What is the biggest, perfectly sharp format you can get from your sharpest negatives?

[A49]

Try again
image on the left 28 mm lens
Image on the right 135mm lens
The hole in the film plae schematic represents the COC

cowanw - Nice citation.

 

[Ian Grant]

It does however work when the two lenses are used on different film formats:
For example, a medium-format 80mm lens at f/8 and a small-format 40mm lens at f/4 have the same opening (10 mm) and produce the same DoF, because the medium-format CoC is twice as large as the other.

There's always been an assumption that the CoC's used for calculating DoF for MF and Large formats should be larger, but that's only really true if comparable sized prints are being made.

Often people are using MF and particularly LF for very much larger prints, in my case I want a 24"x20" print from a 5x4 negative to look at least as sharp as a 10"x8" print from a 35mm neg at close viewing distances.

Back in the early 1970's I calculated my own DoF tables for my Mamiya lenses (MF) after finding that the published tables and also lens/camera markings for DoF didn't give that sharpness I wanted. I did the same when I moved to 5x4.

Another factor that's been missed is the quality & design of the taking and enlarger lenses. 4 element Tessar type designs show significant fall off in sharpness towards the edges and corners as they are opened up.

Ian

 

[Bill Burk]

Hi Ralph,

While it may seem like pointless circular reasoning, I see it more as coming around full circle.

I used your formulae to figure DoF for 40mm f/8 on a 35mm camera w/0.026mm CoC for a subject focused at 4.5m, and then to find the f/stop I would need on 4x5 with 135mm lens and w/0.1mm CoC to get comparable DoF. It is close to f/22.

I like to compare 35mm and 4x5 because I made the switch a few years ago to 4x5 and want to understand the differences I am discovering.

Ian, it's funny you mention Tessar's sharpness fall-off because in the darkroom last night I discovered my negative (that I've held in mind during this thread) is fully sharp to the edges where I thought I had a DoF falloff. I discovered a perfectly sharp branch where the test strip curled up in the air under my Xenar enlarging lens.

 

[Lee L]

For those trying to compare DoF among various formats, apertures, focal lengths, etc, this spreadsheet may be useful:

http://www.schneideroptics.com/software/DOF_Calculator.xls

You can easily copy the working section to have adjacent tables with different parameters for studying variations and differences.

Lee

 

[RalphLambrecht]

For those trying to compare DoF among various formats, apertures, focal lengths, etc, this spreadsheet may be useful:

http://www.schneideroptics.com/software/DOF_Calculator.xls

You can easily copy the working section to have adjacent tables with different parameters for studying variations and differences.

Lee

Lee

Thanks for the link. Looks like a good spreadsheet even that I find the mixture of metric and English units within one spreadsheet a bit odd. Also, the 'blur circle' (CoC) is a bit too liberal for my taste. These numbers aim for a maximum resolution of the human eye of about 4 lp/mm. Most eyes can do better than that and get to at least 7 lp/mm, and detection (not resolution) can easily go up to 20 lp/mm and more. A table with respective CoC values is attached.

 

[Lee L]

Lee

Thanks for the link. Looks like a good spreadsheet even that I find the mixture of metric and English units within one spreadsheet a bit odd. Also, the 'blur circle' (CoC) is a bit too liberal for my taste. These numbers aim for a maximum resolution of the human eye of about 4 lp/mm. Most eyes can do better than that and get to at least 7 lp/mm, and detection (not resolution) can easily go up to 20 lp/mm and more. A table with respective CoC values is attached.

Yes, the mixed units are odd, but not out of keeping with much of photography in that regard, especially in the US. The CoC values are suggested in F3:K3, but you have to enter whatever CoC value you want to use in the cell [B3] that's actually used for the calculations, so you're not stuck with the Schneider default. It's also easy enough to edit and apply a conversion factor so that results are in your preferred units.

I also find it nice that the spreadsheet comes from a source that most folks (I wouldn't dare claim 'everyone' on the internet) should find trustworthy.

