DOF calculators and M8

[wstotler]

Advertisement (gone after registration)

If I input the M8 (which applies the 1.3x crop factor) into a DOF calculator, are the results for a given lens accurate? E.g., crop factor + 35mm lens + inputted distance = valid result. Or, do I need to select the focal length of the 35mm lens as mounted on the M8 (46mm) to get an accurate DOF result? I think it's the former, but can someone confirm? Thanks, Will.

[Advertisement]

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

[2Bébèrt]

first one is correct. The calculator also should use the circle of confusion for the different camera types.

 

Personally I use: Online Depth of Field Calculator

 

There is also an iPhone widget but the M9 isn't included yet.

[jaapv]

I have found over these last years that it is unwise to trust the concept of DOF at all. In a digital world, with lens quality at the current level, it is safe to assume that the only really sharp part of your image is the plane of focus. Personally I find the alternate concept of DOF and hyperfocal distance as expounded here: http://www.trenholm.org/hmmerk/DOFR.html much more useful.

[okram]

They are accurate for a certain enlargement.

 

THIS is a calculator: Schneider Optics, under no.5, there is a link for download the calc.(excel)

[Philippe D.]

Hi Will.

do I need to select the focal length of the 35mm lens as mounted on the M8 (46mm) to get an accurate DOF result?

According to Peter Karbe (see: http://www.photoscala.de/node/7889/pdf), the Summilux 1.4/35mm behave on M8 like a 1.9/46mm.

But as jaapv state, on the M8 sensor (with a circle of confusion of 0.021mm), the really sharp part is the plane of focus.

[!Nomad64]

Spare yourself a headache. There's an interesting article at Luminous Landscape where they treat the DOF issue with M8: Luminous landscape

 

The point is:

 

quote

 

the differences in hyperfocal distances depend on the diagonal differences of the sensors (the so-called crop factor), and 1.33 allows for an easy rule: use the marks of the immediately lower f number

 

unquote

 

Cheers,

 

Bruno

[jaapv]

Advertisement (gone after registration)

Yes, but who takes a photograph assuming you'll never want to enlarge beyond a certain size. What if you decide to exhibit at a later point of time?

[SJP]

Photo's at an exhibit wil hopefully be viewed at a normal distance (1x the diagonal) and then the old rule is still fine, maybe adjust by one stop just to be on the safe side.

 

The main problem is that most people have very big screens that are too close to the eyes (because you forgot to buy a bigger desk for the biggerscreen - this includes me by the way). In addition pixelpeeping is available at a click of the mouse & becomes second nature. This leads to the false impression that the old rules don't apply. The rules are just fine - only the expectations have changed.

[jaapv]

Yes, that is clear, Stephen. Still, the different rendering of the oof areas by sensors as opposed to film and the changes in lens design since the 1930ies strongly suggest that the rules should be at least reevaluated critically. Another aspect, which was totally new to me and that I find highly interesting is this article DOF Revisited , which was posted in another thread. I find the different approach in it highly compelling. Could you please give your opinion?

[jbgeach]

It is important to understand that the DOF tables are for 4x6 images, if you want to go bigger, go down another stop. If you want to go really big go down two. However, if there is time, just focus on what you want clear.

[wstotler]

Thanks, everyone. Some great thoughts--very appreciated. Answers below.

 

 

first one is correct. The calculator also should use the circle of confusion for the different camera types.

 

Thanks for the confirmation--exactly what I was asking.

 

 

They are accurate for a certain enlargement. THIS is a calculator: Schneider Optics, under no.5, there is a link for download the calc.(excel)

 

I've seen and used that before, but thanks for reminding me of the source.

 

I'm currently using an iPhone application called "myPhotoDOF" because it has a visual representation of the DOF that's really easy to view.

 

What I'm going to do is cross-check results from the iPhone application against the spreadsheet (and a few other calculators) just to make sure it's providing results that I'd expect.

 

The reality is that I'm not overly picky about DOF *except* in the rare case when I have two-or-more people that I need to get into a shot, with faces in focus. This can be people side-by-side (but not directly on the focus plane) or many people in two or three rows (a group shot). I could just flip to f/8 or even f/11, but even then I have to determine how far back I have to be so that the DOF will be deep enough to give me coverage. I want just enough DOF to include the subjects' eyes/faces/ears, but not much more.

 

 

It is important to understand that the DOF tables are for 4x6 images, if you want to go bigger, go down another stop. If you want to go really big go down two. However, if there is time, just focus on what you want clear.

