Real world focus shift

[Michael Tyler]

Advertisement (gone after registration)

I'll cut to chase first: For lenses known to focus shift (35lux asph v1, Zeiss Sonnar-C, Noctilux, etc), will you see focus shift at distances greater than five to ten feet, or only if shooting closer than five feet?

 

I ask for clarification, because this is the one detail I don't see spoken of in all the focus shift conversations. I've read Sean Reid's reviews and am surprised how many M-mount lenses show the shift, but all of Sean's focus targets are pretty close up. No one mentions farther distances.

 

I use and love the Canon 50mm 1.2 L lens, an optical design which displays focus shift from 2.0 to 5.6, only at distances of 5 feet or less. Beyond 5 feet, focus shift does not appear in real world shooting. Does the same apply to the Leica lenses?

[Advertisement]

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

[wda]

Increasing depth of field at longer distances can mask the error which is more noticeable at close ranges.

[pgk]

I have the 35/1.4 lux asph (prev. to current version) and have had no practical problems with focus shift that have bothered me even close up. I have recently sold my 50/1.2 Canon lens - it simply wasn't near my 50/1.4 asph M lens wide open and never seemed to have anything like the 'bite' of the M lens wide-open or stopped down, whether due to focus shift or otherwise.

[jaapv]

I am pondering the relationship between incidence angle of the light (i.e. distance) and spherical aberration. It may well be that there is a correlation there that will influence focus shift. Optical experts - come in please.

[Paul J]

Yes 5ft and closer is the short answer. That 50mm 1.2L I have and agree it's a good example of focus shift. It's often dismissed as soft when in actual fact it's sharp.

[pgk]

That 50mm 1.2L I have and agree it's a good example of focus shift. It's often dismissed as soft when in actual fact it's sharp.

Mine simply wasn't as 'sharp' as the 50 'lux, focus shift or otherwise. The 85/1.2 on the other hand was a fabulously sharp lens, but very difficult to focus on precisely the point at which focus was required. I have noticed no focus shift on the 50 'lux asph for what that is worth.

[pgk]

Advertisement (gone after registration)

I popped out today and did a check on my 35/1.4 asph (last version) which suffer from focus shift. Below is a shot from f/2~2.8 which even at 100% shows focus to be exactly where it should be (centre). Another shot at 1.4 is also spot on as is. of course. the f/8 shot. I can't detect any real world focus shift in my copy even at close distances.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[lars_bergquist]

As the diminishing effect at larger distances is an effect of d.o.f. progressively masking the shift, a major variable is how strong the shift is in the lens in question.

 

Stopping my 35mm Summilux ASPH v.1 down to f:4 could make me miss the focus completely at 2 meters, but at 7–10m it became noticeable only by the fact that the sharp zone was displaced backwards (away from the camera) so that everything in front of the object I focused on was unsharp, while there was lots of sharp territory behind it. Now the distribution of sharpness in the picture is an important variable we use in the making of it. So this phenomenon left me with a very real uncertainty that I found increasingly difficult to live with. The upshot was that I ditched the lens and queued up for the 'FLE' version already before it was officially announced. This lens does still have some shift, but it does not 'exhibit' it because it is a lot smaller than that of its predecessor.

 

This suggests that your own shooting habits and your attitude does play a large role. If you don't consciously use the depth distribution of sharpness in the picture as a variable but concentrate on the subject you focus on – an attitude I think characterises most photographers – and either do not stop down much or mostly stop down to mid-apertures, then the focus shift may not even come to your attention. I have read quite a number of posts from people who flatly deny that the v.1 in their possession has any focus shift – which is nonsense, as the shift is an effect of the overall design of the lens, not fault of the individual specimen.

 

LB

[Lindolfi]

The argument that focus shift drowns in the DOF at larger distances is not correct. Another explanation is correct and is already pointed out by the intuition of Jaapv: the angle of incidence of the lightrays at shorter distances makes the focus shift in the focal plane larger than at larger distances. At least that appears to be true for the Nokton 50/1.1 and most likely for the Noctilux 50/1.0

 

Here some measurements:

 

Focus Shift Voigtlander Nokton 50/1.1

 

At 1 meter focus distance the shift from f/1.1 to f/2.8 is from perfect focus to 25 mm backfocus (image distance 52.7802 to 52.707 = 0.073 mm)

 

The DOF of 50 mm focal length at 1 meter at f/2.8 is 17 mm

 

At 5 meter focus distance the shift from f/1.1 to f/2.8 is from perfect focus to 400 mm backfocus (image distance 50.5094 to 50.4722 = 0.037 mm)

 

The DOF of 50 mm focal length at 5 meter at f/2.8 is 494 mm

 

So the focus shift is half the size at 5 meter compared to that at 1 meter in the image plane (0.037/0.073). At 1 meter at f/2.8 the focus shift is 2.9 times half the DOF, while at 5 meter the focus shift is 1.6 times half the DOF. The DOF has been calculated with a very critical measure, but it is clear that the focus shift is about half as bad at 5 meter than at 1 meter.