Lee

 

[RalphLambrecht]

... I also find it nice that the spreadsheet comes from a source that most folks (I wouldn't dare claim 'everyone' on the internet) should find trustworthy.

Lee

The optical principle of DoF is well known, agreed and understood (by most). There are a number of equations floating around, but they usually return almost identical values.

 

[A49]

There's always been an assumption that the CoC's used for calculating DoF for MF and Large formats should be larger, but that's only really true if comparable sized prints are being made.

That is also my approach. If you calculate with small DoFs you are safer. And you can always print smaller to "get" additional DoF. If you want to go to the max in terms of resolution I find it useful to take typical resultion values for each aperture that fit your taking lens. If you have a very good lens then these values could be a good reference for the smallest in practice useful CoC:

f-stop --- resolution --- CoC
11 ------ 80 lpm ------- 0,013 mm
16 ------ 70 lpm ------- 0,014 mm
22 ------ 60 lpm ------- 0,017 mm
32 ------ 45 lpm ------- 0,022 mm
45 ------ 30 lpm ------- 0,033 mm
64 ------ 22 lpm ------- 0,045 mm
90 ------ 15 lpm ------- 0,067 mm
128 ----- 11 lpm ------- 0,091 mm
180 ------ 8 lpm ------- 0,13 mm
256 ------ 6 lpm ------- 0,17 mm

This is for a taking distance of at least 1:20 . If you have a smaller distance than the CoC values can be larger.

Best,
Andreas

 

[Athiril]

Nonsense. You better check the math!

See the focal length in there? Squared!

Sorry, but you are making several fundamental mistakes.

"The point is, the DoF is much larger with the shorter lens (see the numbers above). Using such a lens, one has a better chance to get much more acceptable detail (definition of DoF) into the image."

Your only test in this argument of DoF (dis)advantage is detail, which would be the level of detail, and the zone of focus which becomes the perceived DoF size. There is no clause of anything else needing to be the same, because they are not part of the test. Try to stop straw-manning.

There is nothing wrong with the '1cm' example, it demonstrates DoF changes after the fact. DoF relies upon 2 things after the fact.

1). The image reproduction, if it can actually reproduce the resolution gradient as it is, lower enlargement = flattening of this curve, since the reproduction (whatever method) can't hold as high potential resolution as bigger versions, this is demonstrated in my own image example.

2). The viewer being able to fully perceive the resolution gradient curve, the curve will flatten out for them the further they get away. You may need to be real close up forsome images.

Ralph's example has a high curve, and is missing part of the gradient, as some images do. Try a vertical object on a slight angular tilt stopped down with top and bottom of the object just out of focus.



You can't introduce extra caveats, DoF is defined.


f/x is aperture ratio, not aperture, because f/2 does not equal f/2, because f may differ, f may equal 50 and f may then equal 100, 50/2 does not equal 100/2, it is not the same aperture, but the sme aperture ratio. Aperture is an absolute measurement, not a relative one. Referring to aperture as f/x gives an unknown value for aperture if the person reading that does not have the value for f. The same aperture and focus distance will create the same area of whats in focus and whats not, if you take the same acceptable detail (lpph) into size, then you need to scale up the format too, especially since the argument is that shorter focal length is on a smaller format doesn't provide a DoF (dis)advantage.

To test this, all things must be equal, including the aperture(and not aperture ratio), focus distance, and format to format size ratio must be the same as focal length to focal length ratio. The argument is about DoF (dis)advantage, you do not need to go as far as I did to at least equal speed/sensitivity, it's not an argument about that particular merit.

The format to format ratio of 35mm:6x7cm (5.6x6.7cm) is about 2.02 diagonally, so it'll do.

f/4 50mm 35mm @ 2 meters focus;
close focus: 1.1818m
far focus: 2.222m
DoF: 0.403m

f/8 100mm 6x7cm @ 2 meters focus;
close focus: 1.812m
far focus: 2.231m
DoF: 0.419m

Gee whiz, looks the same to me. With a (neglible) advantage to the longer lens on a bigger format. Looks like shorter lens on a smaller format does not have a DoF advantage.


