75mm APO depth of field

[John Black]

Advertisement (gone after registration)

Just out of interest can anyone, with experience of both lenses, tell me how the depth of field of the 75mm APO-Summicron at f/2 compares with that of the last version 75mm Summilux at the same aperture? Personally, I find the Lux extremely shallow at f/1.4, only very marginally deeper at f/2 and actually focusable with accuracy by f/4. Though maybe when Leica finally get around to CLA my lens that might all change.

 

At F1.4 the Summilux is questionably sharp - for some people, sharp enough. For other people, it's soft. I've owned several 75 Lux's and at F1.4 it can be difficult to see where the sharpest point is. Not always, but sometimes. At F2 the Lux sharpens up to a degree and the central area will look sharp (by most people's standards). At F2.8 the Lux is sharp and the spherical aberrations are gone (from the plane of focus). Bokeh, especially at F1.4 to F2, is what I'd term as "Mandler" - moderate gaussian blur, but a heavy dose of spherical aberrations. The look is more 2-dimensional (but this is also influenced by lighting, so I'm generalizing).

 

The 75 APO is very sharp at F2 at the central region. There is little doubt about where the focus plane was placed. Bokeh at F2 can be quite smooth. The F2 images are more contrasty than the Lux images, so the subject tends to "pop" more. Comparing the draws to Zeiss lenses, I'd say the 75 Lux is more of a Sonnar draw and the 75 APO is more of Planar draw. The 75 APO isn't so happy when shooting into the sun or the sun (light) is just outside the frame. There can be veiling (loss of contrast) and various degrees of flaring. Not that the 75 Lux is any better managed in that regard, but I think people have higher expectations for an APO lens. The 75 APO is a bit of a one-trick pony --- I like its draw at F2, but beyond that its draw is common. As the aperture blades are closed down, the specular highlights can have the "ninja star" shapes.

 

Both lenses can be difficult to shoot. The 75 Lux isn't so sharp wide open, so there's that + focus shift to contend to with. I've owned multiple 75 APOs and their focus calibration is elusive. My current 75 APO was just sent to Leica for calibration (again). I had a 75 APO several years ago that I felt focused very well. I sold or traded it for something - foolish move in hindsight. The 75 APO is a lens I highly recommend trying out in person to make sure its focus matches the camera it will be used with. There are no local M dealers in our area, so I'm stuck with the buy & try (and pray) model.

 

Generally, I prefer the 75 APO over the Lux. The APO is smaller and noticeably lighter. Its minimum focus distance is just .7 meters, so it's also good for close-up shots. Not macro territory, but still pretty close (in terms of a magnification ratio). Its colors are bolder; the 75 Lux coloring is more muted, more subdued. Of course this can be edited one way or the other in post.

 

I sent a 75 Lux to Leica NJ late last year or early this year, their "CLA" was a $435. The lens needed focus calibration and had also developed a slight barrel wobble. I asked them to "optimize focus" F2 at 2.5M, but I don't think they did that. Pretty sure they "optimized" it for F1.4 use.

[lct]

Any DoF table makes assumptions regarding the factors which influence 'sharpness' - they should be considered as guidance rather than anything else. They also assume a flat field and lack of imperfections

Sure but whatever CoC value we choose, for whatever reason, two lenses will always have the same DoF all things equal i.e. with the same aperture, focal length, subject distance and CoC values as i suggested above quite expectedly. Bokeh is linked to DoF of course but it is another story given its subjective aspects which your Zeiss reference (any link?) should confirm if need be.

[Advertisement]

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

[John Black]

Most likely referring to this paper published by Zeiss. Not light reading...

 

Zeiss Bokeh and DOF White Paper

[Michael Geschlecht]

Hello Cheshire Cat,

 

My writing has not been directed to or at anyone's opinion. Everyone is entitled to their own personal opinion. I have been writing about facts. People are not entitled to their own personal facts. Facts are facts.

 

The depth of field for different lens designs of the same focal length, focussed on the same point, of the same subject, is not always the same when the different lenses are set to the same aperture: If depth of field is held to a single set of criteria. Here, the criteria being used is a circle of confusion of 1/30 mm.

 

What we are discussing is: At what point does the APPARENTLY sharp image in a photo cease being considered to be a sharp image when comparing photos of the same subject taken with different lenses of the same focal length set at the same aperture?

 

When a lens contains a(n) floating element(s) this introduces a new set of variables into the mix. Generally this means that the correction of abberations at the plane of focus will be higher at the expense of a more rapidly deteriorating image structure as you move further away from this more highly corrected image plane.

 

Or to put it another way: Less depth of field at the same aperture with a lens of the same focal length focussed to the same point on the same subject.

 

Best Regards,

 

Michael

[pgk]

At F1.4 the Summilux is questionably sharp - for some people, sharp enough. For other people, it's soft. I've owned several 75 Lux's and at F1.4 it can be difficult to see where the sharpest point is.

Exactly. And if you can't precisely see where the 'sharpest' point actually is, how can you assess DoF? And this is exactly how I found my 80 'lux to behave.

