity setting of your rangefinder on a

regular basis, and learn how to adjust it yourself.

[russell_brooks]

I tried this on both my M3 and M2 with the same results. And they

were serviced by Leica in the last 9 months.<br><br>The tricks with

bobbing your head and with hunting for the focus I understand and

have used sometimes. What I am having difficulting figuring out is if

my doorknob will be in focus at 3 meters minus a little bit or at 3

meters plus a little bit. Of course I will make a test. But the

rangefinder shows that both can be considered in focus depending

which direction you start to turn the focus ring from. But there can

only be one correct distance! I just want to find a consistant

focusing method for my noct... thanks...

[michael_darnton1]

If you're asking if the rangefinder delivers different results going

in different directions, the answer is no, not on a camera in good

condition. If the camera is OK, the problem is your eyes or your

method. I'd guess that the correct focus is between that which you

get with the two methods you're using.

[john_collier2]

How new is your lens? The Noctilux is a very heavy lens and as

the focusing helicoil wears, even slightly, the optical unit sags

under the weight. So if you press upwards when you focus the

focus will be affected. Check it again very carefully to see if the

focus shifts slightly when you let go of the lens. If so, practice

using a light touch when focusing. (Thanks to Dick Gilcreast for

pointing this out in his article in the LHSA Viewfinder)

[mike dixon]

Another thing to consider is whether there is any "play" in the

focusing ring. None of my chrome lenses has any, but all of my

black/aluminum lenses have a tiny "dead spot" where the ring turns but

doesn't affect focus. It could be enough to show up as a slight

difference on the focusing scale while not being a difference in the

actual location of the focusing cam.

[andrew_schank]

I read that its best to focus from near to far and stop when the

image looks coincided (and not "hunt" back and forth like with an

SLR). You may want to try it that way and see if the images are more

consistantly in focus.

[andrew_schank]

I just pulled out my fastest lens, a 50mm 1.5 Zeiss Sonnar (LTM with

adapter) and focused it from 1 meter out and from infinity in, on an

object about 6 feet away with my M3. I found the exact same spot on

the focus scale going from both directions, and repeated the results

several times. I wonder if there is some play in your system, either

with the lens or in the camera body.

[andy_piper2]

a) Which extreme? whichever one works - i.e. try actually taking

pictures while doing your 'test' and then see which one is sharpest -

focusing from infinity or from close-up.

 

<p>

 

But definitely 'hunting' for focus is to be avoided - if you have to do

that you might just as well get an SLR. =8^o

 

<p>

 

b) Trying to line up the two rf images is like trying to 'see' depth of

field. We all know that if the circles of confusion are small enough we

will 'see' part of an image as sharp even if it is not PERFECTLY sharp.

In the same way the rangefinder depends on our eyesight - we may see

the RF images as aligned even if they are not PERFECTLY aligned - and

therefore undershoot, or overshoot, the point of perfect focus.

 

<p>

 

This, Russell, is why you get different results focusing 'in' and

focusing 'out' - you are an 'overshooter' by habit and tend to go

slightly past the merge point (which I'll bet is actually 3 meters

exactly.)

 

<p>

 

(Thought experiment for everybody - psychologically do you choose

'correct focus' as the place where the images merge - or the point a

little further on where you see them start to split apart again? )

[russell_brooks]

I did a quick test with my 90mm summicron last night and got roughly

the same results. Hmmm.... The strange thing is both my M2 and the M3

were at Leica last year. And the M3 was there with the noctilux to

match the focus. But both bodies were adjusted by the same man and

probably the same equipment on his side so it could be that... I

dunno. Later this week I will run a test on my friend's .85 body and

see if I get the same results. I'll let you guys know with a post

here.

[michael_darnton1]

Have you checked infinity alignment? I sent my 85/1.5 Canon to

someone who gets mentioned here regularly as the king of lens

repairs, and he sent it back out of whack (along with a couple of

other problems, including all the screws being rattliing loose). I

sent it back to him. His response to me (after he tightened the

screws) was that it was set appropriately and properly to come to

infinity at 200M. With this lens I can tell the difference in focus

between one mile and two, and it's visible in photos wide open--200M

was ridiculously wrong--so I ended up fixing it myself. Because this

lens is the least tolerant of what I have, once in a while I check my

two bodies with it, and adjust them to it, letting the less picky

lenses take care of themselves. You might need to learn to do that

yourself.

[michael_darnton1]

Incidentally, one of the things I learned from this exercise is that

if you align your own RF, don't do it close to a window--the effect

of going through glass is enough to throw things off. Either stand

back 20 feet or so so the two RF windows aren't looking at the glass

from different effective angles and thus seeing slightly differently

refracted views, or open the window.

[russell_brooks]

Ok, I tested it on my friend's .85 M6 TTL. It really looks like it is

the lens instead of the body. The focusing mechanism looks fine but

somehow it is off. I will send the lens into Leica now to be

repaired. But let this be a heads-up to you guys out there with older

fast lenses to confirm the focus...