The sharpness is likely to significantly higher @ f/8 on 6x7cm than 35mm @ f/4, so you'll find that the near the DoF edge, the inner edges on 35mm will be softer, and likely could be below acceptable sharpness, thus reduce the effective DoF.

On top of this, sharpness of the 35mm system is going to have more effiency loss.

Lets both lenses are hitting 70 lp/mm at these settings, and our film is 60 lp/mm. Our total resolution is going to be 45.6 lp/mm.

That gives 1080 lpph (pairs) to the 35mm image, and 2554 lpph (pairs) to the 6x7cm image, or if you want to compare it diagonally since height is a different ratio to diagonal, then 35mm will have 1973 line pairs diagonally, and 6x7cm will have 3982 line pairs diagonally.

On film, we have room to stop down even more to have a much greater DoF than the 35mm image.

At the same focus image, if we stop our 6x7 100mm lens down to f/16, our DoF becomes 0.847m, the diffraction limit becomes 57 lp/mm, its likely to be close to the limit, so we'll say 50 lp/mm, coupled with the film resolution and loss, that puts us down to 38.4 lp/mm, this gives us 3353 line pars over the diagonal, still greater detail at twice the DoF size. To match the 35mm detail, we can afford to drop down to 22.6 lp/mm final resolution. At f/32, our DoF increases to 1.946m at the same focus distance our lens resolution drops to 28.5 lp/mm due to diffraction, our final resolution drops to 25.7 lp/mm, we still have more detail than the 35mm image, close to it, so we're a little above matching that level of detail, with a DoF greater than the 35mm images by a factor of about 5!

Let's change things, and say that you back peddle on the point and introduce new caveats the exposure has to be the same, we go back to f/8.

To match the 35mm detail level, with our lens at 70 lp/mm, including losses, we can still use a film as low was 24 lp/mm. In fact, at we can stop down to f/16 and can use a film thats 30 lp/mm to get the same detail level as the 35mm image and double the DoF.







So the advantage to DoF actually goes to bigger formats, longer exposures are simply not a factor in the argument of "The point is, the DoF is much larger with the shorter lens (see the numbers above). Using such a lens, one has a better chance to get much more acceptable detail (definition of DoF) into the image."

 

[Athiril]

That is also my approach. If you calculate with small DoFs you are safer. And you can always print smaller to "get" additional DoF. If you want to go to the max in terms of resolution I find it useful to take typical resultion values for each aperture that fit your taking lens. If you have a very good lens then these values could be a good reference for the smallest in practice useful CoC:

f-stop --- resolution --- CoC
11 ------ 80 lpm ------- 0,013 mm
16 ------ 70 lpm ------- 0,014 mm
22 ------ 60 lpm ------- 0,017 mm
32 ------ 45 lpm ------- 0,022 mm
45 ------ 30 lpm ------- 0,033 mm
64 ------ 22 lpm ------- 0,045 mm
90 ------ 15 lpm ------- 0,067 mm
128 ----- 11 lpm ------- 0,091 mm
180 ------ 8 lpm ------- 0,13 mm
256 ------ 6 lpm ------- 0,17 mm

This is for a taking distance of at least 1:20 . If you have a smaller distance than the CoC values can be larger.

Best,
Andreas

Your values are all exceeding the diffraction limit, and regardless these resolution figures wont be maintained unless the film is significantly (~3x) higher in resolution than the lens, there will be loss, the higher the resolution the greater the % loss of it will be (with the above caveat of film resolution).

 

[RalphLambrecht]

Sorry, but you are making several fundamental mistakes. ...

Remember, a man without data is just a man with an opinion. Show us your equation for DoF, and use it to prove that focal length has nothing to do with DoF, as you stated in post #172. I love to see it, and it will save you a lot of typing.

f/4 50mm 35mm @ 2 meters focus;
close focus: 1.1818m
far focus: 2.222m
DoF: 0.403m

f/8 100mm 6x7cm @ 2 meters focus;
close focus: 1.812m
far focus: 2.231m
DoF: 0.419m

Gee whiz, looks the same to me.
...

