SL lenses and DoF

[FMB]

Advertisement (gone after registration)

I don't know if this topic has already been treated but in any case....I ask myself which CoC value is used to work out the distances given us on the top screen  of our SL2 and SL2-S showing the DoF that our actual settings determine in each occasion? This value should be different according both camera sensors and beetwing the several focal distances. For each camera and for each lens? Is this value conservative or commercial?

Best regards,

FMB.

[Advertisement]

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

[mike3996]

I am not the expertest man on this subject but it is my understanding that (in simplistic calculations/approximations) CoC is only tied to sensor size, so all full frame sensors can share a same CoC number. Focal length or subject distance plays no role in that approximation.

That's the simplified version.

The complicated answer is that CoC is actually calculated using these following datums: sensor size + (pixel pitch?) + print/output media size + viewing distance of the finished product.  If you make huge prints and give viewers magnifying glasses to inspect details, your CoC has to be very large. On the other hand if you make photographs for mobile phone/instagram consumption you could use very small CoC values. *

In film times the CoC values and DOF scales might have been calculated with 4x6 prints in mind, today with pixel peeping photographers and whoever, the CoC values are very conservative, to the point of making DOF scales unusable.

 

 

*) or small/large, I might have mixed the directions.

[FMB]

But our beloved Leica friends what have they decided do?

Francisco

[UliWer]

Well, there are different ways to find out:

1. Taking a photo at the limit of the indicated DOF: if it is rather sharp, they use a conservative value for the circle of confusion; if it is rather unsharp, they use the traditional one which comes from the times of Max Berek and is not really usuable today.

2. The technical data for Leica-M lenses show graphs for the - nominal - DOF. These graphs are based on the old CoC defined by Max Berek. Does the DOF indicated by the SL differ from these M-graphs? 

If my memory about what Mr. Karbe explained about the 35mm Apo-Summicron SL some weeks ago in Wetzlar, serves me well, he said that the nominal DOF of all the lenses (including the SL lenses) are not realistic today any more.

[FlashGordonPhotography]

It is a myth that pixel pitch affects CoC, until individual pixels are visible at the *print* size used to determine effective DoF. Some smarty came up with the idea that CoC should be the size it takes to make a grid of 9 pixels (a digital circle). That’s not reality. DoF was calculated on what you can see not what a sensor can do. Effective DoF is usually calculated on viewing a 10(12) x 8 print at 12 inch viewing distance. Both the SL2 and SL2S have more than enough resolution that DoF calculations will be identical, because at the size used to calculate DoF your eyesight is the limiting factor. Even at larger print sizes there’s enough resolution (and print dithering) that the eye is the limiting factor as you’ll need to move away from the image to view it at the same angles.

Then again DoF itself is barely an estimate based on what a few grumpy old white men saw in a room over a hundred years ago. It’s not an absolute and need to be interpreted far more widely than a predetermined number. DoF isn’t science. It’s an estimate that forgets to take into account many visual factor that will affect how blur is perceived in a viewed image.

But since you have started a thread on DoF, I will say hi, in about another 6-10 pages.

Gordon

p.s. I’m still trying to work out why people calculated estimated DoF at 100% in Lightroom to then post the final image on instagram……

[FMB]

Ironic Sir, in Spain we say  we are talkin of "cojonésimas" (smallest  part of cojon ="ball"). Even though I've the insane  curiosity about the figures of CoC they are using in this automatic system to give us with that simple way so complicated data. Thank you Leica!!!

No other photographic purpose.
Francisco

[FlashGordonPhotography]

Advertisement (gone after registration)

You could plug the figures the camera gives you into the formula. It should spit out the CoC Leica rely on.

Or email Leica directly. I find them responsive to inquiries.

Gordon

[Manicouagan1]

When I shot mostly with M lenses that have depth of field marking, I remember reading that a pretty good approximation of the practical  depth of field could be made by simply using 1 stop smaller the markings on the lens indicated was appropriate.  If that approximation is roughly correct and the value of 0.030mm was what was used to calculated lens markings then the one stop smaller approximate value for a new CoC is about 0.021mm (.03 divided by the square root of 2).  A similar value comes from using a Nyquist frequency for the diagonal of a a sensor with a 0.005 mm pixel pitch of (0.014mm).  Note all but the very best of our lenses have modest contrast at an MTF of 50 lp/mm(0.02mm).    It is too bad that the newest lenses don't have depth of field marks.

[jaapv]

On 11/17/2021 at 11:47 AM, mike3996 said:

The complicated answer is that CoC is actually calculated using these following datums: sensor size + (pixel pitch?) + print/output media size + viewing distance of the finished product.  If you make huge prints and give viewers magnifying glasses to inspect details, your CoC has to be very large. On the other hand if you make photographs for mobile phone/instagram consumption you could use very small CoC values.

This. Plus the tolerance level of the photographer/viewer for unsharpness. Plus the resolving power of the individual eyes of the user (AKA the competence of his Optometrist) Making DOF a subjective value which defies precise mathematical description and DOF scales nothing more than a guideline.

