Living with focus shift ... how to move successfully from SLR to RF ?

R

Raphael

Newbie
Local time
1:30 AM
Joined
Jan 28, 2009
Messages
5
Good afternoon!

I'm new to the world of rangefinder cameras, and while I have been playing with a VF 135 a few years ago (what a nice lens!), I am still yet to buy my first proper RF camera.

The last years, I used a Pentax MX and a K5ii, mostly for street and landscape picture when travelling, but I was always a bit annoyed by the bulk of the K5 with 20mm 2.8 lens and hood. A few months ago, for travelling with less weight, I bought a beaten up Fuji X-E1, and it happened that I also got a Voigtländer 35mm f/1.4 SC lens. ( I originally wanted the f/2.5 version, but I'm a student and my financial limits are very tight, and I just got a significantly cheaper offer on the used f/1.4 than on any used f/2.5 ). I must admit, photography is for me more about the fun while taking the pictures and later developing the film and the memories associated to the places, rather than resolution or distortions, and I just happen to enjoy using MF lenses much more than the optically stellar current fuji lenses.

So, the X-E1 is fun, but, just like with my SLR setup, I'd like to have a film body to use the same lenses as on the X-E1, in this case an M-mount body, and I already decided to go for the R3m, once I find someone willing to sell his used model ;).

I noticed on the X-E1 that the 35mm f/1.4 lens has quite a lot of focus shift -- which is no problem on a digital camera, neither on a SLR if you can stop down manually, but, I don't know how to handle it on a film RF camera -- the rangefinder coupling is calibrated at f/1.4 ? f/4 ? f/8? And do I have to correct the focus then using the distance scale and my experience of how strong focus shift is, or is there any better way to achieve well-focused images?

Cheers,
Raphael
 
While I have no experience with the specific lens, the rule to master focus shift issues is to make your own focus test, remembering to keep the camera parallel to the test chart and vertically positioned- in other words, the camera has to be upright, and the test chart facing it frontally at a 45 degree angle.
There are a couple of points worth noting: the DOF on film is quite a bit higher than on digital, so it is unlikely you will get much real life problems with such a short FL - you can verify your dof values here: http://www.dofmaster.com/dofjs.html.
The other thing is that unless you plan for explicitly slow photography or want to expose incorrectly your film on purpose, you would be much better off with the R3A or R4A. The R3x has 40mm frames as the widest, and they are very tight, so the body you want is either R2A or R4A.
 
Well, DID you stop down to correct focus before each and every shot when you were using SLRs? Probably not - I certainly have never seen anybody do it...

As a matter of fact, focus shift is a similar issue for RF and SLR - even the Exakta already had optional, crude auto-aperture lenses, so SLRs (past the large format age) had a similar issue of different focusing and exposure environments. If you have ancient, ultra-fast or bad lenses, you'll encounter it, but any reasonably modern and good lens is effectively free from it, unless you pixel peep or use the lens in some scenario it never was intended for (like a fast lens on extension rings).
 
It's a manual focusing ... you have a choice as to where to set the focus. Do a focus test and learn how far off the focus is at various lens opening (it shouldn't be THAT far—if it is, have the lens and rangefinder re-calibrated). Then pick a lens opening and learn what the rangefinder split looks like when the lens is at the correct focus point.

It's easier done than said. Once you're out of the range of near-wide-open and close distances, it all falls into the noise. RFs are not best at near-wide-open and close distances anyway.

I tend to standardize my aperture settings. I'm usually at two stops down (usually long past the focus shifty part) or f/8. So it becomes quite easy to ignore focus shift entirely, just line up the RF and make an exposure.

G
 
Thank you very much for your kind replies so far.

@mfogiel: I expected DOF on film to be shallower, not higher, as for a given focal length, f-stop and fixed magnification (subject size on photo), with larger "sensor" size you have to get closer to the subject, leading to a larger angle between incident rays, and thus a shallower DOF (as the rays diverge "more steeply" when moving out of the focal plane in either direction).

Then, the reason I would like to get a R3m is the 1:1 magnification of the viewfinder, even if it means using the entire finder as a guess of the 35mm FOV.

@sevo, Godfrey, ferider and BillBingham2: indeed I never had to stop down when focusing using the SLR, but the Pentax 50mm f/1.4 has much less focus shift than the VF 35 f/1.4 -- on the X-E1, focus shift is quite easy to discover: when focusing on a close-ish object, let's say 1 meter away, at f/1.4, and then stop down to f/4 or so, the central point I focused on is not sharp any more -- for an ideal lens I'd expect the central point to still be in focus and the DOF increase more or less symmetrically around that central point; instead, the focus point moves outside of the in-focus region. Now this is likely linked to spherical aberration being more evident for large apertures, resulting in something which behaves like an f-stop dependent field curvature) .

Cheers,
Raphael
 
ferider said:
In practice, focus shift is overrated.
Click to expand...

It depends on what lens you are shooting.

I have a current Zeiss Sonnar with focus calibrated for ~f2.8. If I shoot a portrait wide-open near minimum focus distance there is no way that this will look good unless you compensate for the focus shift. The trick of focusing on the ears rather than the eyes or just tweaking the focus ring by a small amount works quite well for me in this case.

But the Sonnar is the only lens I have where this is an issue, and I agree that a lot of people fret over nothing. I also shoot the Canon 50mm f1.2L - and I think that people who complain about that lens' fairly trivial focus shift should be made to use the Zeiss for a while...
 
Raphael said:
...
indeed I never had to stop down when focusing using the SLR, but the Pentax 50mm f/1.4 has much less focus shift than the VF 35 f/1.4 ...
Click to expand...

