Focus shift

[seany65]

I'm probably going to regret asking this as I probably won't understand 99% of the answers, but I've been seeing a little bit about "focus shift" on the old interwebnet, and apparently it can occur when focusing at fast apertures and then shooting at slower ones, eg. focusing at f1.4 and shooting at f8.

It also seems to affect AF more than mf, and when focusing at close distances.

I have a nikkor ais 50mm f1.4 mf, is this lens likely to suffer noticeable focus shift?

It will probably emerge after seeral answers that I should've asked different questions about focus shift, but that's what asking questions is all about.

Any help will be much appreciated, even if not fully understood. 😱

 

[Benjamin Marks]

Focus shift is an optical property that is most obvious in fast lenses and close distances. It particularly affects Sonnar type lenses, which is why you see many references to it on a rangefinder forum as many of the most famous early fast rangefinder lenses were fast Sonnar types. I wouldn't say that the focusing mechanism has anything to do with it or that the Nikon 50/1.4 AiS particularly suffered from it, but more on that below. It is particularly a problem when you are not viewing the subject through your lens, as is the case with all RF cameras. What happens with focus shift is that the point of best focus moves backwards as the lens is stopped down at a particular focus distance. Usually by f:2.8 or so, depth of focus covers the optical effect, so that by that aperture the plane of focus is wide enough that it isn't an issue. And, because having an eye in focus is the current style for a close-up portrait, it is a more noticeable defect in close ups than it is at middle or far distances.

Lens designs chosen for SLR's, like the Nikkor 50/1.4, had to minimize this particular design defect. The reason for this is that focusing is usually done with the lens wide open and then stopped down in the instant before exposure. Ironically, if you are shooting with a RF lens that does suffer from this problem, like the modern Zeiss 50/1.5 Sonnar, you can address it by shooting with a mirrorless camera. Since you are viewing through the lens at taking aperture, what you see is what you get. Many SLR's had stop down buttons or levers so that you could check what was actually covered by your depth of field at the taking aperture, assuming there was enough light to see your focusing screen.

In short, shoot your Nikkor and don't worry about it.

 

[Freakscene]

This explains it for the 28/2 Ultron but what that lens does is what essentially every lens does.
https://www.rangefinderforum.com/forums/showpost.php?p=915561&postcount=26
Focus shift manifests most in optical systems with residual spherical aberration (i.e. under corrected for spherical aberration).

I have no idea about the Nikkor although I have one, but it probably is fairly minimal because even at f1.4 most Nikkors are well corrected. Put it on a digital camera, point it at a target and test it. That’s the best way to know.

It affects different focus types (af or mf) the same when the system focuses open and then stops down without refocusing, or where focus is otherwise independent of the aperture (such as on a rangefinder).

It is a fairly simple concept, but it is often wildly incorrectly described on the internet, or, worse, ascribed to problems with calibration (such as saying a lens that consistently focuses in front of or behind the intended point of focus has focus shift, when actually it is just out of specification).

If you have specific questions, fire away.

Marty

 

[seany65]

Thank you Benjamin and Marty for the replies and for doing it in ways I can understand.

I've seen loads of questions about a wide range of things on the interwebnet (I'm going to keep saying that until it becomes a real word in dictionaries, lol), and I've seen even more replies to those questions, and generally speaking:

75% are really technical.

70% are practically impossible to understand.

76% are completely different to all the other answers.

68% lead to arguments between those posting replies.

57% seem to bear no relation to the original question.

80% are actually replies to whatever other repliers giving info have posted.


Does this seem familiar to anyone else?

 

[Freakscene]

Does this seem familiar to anyone else?

You are welcome. A lot of optical concepts are poorly- or mis-understood. When anyone can post without fact checking or qa/qc problems and inaccurate information emerge.

If you can stick to slower, well-corrected lenses and you will never notice a thing.

Marty

 

[awilder]

If using a film Nikon SLR, the standard central split image and micro-prism focusing aids had a slight focus offset (approximately +0.06 mm) compared to the more accurate surrounding ground glass field to compensate for focus shift at larger apertures. That's because with these aids you're not really focusing at f/1.4 but in reality around f/4. In fact, later published camera test reports from Popular Photography back in the mid-1980's actually posted the amount of focus offset on various cameras they tested. The only drawback is that these focusing aid offset affects critical focus on all lenses so for precise focus, either a B/E screen is advised since they have no compensating focus offset. If using it on a mirrorless camera you focus at the working aperture so focus shift is no longer an issue.

 

[Richard G]

Focus shift is great on the internet and often only survives well there. Even the C Sonnar Zeiss ZM focus shift wouldn't bother a lot of people until they read about it. And many move from blissful ignorance to utter misery such that even leaving the lens on the shelf is not enough and they have to physically get rid of it. I love mine. And it's got focus shift.

 

[Doug]

In Sean Reid's lens reviews he typically checks for and reports focus shift. It often seems to take fairly precise efforts to identify and quantify it. I am not particularly picky about it, so I don't notice this effect in actual use of the same lenses where he identified a little shift, for example the Voigtlander 50mm f/2 Heliar.

The only lens I've actually noticed the focus shift was the C-Sonnar wide open and fairly close (a couple of yards). It was easily counteracted by leaning forward an inch or so after RF focusing. I found it focused right on at f/2.8. I sent it to DAG for 6-bit coding and a checkup, and it came back set for ideal focus at about f/2- f/2.2 -- resulting in no noticeable shift effect wide open, or at any other aperture. It's there, but so minor I don't bother with it.

 

[seany65]

Thanks to all for the new replies and info.

 

[p.giannakis]

I have the AF Nikkor 50f1.4 and have used it wide open in many occasions. No focus shift noticed. The lens is soft wide open by nature, not by focus shift. There is a difference between the AF nikkor 50 f/1.4 and the older LTM Nikkor 50f/1.4 (maybe you read something regarding the latter one). I sometimes use a Jupiter 8 in my Leica, there is a mixture of back focusing due to focus shift and difference between German and Russian LTM standards but it never bothered me to the slightest. It would never stop me from using these lenses, for me it is more a theoretical problem rather than a practical