 

1.33 allows for an easy rule: use the marks of the immediately lower f number

 

Thanks. I generally use the "two stop" rule of thumb if I want to go 8x12 or larger and printing is intended at the time of shoot, especially at close focusing distances. I do tend to shoot wide open, so this means backing off to f/2.8 or f/4, respectively.

 

 

<snip> This leads to the false impression that the old rules don't apply. The rules are just fine - only the expectations have changed.

 

This is a very interesting point and true. I appreciate you bringing this up. I think the old rules are still "valid." But I do think they aren't really relevant as expectations *have* changed.

 

But. . . . there's this: In practice, for 4x6s the old rules absolutely apply. Shooting good photos and leveraging awesome Leica glass leads to kick-ass 4x6s that blow away point-and-shoot 4x6s. 4x6s? Who needs those? Well, 4x6s are still relevant because for the family, 4x6s are still the "standard." For onscreen viewing, also, at smaller sizes (like for E-mail), I also think the rules apply--smaller shots are like 4x6s. Again, the M8 + Leica glass blows away my "local" competition of POS cameras (and misused prosumer Nikons).

 

What I generally try to do, though, is shoot with an 8x12 expectation. A more stringent workflow can be applied to get me to a very credible 16x24. Generally, though, I'm shooting for 8x12.

 

The key thing your post points out for me is that I need to get a handle on the 16x24 "standard" requirements for DOF. As I think that is where my expectations really lie. Yes, it will vary depending on subject and intent, but I need to understand the rules well enough that I can get what I want out of 16x24 shots, with crispness, when that's desired. And with more deliberate intent than "dial in f/8." So, thanks for bringing it up in the context that you did.

 

 

I have found over these last years that it is unwise to trust the concept of DOF at all. In a digital world, with lens quality at the current level, it is safe to assume that the only really sharp part of your image is the plane of focus. Personally I find the alternate concept of DOF and hyperfocal distance as expounded here: DOF Revisited much more useful.

 

Jaapv--thanks. The article is very interesting and thank you for providing it. I can say that it's difficult for me to quickly apply the rules he provides in my head. And there's the issue for me of guestimating distances, also. It would be quite interesting to see the equation he provides rolled-up into an application where you can input some basic information (like with other DOF calculators) and then see graphically what's in and what's out, per his rules.

 

 

To put this in perspective, I ran across this over the weekend: "Back in the relatively early days of 1925 . . . my new small camera produced what I thought were some really remarkable results: pictures which enlarged up to 4 x 3 in. and were in some cases *almost* as good as those taken with my [quarter plate camera]." - H.S. Newcombe, 35 MM. Photo Technique (1946)

 

Cheers,

Will

[wstotler]

I need to get a handle on the 16x24 "standard" requirements for DOF. As I think that is where my expectations really lie. Yes, it will vary depending on subject and intent, but I need to understand the rules well enough that I can get what I want out of 16x24 shots, with crispness, when that's desired.

 

 

Conclusion

-------

The final CoC that I computed to use when calculating acceptably sharp DOF in 16" x 24" prints was .018mm instead of the M8's recommended CoC of .023mm.

 

Note: The .018mm CoC corresponds to roughly a 1.6x sensor crop factor (rather than 1.33x) in my iPhone application. I could choose "15 microns" (2x) or 19 microns (1.6x)--I went with the 2x crop factor ("15 microns") to be more conservative.

 

 

Assumptions

-------

* Final 16"x24" print; 60-degree diagonal angle of view = a minimum viewing distance of about 20" (50mm)

* "Perceived" target of 5 lp/mm resolution as viewed from 25mm (this appears to be a constant value measuring/describing targeted quality)

* Enlargement of 22.6x (M8 sensor is 27mm x 18mm; 24" is 610mm; therefore 610mm/27mm = 22.6x)

 

 

Calculation (per Circle of confusion - Wikipedia, the free encyclopedia)

-------

50mm / 5lp / 25mm / 22.6x = 0.01765 CoC or 0.018 CoC, rounded up

 

 

Expectation

-------

If I use the "15 microns" CoC setting (and focus appropriately) the DOF that's reported in my calculator should be about the DOF I see in my 16"x24" print, when viewed from about 20" away. Likewise, printing 8"x12" or 4"x6" shots should net really sharp images in the DOF areas, which will be perceived to be deeper than in the 16"x"24" print.