Because SLR cameras since 1960 have incorporated auto-diaphragm operation to allow viewing and focusing with a bright viewfinder, it has been a design priority with SLR lenses to minimize focus shift when the aperture is stopped down. It would be a disaster if SLR lenses exhibited focus shift since you are reliant upon the image formed by the lens to set the focus correctly, every time.

RF cameras do not view through the lens and are not constantly opening and closing the aperture; controlling focus shift has not been as large a priority in lens design as a result. MOST do pretty well; some are notably prone to a lot of shift. I haven't heard this to be an issue with the Nokton 35/1.4 SC ... It isn't with my Nokton 40/1.4 MC, which is a very similar design. But eh? stuff happens.

I wouldn't obsess over it. Get your RF body, fit your lens, and go take some pictures to test it and get to know it. At most it should require a minor tweak to the focus setting. Don't rely upon what your TTL electronic camera sees as the RF camera and RF focusing mechanism will see things differently...

G
 
sevo said:
Well, DID you stop down to correct focus before each and every shot when you were using SLRs? Probably not - I certainly have never seen anybody do it...

As a matter of fact, focus shift is a similar issue for RF and SLR - even the Exakta already had optional, crude auto-aperture lenses, so SLRs (past the large format age) had a similar issue of different focusing and exposure environments. If you have ancient, ultra-fast or bad lenses, you'll encounter it, but any reasonably modern and good lens is effectively free from it, unless you pixel peep or use the lens in some scenario it never was intended for (like a fast lens on extension rings).
Click to expand...

Yup, my Nikon 50 1.4 AI/AIS lenses focus shift from 2.8- 4.
 
Raphael said:
........ when focusing on a close-ish object, let's say 1 meter away, at f/1.4, and then stop down to f/4 or so, the central point I focused on is not sharp any more -- for an ideal lens I'd expect the central point to still be in focus and the DOF increase more or less symmetrically around that central point; instead, the focus point moves outside of the in-focus region. ....
Click to expand...

When you are a subject close up the actual zone of focus is not as wide as say when you are three meters away. As such any movement on your part or that of your subject after you focus (e.g. when you are adjusting the aperture) might actually be the cause of what I would cause a major focus shift (taking the point of focus so far out that your initial plain of focus is no longer in the acceptable zone of focus).

I'd be hard pressed to think that a modern lens would have such a major focus shift if it were still in factory spec. While not connected in any way, you might bump the focusing ring when you adjust the aperture.

To test this concern get a tripod, even a cheap one will do. Find a yard stick, tape measure, some with regular marking. Set the camera so it is on the same plain and get some small objects to place in the frame. Place them at different distance from the focal plain across the frame. Say closest on the far right and then you place them closer to the camera the are located further left in the frame. Place the item you will focus on right next to the ruler and then items closer to the camera to the left of the frame.

Take a shot wide open, then carefully stop down, one stop a frame till you are as far as you can go.

Let us know how you make out with the tests.

Best of luck.

B2 (;->
 
I use the Nokton 35/1.4 on film, my main-lens next to the 21mm Skopar...

My suggestion: just shoot the damn thing ;) don't think about focus-shifting or whatever. It's more likely that my own misfocusing, the cosine-error or small body movements at short distances will throw my focus off then any focus shift with this lens.
Focus shift with the Nokton on film is really overrated. I never experienced it in reallife shooting... I once checked carefully for it: yes there is some shift, you can spot it at 0.7m distance and even less at 1.0m. It's "critical" at f2.8 and f4.0 (f2.0 is spot on with my M2 & Nokton, at f1.4 it's slightly front focusing) but even at those stops the depth of field at 0.7m and 1.0m still covers the focused point.

Some time ago I've read too much about the Nokton's shift... so I tried compensating for it at 2.8/4.0... _this_ resulted in lot of misfocused shots :rolleyes:
 
Good evening -- I'm happy to hear that trying to correct for focus shift for this specific lens actually makes it worse ;).
Either way, here are some samples taken with the X-E1 (100 % crops) showing the focus shift
https://www.dropbox.com/sh/wumsnwww7pvjiq2/AACCHn6spQaVZHsa92kDW2O_a?dl=0

Once I found a used R3m (hopefully in a not too far future), I'll report here about the results.

Cheers, thanks, Raphael
 
I had problems with this lens. Suddenly I had focus problems I never noticed with other lenses. I sold the lens and picked up a 35mm Ultron f/1.7 instead. Maybe you should look at getting another lens - just saying; I am sure others will have a different opinion.
 
"@mfogiel: I expected DOF on film to be shallower, not higher, as for a given focal length, f-stop and fixed magnification (subject size on photo), with larger "sensor" size you have to get closer to the subject, leading to a larger angle between incident rays, and thus a shallower DOF (as the rays diverge "more steeply" when moving out of the focal plane in either direction). "

On film or digital, a 35mm f1.4 lens stays the same in terms of the angle at which the rays hit the focus plane, and the subject distance is an extra variable. However, smaller formats need to be enlarged more, hence a cropped digital sensor will require a smaller CoC=less DOF, and as mentioned above, film emulsion has physical depth, as opposed to a flat surface of your sensor, so = more DOF on 35mm film again.
Unless you have a very flat face with protruding eyes, I think you will not find the R3x VF satisfactory for 35mm.
 
I have the 35mm f/1.4 classic and thought I had focus shift until I did the tripod test, then I realized that it was my lack of skill at focusing that was the issue, so I stopped worrying about it.

Also, how do to get focus shift on an X-E1? You are looking through the lens when focusing.
 
You must log in or register to reply here.
Back
Top Bottom