 

The fact that the focus shift is halved at 5 meter compared to 1 meter both in the image plane and in the object plane compared to the DOF shows that most of the effect we see is due to the incidence of the light rays and not due to drowing in the DOF. This makes sense: DOF is just the projection of the allowable image shift into the object space. The image shift is caused by spherical aberration and the relative high contribution of the periphery of the lens elements when wide open compared to closed down.

 

This implies that focus shift remains an issue at infinity, unless it reduces to zero (0.073 mm at 1 meter, 0.037 mm at 5 meter, 0.000 mm at infinity). I will look into that later on. It may be that the designers simply have optimized for zero focus shift at infinity and could not get rid of the rest for closer focus ranges.

 

Other designs (Sonnar, Summilux) may have different shifts in the image plane as a function of object distance. Measurements are needed to see how this comes out.

[pgk]

I have read quite a number of posts from people who flatly deny that the v.1 in their possession has any focus shift – which is nonsense, as the shift is an effect of the overall design of the lens, not fault of the individual specimen

I'm far from denying it BUT I can't find it. I suspect that the actual adjustment of each individual lens may have something to do with this because so far I have failed to find focus shift in my images - and I am critical.

[Guest #12]

...

 

Why back in the image space, where little shifts are big shifts? You measured the shift in the object space, right? That is where your subject is...at one meter a shift of two centimeters is a little shift--eyes not quite as startling but hair will be sharp and picture will look about right. At 5m, a shift of 0.4m is a big shift; will not quite look right.

[Lindolfi]

Why back in the image space, where little shifts are big shifts? You measured the shift in the object space, right? That is where your subject is...at one meter a shift of two centimeters is a little shift--eyes not quite as startling but hair will be sharp and picture will look about right. At 5m, a shift of 0.4m is a big shift; will not quite look right.

 

Back in image space since sharpness in the image is determined there. That's where calculation of DOF (in object space) comes from. An object at the border of DOF at 1 meter will be as unsharp as a 5 times scaled up object at the border of DOF at 5 meter. And that is because the distance of their images to the focal plane will be the same.

[pgk]

OK I've had another really good look at my images and yes there is in fact a small amount of focus shift in the f/2~2.8 shot however the point I focused on is still sharp although it does not lie at the centre of the depth of field - you really have to look carefully to determine this. So I would still say that it is perfectly acceptable although if you were being pedantic the image's depth of field characteristics are not as one might have expected. Its certainly of too little consequence to make me think of changing the lens for the newer one so in terms of real world focus shift I would consider it to still produce thoroughly acceptable images.

[SJP]

If I understand correctly focus shift, in fact, is focal length shift.

 

The lens to sensor distance does not change for a given object focused using the rangefinder, but the actual part of image that is tack sharp moves around a bit depending on the aperture setting. I.e. the focal length depends (very slightly) on the aperture.

 

If that is that case then the whole exercise can be reduced to calculation of the DoF of various lenses with (slightly) different focal lengths. I would expect that using the standard 1/f = 1/v +1/b to calculate how the object distance (v) changes probably is enough, the DoF will not change dramatically only shifted compared to the nominal boundaries. (E.g. if the DoF is 0.95 - 1.05 m on paper, then it could be 1.00 - 1.10 m in practice, shifted by 5 cm & same DoF range of 10 cm.)

[jaapv]

Not quite intuition ; I was 90% sure I was right. So I claim 9 Brownie points out of 10

The argument that focus shift drowns in the DOF at larger distances is not correct. Another explanation is correct and is already pointed out by the intuition of Jaapv: the angle of incidence of the lightrays at shorter distances makes the focus shift in the focal plane larger than at larger distances. At least that appears to be true for the Nokton 50/1.1 and most likely for the Noctilux 50/1.0

 

Here some measurements:

 

Focus Shift Voigtlander Nokton 50/1.1

 

At 1 meter focus distance the shift from f/1.1 to f/2.8 is from perfect focus to 25 mm backfocus (image distance 52.7802 to 52.707 = 0.073 mm)

 

The DOF of 50 mm focal length at 1 meter at f/2.8 is 17 mm

 

At 5 meter focus distance the shift from f/1.1 to f/2.8 is from perfect focus to 400 mm backfocus (image distance 50.5094 to 50.4722 = 0.037 mm)

 

The DOF of 50 mm focal length at 5 meter at f/2.8 is 494 mm

 

So the focus shift is half the size at 5 meter compared to that at 1 meter in the image plane (0.037/0.073). At 1 meter at f/2.8 the focus shift is 2.9 times half the DOF, while at 5 meter the focus shift is 1.6 times half the DOF. The DOF has been calculated with a very critical measure, but it is clear that the focus shift is about half as bad at 5 meter than at 1 meter.