 

If anyone has any comment (or finds a math blunder), feel free to let me know. My math skills are often not the best.

 

Thanks,

Will

 

P.S. What about the "choose DOF then go one stop more" rule? It's fairly close--not exact, but close. In practice, I could set the M8 in the DOF calculator (23 microns), get the DOF on the calculator I want, then flip the aperture one stop more on the M8. (E.g., if the DOF I see on the calc requires f/4, I'd just bump the aperture ring to f/5.6 to get the equivalent of a "15 micron" CoC.) So, the rule of thumb is mostly true--but I like putting the "15 micron" setting straight into the calculator so WYSIWYG. I think I just smartly applied EV comp to my DOF calc.

[ho_co]

I have found over these last years that it is unwise to trust the concept of DOF at all. ...

Agree 100%. Best formulation I've seen yet on the topic.

 

1) The fact that the forum sees so many threads on depth of field is a good indicator that there is more to it than a simple calculator can cover.

 

2) The fact that DoF arises in the mind, and not in the physical world, also points that direction.

 

3) The fact that most (all?) manufacturers have quit putting DoF markings on their AF lenses indicates IMO that the concept may have outlived its usefulness.

 

4) Considering the number of options available on "modern" cameras and the ease of incorporating hyperfocal calculations automatically, the fact that no manufacturer has put a "hyperfocal focus" mode on his cameras is another indicator that the concept is somehow deficient.

 

5) DoF still made some sense with color film, where the differing focus of various colors on various film layers could help extend apparent sharpness. With digital, this "depth of focus" can no longer be used to the same effect because of the plane sensor.

 

6) Every(?) article on DoF points out that its underpinnings are today out of date, but honored by tradition.

 

7) Nevertheless, the Merklinger article Jaap mentions (DOF Revisited) was written before the popularity of digital.

 

8) Following Merklinger, Gary Ferguson pragmatically demonstrates in his LuLa article Digital Focusing Part One that "nothing in photography destroys quality faster than departing from the plane of exact focus."

 

 

In other words, the usefulness of the idea of DoF had been waning as we made ever bigger prints from film and is doing so now even more quickly with the advent of digital technologies.

 

 

Although the formulas are arguably still as valid as ever and as much fun to program, I think the idea itself is becoming ever less tenable. Nice rule of thumb for small prints, but as Jaap said, "it is unwise to trust the concept of DOF at all."

[wstotler]

I wrote this for the Epson R-D1, but the its assumptions remain relevant for the Leica M8, albeit the crop factor of the Epson R-D1 is slightly more than the M8: Epson R-D1: Depth of field Note the link to the Zeiss publication under the heading "That's not the whole story" (see page 3 of the PDF).

 

Rich, that's a quite well-written, detailed article. (The supporting Zeiss PDF was interesting, also.) Wish I had seen the page prior to my math spree yesterday. Thanks for posting it originally and then linking to it above.

 

 

In other words, the usefulness of the idea of DoF had been waning as we made ever bigger prints from film and is doing so now even more quickly with the advent of digital technologies. Although the formulas are arguably still as valid as ever and as much fun to program, I think the idea itself is becoming ever less tenable. Nice rule of thumb for small prints, but as Jaap said, "it is unwise to trust the concept of DOF at all."

 

Respectfully, Howard, I understand the sentiment but I don't agree. Just chucking DoF overboard seems to me like throwing out the proverbial baby with the bathwater. And, Jaapv, I also respectfully disagree about "trusting" the concept of DoF at all.

 

As I said above, I did look at the alternative Merklinger article, and I do believe it could work, but it's complicated to work out in my head. (For others, maybe it's a snap. For me, not so much.) If I had a "Merklinger" application on my iPhone, I'd sure give it a go. Additionally, selectively deciding what to sacrifice is something you have to do with the "old" concept of DoF anyway--where do you want to put it to get what you want from it?

 

Just so I don't seem to be contrarian, everyone appears to agree on the following. . . .

 

(1) The DoF scale markings on Leica lenses are "out of date."

- For me, this is a "yes" and a "no." I understand this sentiment, but DoF scale markings *are* accurate for 4"x6" prints. Everyone seems to agree on this and it's a caveat. They are significantly less accurate for 8"x10" prints. And they are disappointingly inaccurate for 16"x24" or larger prints. That's by modern standards and expectations. Is this a flaw? No. One just has to understand that they are "legacy" markings and apply to 4"x6" prints and adjust aperture accordingly as needed. If one is going to use them much at all. . . . I use them on the 18mm Super Elmar-M fairly regularly because I'm shooting that lens at f/8 and f/11, but not so much with my other lenses.

