very basic "focus shift" question

[uroman]

Advertisement (gone after registration)

I have a very stupid, and perhaps very basic "focus shift" question

 

I was using the old noctilux and the CV nokton this weekend at a museum. First, let me say that both lenses are great for low-light museum shots...

 

If i focus with the lenses set to an aperture of 2.0 or 1.4, and then shoot without changing the aperture, then the lenses seem very sharp. So in other words, is it true that if you dont change the aperture after focusing, then you dont get "focus shift"?

 

I had no probems with sharpness at any small aperture from wide open to f2.8 but in ALL cases, I first moved the aperture ring to the desired aperture, then focused, then fired the shutter...

[Advertisement]

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

[Jeff S]

The sequence of setting aperture first, or focusing first, is not relevant.

 

Focus shift generally occurs with very fast lenses (as in your samples) when stopping down from the widest setting (or two). In other words, many fast lenses focus dead-on when shot wide open (if calibrated correctly). But when using smaller apertures, typically the next couple of apertures, the lens shifts focus...either toward the front or the back of the subject...even though the viewfinder shows good focus. Then, after stopping down further, the lens again focuses fine. This is an inherent design issue in many fast lenses, including even a 35 Summicron asph.

 

So, the aperture being used is the determining factor, not the sequence of actions.

 

For me, this is only important if it creates a problem in the final print. Often it's of no consequence.

 

 

Jeff

[uroman]

thank you

i did not have any "shift" issues with both lenses... yet people say that these lenses have shift. notably, i was VERY impressed with the nokton esp for its price. I expected both to have issues with shift, but couldnt get any...

[bpalme]

You can do a test on a yard stick. Focus on the same point, take a pic... stop down, take another pic. ect.

[Jeff S]

thank you

i did not have any "shift" issues with both lenses... yet people say that these lenses have shift. notably, i was VERY impressed with the nokton esp for its price. I expected both to have issues with shift, but couldnt get any...

 

To be sure, you would need to shoot carefully on a tripod, with a target and measurement scale. But, as I said, if you don't see problems in your prints, there is certainly no need to go searching for problems that don't matter.

 

Jeff

[lars_bergquist]

Those 'yardstick tests' can lead to lots of confusion and paranoia. I would suggest a more practical method.

 

Put your camera on a tripod and focus on a target with, say, some text on it with a range of type sizes. The distance should be 1 – 2 meters. A bottle label does often work fine. Check by chimping that you have the best focus you can get with the lens wide open. Now, without changing the focusing, stop down to 2.8 and take a new picture. Follow up with an exposure at 5.6. The exposures should be evaluated at least on a good computer monitor, possibly by printing. If at 2.8 the fine print suffers a lack of contrast compared to the wide open frame, then you have a problem. At 5.6, increasing depth of field will normally cover up for the fact that the plane of best focus has moved.

 

Every spherical lens, be it a single lens or a composite objective lens, has spherical aberration. This comes from the fact that rays passing through the peripheral parts of the lens come to a different focus than rays through the center or near center. It is clear that fast lenses are wider – have a larger diameter – than slow lenses, so they should have more spherical aberration.

 

Photo lenses with much spherical aberration are traditionally calibrated for the best compromise focus when wide open, i.e. the diffuse spot that a point source of light creates on the sensor or film is as small as possible. But when you stop down, the peripheral parts of the lens are cut off and the rays focused by the axial part are more prominent. The compromise changes – but not your focusing. The best compromise is now a bad compromise, and you lose definition.

 

This holds for all fast lenses, not only for rangefinder glass. SLR cameras have the same problem, because their lenses are focused – visually or by an AF mechanism does not matter – when wide open and are stopped down for the exposure. Aspherical lens surfaces and/or floating elements give a lens designer more parameters to work with, and thus more leeway to improve the state of correction, including the correction of s.f.

 

The still somewhat non-spherical old man

[lct]

Advertisement (gone after registration)

Those 'yardstick tests' can lead to lots of confusion and paranoia...

+1

[jaapv]

thank you

i did not have any "shift" issues with both lenses... yet people say that these lenses have shift. notably, i was VERY impressed with the nokton esp for its price. I expected both to have issues with shift, but couldnt get any...

If you have no problems why are you worrying?

Btw- your method will work - but only on an SLR (or EVIL) with the lens stopped down to the final aperture.