Osuna breaks into the big time on LuLa

[ho_co]

Advertisement (gone after registration)

I would think this had already been posted but didn't see it at a quick glance.

 

Forum member Rubén Osuna et al have published on LuLa a very nice article covering a number of factors that distinguish the M8 from other top-line instruments:

 

Luminous landscape

[Advertisement]

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

[Riley]

never saw it before myself, truly a worthwhile read

the most explicit explanation of the optical criteria i have yet seen

[marknorton]

An excellent piece of work, it's very welcome to see such a lucid explanation of the various design issues.

[lct]

...Forum member Rubén Osuna et al have published on LuLa a very nice article covering a number of factors that distinguish the M8 from other top-line instruments...

I don't know where they found out their "typical circle of confusion of 33 microns" for film but i agree that "the circle of confusion of the Leica M8 is 1.33 smaller than the circle of confusion of a film based Leica M".

That is was i tried to explain, Howard, remember?

Anyway, i'm happy to see that they advise to use the DoF markings of the nearest faster f stop as i suggested myself.

Who says thank you LCT?

Just kidding, Ruben, happy to have been useful if any.

[lars_bergquist]

I did not like the part about DoF. Here's why:

 

In order to be seen as sharp, an image must resolve at least the amount of detail that the eye does. There is an old rule of thumb that says that detail 1/10 of a mm apart must be clearly resolved in the print, when seen at optimum distance (which is usually c. 30 cm, i.e. normal reading distance -- we try to read at that distance because it -is- optimal!). So, a neg which was to be printed contact, i.e. same size, had to offer resolution of that order. It is a fact that many older MFroll film cameras had DoF tables computed for a circle of confusion (CoC) of that value. It was Leitz (probably in the person of Max Berek) who introduced DoF tables and scales based on 1/30 of a mm, and this has been the industry standard for 35 mm cameras for half a century now.

 

It seems that Berek based his CoC on the assumption that the tiny Leica neg would be enlarged to a size to be comparable to the prints produced by the major competitor. And that in the 1920's was a roll film 6X9 cm camera -- and most amateurs even around 1950 were satisfied with contact prints from that size! That meant an enlargement of 3X, giving a print of around 7X11cm.

 

This standard is of course totally obsolete now. Nobody has however dared change it. I would really discourage anyone from basing any shooting on it, unless you want to limit yourself to print sizes smaller than the corner-lab's standard of 10X15 cm or 4X6". I find that many people are puzzled by (often) unsharpness in the far distance, and as they have based their focusing on the supposedly bomb-safe DoF scale, they start hypothesizing about an alleged inability of wide angle lenses to draw sharp detail at infinity!

 

This computation of course is true of the M8 sensor size too, and in spades. My own practice is to halve the aperture values, so that a hyperfocal distance for 1:11 is actually based on the scale values for 1:5.6. I recommend it.

 

The old man from the Age of Measuring-tape Focusing

[rosuna]

THANK YOU LCT!!!!!!

 

You were right!!!

 

I only made an explicit analysis adapted to the case of the M8. The original one can be found here (by Bob Atkins):

 

Digital Depth of Field

 

and here:

 

Photo.net Depth of Field and the Digital Domain by Bob Atkins

 

The 33 microns CoC is quoted in Erwin Puts' site:

 

Kodachrome

 

... it is a conventional variable, in any case.

[lct]

Advertisement (gone after registration)

THANK YOU LCT!!!!!!...

Ruben:

[lct]

Lars, usual CoC values are still used on current DoF markings and they work don't they, so how could they be obsolete in the same time?

[lars_bergquist]

Lars, usual CoC values are still used on current DoF markings and they work don't they, so how could they be obsolete in the same time?

 

As I wrote, they ARE obsolete -- unless you restrict yourself to very small size prints. For the conventional value of the circle of confusion is based on an assumption of the final print size. For instance, 1/30 of a mm or 33 micron is OK for a 7X11 cm print. But if you go for an 18X24 cm or 8X10" print, you enlarge not only the image. You will also enlarge small amounts of previously invizible fuzz, until it is no longer invisible.

 

In other words -- they are obsolete, because they DON'T work.

 

The old man from the Age of Guess Focusing

[sean_reid]

Enjoyed the article Ruben, congratulations.

 

Cheers,

 

Sean

[lct]

...they are obsolete, because they DON'T work...

Well let's say that they don't work for you and they do for me then.

[Woody Campbell]

Great article. I agree with Lars on the 2-stops issue that given the size prints I'm making now - this works very well for me.

 

There is an equivalent issue on the rule of thumb on minimum shutter speed (1/focal length): It continues to work well in digital age (better for Leica M shooters, worse for SLR shooters) as long as you use the 35mm equivalent focal length. If you are printing larger prints it's best to add a stop as a safety factor. Again, this works well for me.

[carstenw]

As I wrote, they ARE obsolete -- unless you restrict yourself to very small size prints. For the conventional value of the circle of confusion is based on an assumption of the final print size. For instance, 1/30 of a mm or 33 micron is OK for a 7X11 cm print. But if you go for an 18X24 cm or 8X10" print, you enlarge not only the image. You will also enlarge small amounts of previously invizible fuzz, until it is no longer invisible.

 

In other words -- they are obsolete, because they DON'T work.

 

The old man from the Age of Guess Focusing

 

Lars, you don't view a large print from the same distance as a small print, so that ought to take care of that argument. The viewing distance is roughly proportional to the size of the print...

[sean_reid]

My own feeling, FWIW, is that there are so many variables that can come between the capture of the subject and the final print that any kind of DOF scale is a very rough estimate that may or may not apply for a given photograph. Among other things, the definition of what is considered "sharp" is always somewhat arbitrary even when its been conventionalized. I'm in favor of empirical experience as the best way for a photographer to get a feel for how DOF plays out with his or her various subjects, lenses, cameras, print sizes, etc. I appreciate that what LCT and Ruben have tried to do, however, is to clarify what might be a starting point worth considering.

 

Cheers,

 

Sean