Yes, stopping down by two more stops (!) will give you the same DoF for the larger format. Hardly the same in my view, and by the way, I already said that in post #179. You may not have to reveal your equation after all. I see you use the same or one that is very similar to mine.

...So the advantage to DoF actually goes to bigger formats ...

Very creative conclusion. As I said all along, at the same f/stop, it's the other way around. Your numbers show that to be true as well. What are you arguing about? The difference between f/stop and aperture?

 

[RalphLambrecht]

...
f-stop --- resolution --- CoC
11 ------ 80 lpm ------- 0,013 mm
16 ------ 70 lpm ------- 0,014 mm
22 ------ 60 lpm ------- 0,017 mm
32 ------ 45 lpm ------- 0,022 mm
45 ------ 30 lpm ------- 0,033 mm
64 ------ 22 lpm ------- 0,045 mm
90 ------ 15 lpm ------- 0,067 mm
128 ----- 11 lpm ------- 0,091 mm
180 ------ 8 lpm ------- 0,13 mm
256 ------ 6 lpm ------- 0,17 mm
...

The top half of your list does not exceed the resolution limit. The bottom half 'rides' on the edge. However, your resolution numbers are achievable and realistic for good optics and contemporary fine-grain film.

 

[Ian Grant]

Sometimes the theoretical doesn't match the practice. Issues like diffraction limitations of smaller apertures which are always being widely quoted aren't borne out by the stunning prints made by many photographers at aperture that others theorise won't produce good definition/sharpness.

Issues like DoF at wider apertures become less important when lens definition often drops off at wider apertures because of design/type anyway.

There's far too much theory and not enough practice, it's not what's in spreadsheets, text books, manufacturers data that really matters to the final images, it's what comes for learning craft not maths and spreadsheets.

Craft needs to be honed until it's intuitive.

Ian

 

[RalphLambrecht]

Very true, Ian. That's why I went through the effort of making sample pictures, but it did prevent the critiques from refusing to accept the facts.

 

[Athiril]

If anyone is ignoring th facts, it is you. Your whole concept of DoF was contradictory to the definition of DoF to begin with, stating it's 'locked in' after exposure, when in fact it is variable.


f/4 to f/8 is 2 stops in exposure, not over. Exposure difference is not a component of the test. Only detail was a component of the test, and your point was that smaller focal lengths on smaller formats have a DoF/detail advantage, I disproved that, there's no reason not to stop down as far as you want, as long as at least the same or greater detail is there with as wide or greater DoF, then it satisifies the subject of what your point was (DoF, detail).

But in any case, the same aperture, provided the same DoF, with greater detail. So your point is wrong anyway, but your methodology and concept were completely flawed to begin with.

And in our case, f/4 = f/8, as 12.5 = 12.5, because f stands for a friggin' variable, called focal length.

As you can see,

f/4 50mm 35mm @ 2 meters focus;
close focus: 1.1818m
far focus: 2.222m
DoF: 0.403m

f/8 100mm 6x7cm @ 2 meters focus;
close focus: 1.812m
far focus: 2.231m
DoF: 0.419m

As you can see, we have the same DoF from the same aperture.


The point is, according to you. "DoF is much larger with the shorter lens (see the numbers above). Using such a lens, one has a better chance to get much more acceptable detail (definition of DoF) into the image."

DoF does not change with focal length.

Only two things influence DoF:
1. The gradient curve of resolution (perpendicular to the lens/film axis/into the picture, rather than across the negative) of the viewable image.
2. How much of that gradient curve of resolution you are perceiving from the viewable image.

1: Is created during exposure and then preparation for viewing (scanning, printing, projection, light table, whatever), and that is fixed, unless you for example later on make a sharper/bigger print.
2: Occurs during viewing of the viewable image, it is based upon your eyesight and how closely you are looking at the viewable image, how much you can perceive from it will drop with longer viewing distance, so you will not be able to perceive a sharpness differential between two certain adjacent areas as you would up close.