 

(2) DoF exists.

- Nobody has made the claim that DoF does *not* exist. The problem is with quantifying it in a predictable and useful way.

 

(3) Maximum sharpness in any image is where the focus plane was placed.

 

(4) When stopping down, the plane of maximum sharpness does not expand.

- When you stop down, you are expanding the "acceptably sharp" area *forward of* and *also behind* the plane of maximum sharpness.

- The "acceptably sharp" area that expands is *not* uniformly sharp. It becomes sharper when closer to the plane of maximum sharpness and less sharp further away from the plane of maximum sharpness.

- This is a key point that I understood but didn't completely get. Diagrams typically show DOF as a solid mass that expands when you stop down. They should instead show a single plane of maximum sharpness and a gradient-type foggy area moving away from that plane.

 

(5) The area of "acceptable sharpness" *forward of* and *also behind* the plane of maximum sharpness is *relative*--"acceptable sharpness" is based on subjectivity by the viewer.

 

I went through the basic math (post above) and came up with the "15 microns" circle of confusion (CoC) for use with my DoF calculator and the M8. (This is instead of the "23 microns" default for the M8.)

 

I will use this in my DoF calculator when I need "acceptable sharpness" to the front and rear of the plane of focus where maximum sharpness exists. I can dial in my focusing distance and estimate reasonably accurately how much DoF I'm getting--and adjust aperture accordingly so I can get more (or less!) DoF, based on what I need. (I don't need everything to be sharp--just what I want.)

 

The "15 microns" setting ensures that when I print at 16"x24" what I expect to be sharpest (plane of focus) is sharp and the "acceptably sharp" area to the front and back of the plane of focus is also sharp enough, based on standard viewing distance and the viewer's ability to resolve lp/mm.

 

Printing at smaller sizes will net even "more" DoF in the image.

 

I attached a quickie graphic to this post that I haven't seen before. I personally think this is an important concept that gets lost when the concept of DoF is graphically represented on the Web, in calculators, etc. (I also think it's at the root of a lot of disgruntlement about DoF.)

 

Thanks,

Will

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!

[footnoteblog]

Good job, Will.

 

None of this really matters unless you're shooting 4x5 with tilt-shift anyway. You know, real photography. *DUCKS*

[jaapv]

Will, I don't think we disagree at all. My gripe with the way DOF is presented is that users that have given less attention to the subject think that it is an absolute. The way you wrote it in your post concurs fully with my position. Sharp is in the plane of focus and everything in front or back of that depends on the individual level of acceptance. In my book that makes it subjective and thus not a rule you can trust. DOF is in the eye of the beholder.

[SJP]

+1 and yet it exists and is standardised, just like the metre, ampere, joule, erg, BTU, lbs. and all the other stuff we like so much.

[jaapv]

So it is, Stephen. I'll buy you one metre of beer next time we meet.

[Advertisement]

Hallo,

Look here: Leica M

[wstotler]

Good job, Will. None of this really matters unless you're shooting 4x5 with tilt-shift anyway. You know, real photography. *DUCKS*

 

Thanks, Jack. My 28mm f/2.8 PC-Super-Angulon-R and Novoflex M to R adapter is on order.

 

Jaapv--thanks for the clarification. I think we are in agreement except I do think DoF behavior is a constant and can therefore be "trusted" and leveraged by the photographer, provided that the photographer understands what he or she is dealing with from the shot through to the print.

 

Now whether or not the DoF can get the photographer what he or she wants, given the constraints of the environment and scene (not even mentioning the constraints of the diffraction limit), is another matter entirely.

 

Nevermind that a critically unsharp image of something really interesting is worth much, much more than a technically perfect, 16"x24" shot of something boring.

 

See attached image for clarification about how much issues like DoF really matter to the "real world."

 

Cheers,

Will

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'm with you, Will. We're on the same page.

 

I do disagree with your statement #2, "DoF exists." Technically, it doesn't exist (in the physical world, definable by formula), but is perceptual (arises in the eye and brain; the formula is approximative).

 

The formulas are still usable today, but as you say, they are less applicable with larger blow-ups. Film, sensor and lens resolution have made those enlargements possible, undermining the suppositions of the traditional values.

 

My point is simply that DoF is fuzzy and no one should take seriously a depth-of-field calculation without understanding its assumptions.