[BernardC]

On 11/21/2021 at 1:49 PM, Manicouagan1 said:

When I shot mostly with M lenses that have depth of field marking, I remember reading that a pretty good approximation of the practical  depth of field could be made by simply using 1 stop smaller the markings on the lens indicated was appropriate.

That's how I do it. Stop-down 1 as a rule of thumb, go 2 stops if I plan to print big.

It really depends on your needs. The M depth of field marks are probably fine if your images will be viewed on a phone, but you'll need more depth of field for larger output.

[Jean-Michel]

On 11/17/2021 at 4:11 PM, FlashGordonPhotography said:

It is a myth that pixel pitch affects CoC, until individual pixels are visible at the *print* size used to determine effective DoF. Some smarty came up with the idea that CoC should be the size it takes to make a grid of 9 pixels (a digital circle). That’s not reality. DoF was calculated on what you can see not what a sensor can do. Effective DoF is usually calculated on viewing a 10(12) x 8 print at 12 inch viewing distance. Both the SL2 and SL2S have more than enough resolution that DoF calculations will be identical, because at the size used to calculate DoF your eyesight is the limiting factor. Even at larger print sizes there’s enough resolution (and print dithering) that the eye is the limiting factor as you’ll need to move away from the image to view it at the same angles.

Then again DoF itself is barely an estimate based on what a few grumpy old white men saw in a room over a hundred years ago. It’s not an absolute and need to be interpreted far more widely than a predetermined number. DoF isn’t science. It’s an estimate that forgets to take into account many visual factor that will affect how blur is perceived in a viewed image.

But since you have started a thread on DoF, I will say hi, in about another 6-10 pages.

Gordon

p.s. I’m still trying to work out why people calculated estimated DoF at 100% in Lightroom to then post the final image on instagram……

Yep!

Still on page 1!

For fun, I looked up the subject in my first year photography school manuals: Langford Basic Photography, (1967) and Ilford Manual of Photography (1968). Essentially, the eye of the average viewer "…  cannot see any difference between a circle of confusion and a true point if the diameter of the circle is less than one-thousandth of its distance to the eye. To put it in another way, a print will appear perfectly shaped if the circle of confusion corresponding to each and every point on the subject subtends at the eye an angle not greater than 1 in 1000, i.e,. 3 minutes of arc. This is equivalent to circle one-hundredth of an inch in diameter at 10 inches, the nearest distance of distinct vision …" (page 98, lford Manuall of Photography, 1968).

On page 100, there is information about conditions for "correct" and "comfortable" viewing. :The "correct" viewing is the focal length (f)  times the degree of enlargement (e). In that case the permissible circle of confusion in print is:  f x e /1000. The "comfortable" viewing or minimum viewing distance is the diagonal of the print, in which case the permissible diameter of the circle of confusion is 1/1000 of the diagonal of the print.

The scales on my 35 mm version 1 summicron and 50 mm 'rigid' are exactly the same as the on my current versions of both lenses. The scales were definitely calculated using a  method similar to those in the Ilford an other manuals of decades ago. Even in film days, I tended to the one stop wider range for an approximation of the depth of field. Focus on th energy object, focus on the far object, find that range and set the f/stop that should work; or just focus on the main subject and let the rest fall where it may.

The final proof is in the print, and only in the print, because if you do not print it, it does not exist. 

Then, there is depth of focus 😀!

[SrMi]

Has anyone measured if the numbers on the top LCD are correct/usable? I did some measurements with laser distance meter, SL2S says 2.8m, laser says 2.2m.

[FlashGordonPhotography]

4 hours ago, SrMi said:

Has anyone measured if the numbers on the top LCD are correct/usable? I did some measurements with laser distance meter, SL2S says 2.8m, laser says 2.2m.

Good question! I haven’t. Mostly because I don’t rely on that feature. That’s because I only have three apertures I use. Thin, somewhere in the middle and lots. That’s been enough to support me for the last 30 years….

But maybe I’ll have a look, just to see.

Gordon

[beewee]

6 hours ago, SrMi said:

Has anyone measured if the numbers on the top LCD are correct/usable? I did some measurements with laser distance meter, SL2S says 2.8m, laser says 2.2m.

I can’t speak for close focus distance but on the SL2-S the far focus distance when it shows infinity, I find things at infinity (i.e. stars) are sharp.

[FMB]

Following your advises I've tryed to compare the figures of top of cameras screen with the usual mathematic ecuations and did not have any success  and consistency with the results. They apparently say those Leica used seem are not superior at one CoC  of 0,020 mm.???

The  "aficionados" as me are not actually much worried by this parameter but, I insist, I would like to know which are the figures uses Leica in this case if the CoC has been taken in account. I think of the very real possibilities we have to use every day more big apertures in our photos as the new SL Summicrons permit. The accuracy of their thinest DOF is very relevant.

Francisco