The resolution gradient available (into distance) on the neg does not change with focal length. It changes with aperture.


You insist DoF changes with focal length, but all you are really doing is changing the size of the hole in the lens (aperture), which is what ultimately controls DoF, not focal length. Even though f/x gives the value for aperture, it really isn't aperture, it's an expression of focal length over aperture ratio, you mistaking that for aperture, and not isolating it. You are letting aperture vary as well as focal length, then claiming it is focal length that is giving a different DoF.

What you are doing is saying, comparing f/4 and f/4 of different focal lengths, you are saying "let's keep the aperture at 1/4th of the focal length", you are varying the aperture and not keeping it constant between the shots with the 2 lenses. You're not isolating focal length, yet you claim it's focal length that determines DoF, that is like saying focal length determines perspective rather than distance to subject.

Isolate focal length by keeping the aperture constant, the size of the area of focus will remain the same, it is physics. You can't violate that. What you are stating is (incorrect) opinion and pseudoscience, not fact. You need to stop posting it, you can keep your beliefs but do not spread misinformation to others.

 

[Athiril]

Very true, Ian. That's why I went through the effort of making sample pictures, but it did prevent the critiques from refusing to accept the facts.

You haven't posted any real facts, just "facts" according to you. IE: pseudoscience.

 

[A49]

Your values are all exceeding the diffraction limit, and regardless these resolution figures wont be maintained unless the film is significantly (~3x) higher in resolution than the lens, there will be loss, the higher the resolution the greater the % loss of it will be (with the above caveat of film resolution).

Athiril, i have to add that my chart is meant for large format. With MF or 35 mm much higher resolution values are possible due to the far more opened apertures that you can shoot with.

I know the problem of systemic resolution, but B/W film three times higher exist for all usual formats: Adox / Rollei Ortho 25 (a kind of Technical Pan Film with another sensibilization curve). That film resolves 330 lpm (probably at high 1000:1 contrast) which is more than 3x80 lpm. So at least for high contrast details it fulfills your demand. There is also the Adox CHS 25 in 35 mm and in 4x5 inch that is claimed to resolve 800 lpm (with high contrast). Even Tmax 100 and Tmax 400 resolve 200 lpm. In the colour film area we have the Ektar 100 that resolves 200 lpm (with high contrast).

The resolution values for f/11 , f/16, f/22 I put in the chart are from Perez / Thalmann test that they did with Tmax 100 in D-76. That means they are already values for a system´s resolution (LF lens, LF camera and Tmax 100 in D-76). I distilled average values for the best lenses they had in their test. The values for apertures smaller than F/22 are theoretic diffraction limits that I took from a chart in the book "Tabellenbuch Fotografie" by Werner Wurst. They are for a shooting distance of 100xfocal length of the taking lens (maybe for 1:20 at which Perez / Thalmann tested, the diff limits are about 5% smaller) and for monochromatic light at 546 nanometer. If you shoot with apertures smaller than f/22 almost every lens (that is at least anastigmatic) will perform nearly at diffraction limit.

Maybe the values in my chart are slightly optimistic. In LF real life you can add to the CoC values 50 % to 100 % and calculate with them, since misalignments of film plane and ground glass degrades the resolution performance at least for the more opened apertures. On the other hand if you shoot with (very) slow high res films at higher f-stop numbers the exposure times often become longer than 1 second and your in danger of (micro) vibrations that decrease your theoretic possible resolution.

I don´t wont argue about percentages of theoretical and practical resolution limits or diffraction limits. It´s not worth the times. I hope you agree at least with that: If you shoot LF and you use the lpm and CoC values from the chart I posted to calculate DoF you surely will not rely on a too large theoretical DoF .

Best regards,
Andreas

 

[A49]

Sometimes the theoretical doesn't match the practice. Issues like diffraction limitations of smaller apertures which are always being widely quoted aren't borne out by the stunning prints made by many photographers at aperture that others theorise won't produce good definition/sharpness.