Sure but whatever CoC value we choose, for whatever reason, two lenses will always have the same DoF all things equal i.e. with the same aperture, focal length, subject distance and CoC values as i suggested above quite expectedly. Bokeh is linked to DoF of course but it is another story given its subjective aspects which your Zeiss reference (any link?) should confirm if need be.

Yes, but to repeat, the equation is simplistic and doesn't take all variables into account and is only valid for 'perfect' lenses.

[thighslapper]

How many angels can you get on the head of a pin ? ........

 

This is fast degenerating into a classic 500 post Leica discussion .....

 

The poster originally voiced a 'subjective' opinion.......

 

Appreciation of the shallowness of DOF is usually gained by the quantity of duff OOF photos you have ........ and that in turn is a product of the difficulty in focussing with the lens and combined RF and lens calibration.

 

In my experience the 75/2 ranks close to the 50/1.4 in miscalibration stakes ..... presumably by virtue of the floating elements ..... mine is just back from Solms/Wetzlar for the second time and is finally ok for close and at infinity ........ I sincerely hope it is ok at intermediate distances ..... but I'm not holding my breath ...

 

If the bits you intend in an image to be in focus are ..... I'm not sure that you pay much attention to the DOF at all .... or for that matter can easily compare it between lenses..... subjectively or otherwise .....

[Michael Geschlecht]

Advertisement (gone after registration)

Hello Paul & LCT,

 

Once again, you are certainly allowed to have your own personal opinions about things but you are not allowed to have your own individualized personal facts to support that opinion.

 

Facts are facts:

 

Sometimes lenses don't follow general assumptions.

 

1. Lenses of the same focal length set to the same aperture focussed on the same object can have different depths of field dependent on lens design.

 

Example: 75mm Summarit F2.5 focussed at Infinity. Look at the depth of field scale on the lens at F16. it is 8 meters to Infinity. Compare this with the 75mm Summilux focussed at Infinity. Look at the depth of field scale at F16. It is 10 meters to Infinity.

 

2. Different lenses of the same focal length extended to the same Image : Object Ratio and set to the same aperture can require different exposure factors dependent on lens design.

 

Example: 135mm F4 Elmar extended to 1 : 1 requires an exposure factor correction of approximately 4X the indicated exposure for any aperture set. The 135mm F4 Tele-Elmar also at 1 : 1 requires an approximately 6.5X exposure factor correction at any aperture set.

 

3. The same lens focussed at a different distance can change its effective aperture without anyone adjusting its aperture dial with certain lens designs.

 

Example: 280mm APO F2.8 is a F2.8 lens when focussed to Infinity. When it is focussed at its closest setting the F2.8 lens becomes an F2.2 (two point two) lens.

 

Optical rules, whether depth of field, exposure compensation for closeups, or exposure compensation in general circumstances, are correct in their standardized forms for many lenses in many situations. They are not correct for all lenses in all situations.

 

I hope this has been helpful.

 

Best Regards,

 

Michael

Edited September 15, 2014 by Michael Geschlecht

[lct]

Sorry Michael but i am not convinced about 75mm lenses for the reasons suggested above and adding variables is not the easiest way to compare DoFs or anything else IMHO. If the full aperture of a f/2.8 lens is not actually f/2.8 but f/4 for whatever reason it is then an f/4 lens that will have the same DoF as another f/4 lens of same focal length other things being equal that's all. Now i won't bother you and other LUF members by repeating the same things ad nauseam so let's agree to disagree on this point if you don't mind.

[jankap]

+1

Jan

[pgk]

A few centuries ago the world was flat. Everyone knew that, except.....

 

We now apparently know that despite variables such as differing lens design, image flatness, aberrations, aspheric elements, diaphragm shape and more, that all lenses of the same focal length produce identical depth of field at the same aperures. Don't we;)?

[lct]

Not if your variables produce significant differences in focal length and/or aperture of course but when comparing high quality lenses like Leicas we should get basically the same results IMHO. I'll try to retrieve some Leica's technical datas of same focal length M and R lenses and post them here for those interested.

[Peter Branch]

A few centuries ago the world was flat. Everyone knew that, except.....

 

We now apparently know that despite variables such as differing lens design, image flatness, aberrations, aspheric elements, diaphragm shape and more, that all lenses of the same focal length produce identical depth of field at the same aperures. Don't we;)?

 

Of course we know that - we read it on the Leica Forum - must be true.

 

Except, of course, that its a load of rubbish. The differing image "quality" obtainable by using different lens designs has been known about since the dawn of the photographic age.

 

My concern is that, with some honourable exceptions, there are too many contributors who seem unwilling to learn anything from the contributions made by members who do actually know a bit about this subject - or indeed from the literature such as the excellent Zeiss paper on the subject.

[Peter Branch]

Not if your variables produce significant differences in focal length and/or aperture of course but when comparing high quality lenses like Leicas we should get basically the same results IMHO. I'll try to retrieve some Leica's technical datas of same focal length M and R lenses and post them here for those interested.

 

I suggest you start by looking at the published DoF data for the "R" and "M" lenses with relatively short focal length - its quite instructive.