 

The fact that the focus shift is halved at 5 meter compared to 1 meter both in the image plane and in the object plane compared to the DOF shows that most of the effect we see is due to the incidence of the light rays and not due to drowing in the DOF. This makes sense: DOF is just the projection of the allowable image shift into the object space. The image shift is caused by spherical aberration and the relative high contribution of the periphery of the lens elements when wide open compared to closed down.

 

This implies that focus shift remains an issue at infinity, unless it reduces to zero (0.073 mm at 1 meter, 0.037 mm at 5 meter, 0.000 mm at infinity). I will look into that later on. It may be that the designers simply have optimized for zero focus shift at infinity and could not get rid of the rest for closer focus ranges.

 

Other designs (Sonnar, Summilux) may have different shifts in the image plane as a function of object distance. Measurements are needed to see how this comes out.

[Guest #12]

Back in image space since sharpness in the image is determined there. That's where calculation of DOF (in object space) comes from. An object at the border of DOF at 1 meter will be as unsharp as a 5 times scaled up object at the border of DOF at 5 meter. And that is because the distance of their images to the focal plane will be the same.

 

I thought the question was about the absolute shift in distance on the subject side of the lens (not how "o.k." that shift will be.) Which in your example increased with the subject distance, but not sure that will be true in general.

 

So I'm not sure the question requires any depth of field, if it can even be answered.

[Guest #12]

...Another explanation is correct and is already pointed out by the intuition of Jaapv: the angle of incidence of the lightrays at shorter distances makes the focus shift in the focal plane larger than at larger distances...

 

...if the lens is a single spherical surface?

 

If you look at s.a. curves for real lenses ("focus shift in the focal plane" vs aperture) they are all over the place, and it doesn't look like things work this way. But who knows...

[Guest #12]

If I understand correctly focus shift, in fact, is focal length shift.

 

The lens to sensor distance does not change for a given object focused using the rangefinder, but the actual part of image that is tack sharp moves around a bit depending on the aperture setting. I.e. the focal length depends (very slightly) on the aperture.

 

If that is that case then the whole exercise can be reduced to calculation of the DoF of various lenses with (slightly) different focal lengths. I would expect that using the standard 1/f = 1/v +1/b to calculate how the object distance (v) changes probably is enough, the DoF will not change dramatically only shifted compared to the nominal boundaries. (E.g. if the DoF is 0.95 - 1.05 m on paper, then it could be 1.00 - 1.10 m in practice, shifted by 5 cm & same DoF range of 10 cm.)

 

I'll reiterate...if you want the focus shift, you can get it on one side of the lens or the other...in one way by tracing rays of different heights, or another way by measuring the shift with a ruler like Lindolfi did. But this doesn't have anything to do with depth of field or the lens equation.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[Richardgb]

If you look at s.a. curves for real lenses ("focus shift in the focal plane" vs aperture) they are all over the place, and it doesn't look like things work this way. But who knows...

 

A primer in the effect of stopping-down on focus-shift and MTF curves, see pp24-27 of Nasse, H.H. (2008) 'How to read MTF curves', published by Zeiss as no. 30 ('Measuring lenses objectively - Part 1') in their Camera Lens News series and available to download from the Zeiss website.

[tobey bilek]

diglloyd - Focus - Focus Shift and Spherical Aberration

 

A nice well written explanation and definition. Clearly some think focus shift is the RF being wrong at certain distances. It happens when the lens is stopped down and focus moves. Leica sets the RF to work with a full open aperture. If aperture is closed, then the focus point moves. If you want the focus correct at 5.6, then you must focus at 5.6, something not achievable with a RF camera.

 

Low quality enlarging lenses used to have focus shift and we were always told to focus stopped down.

 

I have a Leica lens book by Rudolph Seck where he runs a comparison of 50 Summicron, 50 Lux (not current) and the Noctilux 1.0. The Noct clearly shows an inferior image at 5.6 compared to the Summicron. The error can be rectified by moving the image slightly out of focus and it is then equal to the best of three, Summicron.