I never heard anyone say that a picture lacks resolution. If a picture has sharp edged details, a good acutance, nobody, even experienced photographers will never complain about imperfection.

There's far too much theory and not enough practice, it's not what's in spreadsheets, text books, manufacturers data that really matters to the final images, it's what comes for learning craft not maths and spreadsheets.

As I said before. (Some) forums are mainly about theory and technical facts. I think it never hurts to know something about the chemistry and physics that are connected with photography. But when you are truly inspired by a sujet and you take the pictures, your mind should be free of all this theory and you should concentrate on what you see on your ground glass.

Best regards,
Andreas

 

[RalphLambrecht]

You haven't posted any real facts, just "facts" according to you. IE: pseudoscience.

Athiril

Why the insults? I posted pictures and published equations. Where is your equation to prove focal length has nothing to do with DoF? It's still missing! When you eventually come up with it, will it show f/stop and aperture instead?

Your aperture vs f/stop statement is just a play on words. The term 'aperture' is commonly used for 'aperture setting', meaning f/stop. We all know it's the ratio of focal length over aperture opening. People set their lenses to f/stops or at a certain aperture (setting) and they mean the same thing when they say it. It's a common abbreviation. Get over it. Do you say 'fax' or 'facsimile', 'modem' or 'modulator/demodulator'? Why this nit-picking nonsense?

 

[A49]

As you can see,

f/4 50mm 35mm @ 2 meters focus;
close focus: 1.1818m
far focus: 2.222m
DoF: 0.403m

f/8 100mm 6x7cm @ 2 meters focus;
close focus: 1.812m
far focus: 2.231m
DoF: 0.419m

As you can see, we have the same DoF from the same aperture.

As I can see we have the same DoF at different apertures in terms of F-stops .

In my point of view it makes no sense to mess up the use of f-stop values. You are right: The absolute diameter of the aperture opening of a 50 mm lens at f/4 and a 100 mm lens at f/8 is the same: 12,5 mm. But there are not many photographers that uses absolute diameters in practice.

The negative format does not play any role for the resolution kept on the film. It is only important for the maximum, still sharp print size that you can later obtain from the negative. So Ralph is right, when he says, that the DoF in terms of resolution is fixed in the negative. But Ralph is not right, if you look from the print format´s point of view. Then the DoF seems not to be locked in the negative and the DoF we observe in the print changes with the enlargement ratio for the print. There is no antagonism or paradox in it.

Best,
Andreas

 

[RalphLambrecht]

Andreas

There is nothing wrong with your numbers! Athiril's statement that all of your resolution values are above the diffraction values is not correct. If you take the near-IR threshold of 650 nm, the smaller f/stops are getting close to being borderline, but take the average of 555 nm, and your numbers are just fine. Also, your lens resolutions stated are well within the system performance of good equipment and materials.

 

[onnect17]

The importance of looking at the numbers is not new.

I almost done reading some papers by Ferdinand Hurter and Vero Charles Driffield.
Here's a fragment of one of the writings (1903)
... we had long felt that art in photography ceased to play any part the moment the cap was removed from the lens, and that every subsequent operation, whether exposure, development, printing, enlarging, was strictly a matter of science, and amenable to calculation. While we quite realised that the artist will produce the best picture, we contended that the scientist will produce the best negative. The photographer, therefore, who combines scientific method with artistic skills in the best possible position to produce good work...

 

[RalphLambrecht]

Beautiful! Goes along with a quote I used:

The production of a perfect picture by means of photography is an art. The production of a technically perfect negative is a science.— Ferdinand Hurter

Where did you find it? I have copies of all their papers. Which one is it?

 

[A49]

...The photographer, therefore, who combines scientific method with artistic skills in the best possible position to produce good work...

I absolutely agree with that.

Best regards,
Andreas

 

[onnect17]

The book is titled "The photographic researches of Ferdinand Hurter & Vero C. Driffield"
You can find used copies in Amazon for $5 + shipping.
http://www.amazon.com/photographic-...=sr_1_1?s=books&ie=UTF8&qid=1297734069&sr=1-1