Differential Focus Errors

[lct]

Advertisement (gone after registration)

My Colorplan lens takes slides curvature into account in its optical formula. Could it be the same for some view lenses as well?

[Advertisement]

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

[Guest #12]

Yes, that is exactly what I'm saying.

 

now I'm confused...I thought your argument is

 

film is more forgiving of focus errors,

so digital equipment needs a stricter tolerance.

 

If this is true, then there will be some experimental demonstration of it somewhere?

 

And for work where focusing at wide apertures is everything, there would be some advantage to a film camera?

[ho_co]

... Solms adjusted all the newer lenses free of charge and charged quite heavily for the older lenses...

Teddy--Were these "newer" and "older" lenses all bought new by you?

[Guest #12]

My Colorplan lens takes slides curvature into account in its optical formula. Could it be the same for some view lenses as well?

 

Yes, but I don't know the models. I would love to know. There is a Minolta lens with adjustable field curvature.

[ho_co]

That was a funny one, LCT.

 

Every year, Kodak improved the quality of their projection lenses. Every time they did, Leica had to come out with new demo slides to show the advantage of the Colorplan over the Ektar.

 

Kodak continually improved the flatness in which their mounts held the film. Every six months to a year, the US cardboard mount was changed to hold the film flatter.

 

And then suddenly, Kodak came up with the genial (I know, you can't say that in English) recognition that they should quit improving the mount and instead adjust the lens to compensate for a general film curvature.

 

It was brilliant, and forced Leica to do the same with the Leitz Colorplan CF.

 

Since you Europeans tended to use glass mounts, this little cat-and-mouse game may not have been recognized on your side of the Atlantic, but it was a big deal for Leitz USA.

[lct]

Thanks Howard. I did not know the Leitz vs Kodak story but i wondered if "film" view lenses did not take curvature into account in their formulas to the point where flatness of digital sensors might become a problem. I guess not but i was (am) just curious.

[ho_co]

Advertisement (gone after registration)

The flatness of film is unpredictable. Leica fought it with the screw-mount cameras by means of the tightness of the pressure-plate, then with the initial M3 with the glass pressure plate, with the SL with a small "film tunnel," with the R3 (IIRC) with a larger one.

 

Nonetheless, you can see the film's unflatness by putting a waste roll into the camera, removing the lens, opening the shutter on B, and looking at the film. It's amazing when you do that that anybody's lenses produce a good image. The effect is even worse on view cameras, simply because there's a much larger area that can't be held flat.

 

But of course, you can't predict how it will curve, so you produce lenses designed to project a planar image.

 

Because the film wasn't that flat anyway, the differential focus of some lenses (like the 35/1.4 asph) didn't need to be compensated. In fact, it was hardly noticed until a truly plane image substrate appeared in the digital era.

 

(I may be answering a question you didn't ask; if so, my apologies.)

[lct]

Yes that's what i suspected as well. We would see the problem ourselves if focus errors were caused by the flatness of digital sensors. AFAIC i'm lucky enough that most of my M and R lenses focus properly with digital.

[Guest #12]

o.k., now I have two people saying

 

1. curved film is more sensitive to focus error than flat film.

2. digital is more sensitive to focus error than film.

3. digital is more sensitive because film is curved.

 

so I am obviously misreading. How do I prove #2, i.e. what is some reliable source for this idea--some BJP article, some experiment, anything?

[ho_co]

Comprehensive LFI article from a couple years ago, where they and Leica discussed the topic in great detail and even said that if they were designing the 35 Summilux for digital, they'd have given it a floating element.

 

That exact thing happened a year or so after the article, when the current 35/1.4 replaced its predecessor.

[Guest #12]

thanks...I had no idea they did so many tech articles

[jaapv]

For 4x5 film there used to be a vacuum film holder by Schneider, to keep the film as flat as possible. The trouble with such large film formats is twofold, the DOF is razor thin and as one tilts the camera forwards the film will sag in the middle. All totally obsolete by now of course with MF digital backs.

[pixelman]

Just ran some tests with a few lenses and the M9 on a tripod.

 

Sadly, it seems that my new M9 is front focusing and would appear to need some RF adjustment.

 

Here are 2 examples first is a brand new 50mm 'cron, second is a type III 50mm 'cron, both at 5ft.

 

Guess it is a trip to NJ for the new M9...

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!

[ho_co]

I take it those are both shot at maximum aperture, right?

[pixelman]

Yes, both wide open @ f/2 on the 'crons.

[pixelman]

Here is the full frame view of the first test image for context (from the newer 50mm 'cron).

 

I used an iPad to document which lens was which. The case I have for the iPad worked quite well over the rail of a deck chair.

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!

[ho_co]

Thanks.

 

That looks like an almost negligible difference to me, but it's definitely not the way it should be, and it should be taken care of.

 

You've got similar results with two lenses, so it's likely the camera, as you said. Sorry; that's a disappointment. But they'll get it turned around quickly for you, I'm sure.