 

You might also try looking at the data for different "M" lenses of the same focal length at the same aperture even if the maximum apertures are different. They are all available on the Leica website as the last pages of the technical data for each current lens.

 

You will observe that they are different - mainly, but not exclusively, because of the differing lens designs - Leica actually know a thing or two about these things.

Edited September 15, 2014 by Peter Branch
Missing words

[lct]

You might also try looking at the data for different "M" lenses of the same focal length at the same aperture even if the maximum apertures are different. They are all available on the Leica website as the last pages of the technical data for each current lens.

I will begin by lenses of same aperture and focal length if you don't mind but i'm pretty sure that results will be the same roughly.

[pop]

There's no need for sarcasm, I think.

Depth of field (as marked on the lens barrel) is reckoned from the diameter of the "circle of confusion". This is but one of several parameters which contribute to the overall image quality. It just says how large a point may be in the image so that we still accept it as a point and not as a disc. Obviously, whether any individual photographer will find an image sharp enough if significant parts of the subject are near the borders of the depth of field will depend on the preferences of the photographers. That has been so since the invention of the term DOF.

Lenses of the same focal length have the same depth of field at the same aperture, provided their (technical) design is the same, I think. It's all defined by the angles of the light rays which meet at the image of any given point.

Whether an image "looks" sharp to any given viewer under any given magnification depends on many factors; the thickness of the points and lines in the picture is but one of them, albeit a prominent one.

The system would break down, of course, if a lens was incapable of producing points which are small enough even in the focus plane. A new term would be needed, perhaps "depth of blur".

[pgk]

Can I throw another spanner into the works here? Out of curiosity, why should an aspheric element help in lens design and how does a design incorporating an aspheric surface, or indeed several, differ from one which is based on purely spherical elements?

 

I have some observational data myself (and have discussed this with others, some of whom have considerable expertise/experience in lenses) but am curious as to what people think the effect of such a lens surface actually is.

[lct]

50/1.4 M and R lenses to begin with. Next ones will be 35/1.4s if i can retrieve them.

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!

[01af]

First of all, depth-of-field most definitely is nothing but physics. No magic, sorcery, or witchcraft involved.

 

And then, for any given maximum circle-of-confusion, depth-of-field depends on many factors. The most obvious are focal length, distance, and aperture. Second-level factors include vignetting, field curvature, and pupil magnification. In particular, when a lens has some vignetting (as most lenses have) then depth-of-field won't be even across the frame; instead, it will be narrower at the center and wider near the edges. Third-level factors involve things like the degree and the characteristics of the residual spheric aberrations, and possibly more.

 

Those 2nd- and 3rd-level factors are the reason why two different lenses of the same focal length can yield slightly different depths-of-field, even when set to the same aperture and focused at the same distance. Depth-of-field calculators take only the 1st-level factors into account and ignore the others ... simply because the others usually are unknown to the user. Still another factor, by the way, is the simple fact that most lenses have true focal lengths that are slightly different from their nominal values.

[lct]

35/1.4 M and R lenses. My plea is over... for now .

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!

[Advertisement]

Hallo,

Look here: Leica M

[Peter Branch]

First of all, depth-of-field most definitely is nothing but physics. No magic, sorcery, or witchcraft involved.............

 

You are correct - the properties of the ariel image produced by a lens are totally and solely determined by physics and differ according to the design often in highly complex ways.

 

(As an aside the DoF of wide-angle lenses may, or may not, increase towards the edges. Many, (actually most), modern lens designs significantly increase the pupil magnification at wider angles which tends to decrease the DoF - yet another higher order effect.)

 

The ariel image is projected onto a sensor. In the days of film this could have a noticeable effect on the way that image was captured and recorded. Claims were made that some films could "increase DoF". These were usually those that rendered everything slightly blurred / diffuse / atmospheric - pick a word!

 

With digital sensors this sort of thing has greatly decreased but does still occur as one of the implied claims made both for and against increasing the number of pixels. This is still just about within the realm of physics.

 

The resulting digital file is subject to all sorts of manipulation and transformation with perhaps the most significant being sharpening. It is highly instructive to use the sharpening tool in CS6 on the parts of the image that are clearly, but only just, out-of-focus. Suddenly the observed / perceived DoF can be significantly increased.

 

At this point I think one has to accept that the physics is no longer the sole determinant and that visual and perception effects become significant. These are just as complex as the physics of the lens design - but much less quantitative.

 

My personal view is that digital sensors and very highly corrected lenses have rendered most, if not all, of the theory of DoF based on a given size for the "Circle of Confusion" much less useful. Under some conditions, with some subjects, and with some lenses, which include the 75 APO, the plane of sharp focus can be brutally clear, particularly in large prints, and everything else is very obviously less sharp. Thus the original premise that the DoF defines a quantifiable range over which the sharpness will be indistinguishable from that of the actual plane of focus breaks down. The observable DoF under these conditions approaches zero even with wide angle lenses.

 

The OP does not make specific whether the DoF differences that have been detected and are the subject of this thread affect the quality of images or makes focus accuracy much more critical. These are obviously related topics but they are not the same.