The Maths of a Fine B&W Print

[Photo Engineer]

But what kind of output would you need to see this difference in practice? I mean there are a LOT of former medium format users who swear by their high end DSLRs these days, and it seems their output is fine for their applications incl large prints.

In other words, is there some other equalizing or mitigating factor between the two different media?

Digital will always include a degree of inferiority due to the aliasing of side-by-side sensors. I know that there are stacked sensors but then there is the square nature (regularity) introduced and the inherent noise. All of these are things I mentioned previously.

In effect, it was a bit tongue in cheek, but I really meant to imply equal magnification, but not all other things would be equal, and therefore my qualification at the beginning of this sentence.

Digital sensors, at their best right now are on the order of 10x the size of equivalent speed film grains, and are packed less densely in the sense of how and where they are located in a sensor as opposed to film emulsion grains.

Regardless of this, a sensor would have to be X times the size of a similar film grain plus modifiers to compensate for aliasing, noise and randomness. So, my little comparison was just a rough illustration.

In actual practice, I would guess that modern sensors would have to be larger than 4x5 probably, but IDK. If a LF photographer said that digital APS sized sensors can match LF, he/she is blind. A MF sensor might be very good, especially a scanning sensor due to the huge number of pixels possible. These emulate a scanner which is also quite good. But then, you are limited to still life or landscape photogrpahy.

PE

 

[Michel Hardy-Vallée]

Besides the resolution debate, what I find actually more relevant about sensors is how they render colors.

MTF-addict Erwin Puts has a test report that illustrates well what I mean : http://www.imx.nl/photosite/zeiss/test85/t004.html

If you compare the film & the digi photos, it seem that digi gets a better resolution, but on the other hand the colours just look awful and sickly.

In other words, the qualitative aspect might be more important to make a fine print than sheer resolution power.

 

[DrPablo]

Honestly, what I've come to conclude is that for about 99% of photography it's more about which workflow you prefer than any quantifiable difference. I mean I posted a cyanotype on FredMiranda, and someone whipped up a Photoshopped digital version; and granted it didn't look real, but I'm sure with about an hour of PS work, and then printing it on watercolor paper, it would look pretty good. That's not to say there wouldn't be some measurable differences in the final product, but they would be pretty slight. The major difference in my mind is that real cyanotypes are fun, and making a digital one, however "real" looking, would be unbridled drudgery.

In the absence of special applications, I'm not sure it really matters which medium you choose. I mean it doesn't take the resolution of 120 film to fill up a wedding album and make a few 16x20s for people, so I can see why most wedding photographers have jumped ship from film (especially because the material costs get recouped pretty quickly with their digital equipment).

So I may be making an excessively reductionist statement, but I don't think the quantifiable differences between film and digital are often important aesthetically -- I think it just has to do with preferred workflow, and how well you can actualize the "look" you want in your pictures.

 

[Jed Freudenthal]

I am VERY interested!!

Would it be possible to scan and post both MTF charts, so that I - we - might compare them?

BTW - The "DISTAGON 40/40"? - are you referring to the Distagon CF f/4 FLE or the older lens without the two "floating" front elements... and are you comparing a 40mm lens to a 90mm?

It is a Distagon HTF 40/40 coated lens in comparison to the Super Angulon uncoated lens (probably 40 years old). The Super Angulon was far better in the case of landscape photography. I never used the Distagon for landscape photography anymore. However the Distagon was good for other applications.

Jed.

 

[Thomas Bertilsson]

Original Thread Reply

I couldn't contribute to what you're asking for if I wanted to. My photography is about as unscientific as it can be.

What I'd like to contribute, however, is that I burn with passion for film photography. It is my creative outlet, I love the look of Tri-X in Pyrocat-HD printed from MF negs to about 10" square. I have seen many digital photographs that rival or surpass the quality I'm able to achieve, but none that I would have liked to create myself. The ability to hold the negative in my hands, study it, and finally shine light through it to print it, is something that is pure magic to me. Every part of the process is tangible, it physically exists.
A digital capture is a bunch of electronic charges, you can't see it, you can't touch it, and you have to stare at a computer screen to do anything with it.

Tell them why you love film photography. Tell them why it is your medium of choice. Tell the audience why you burn with passion for chemistry, darkrooms, and safelights. After all, it's the finished print in your hand that matters, right, and the process to produce it?

- Thom

Hi all,

I am giving a talk on B&W prints to people who are digital photographers who know nothing about traditional B&W process. So I need to know some of the maths involved to enable them to understand how a B&W print can have so much detail in the shadow and highlight areas. If anyone knows the maths involved here of the difference between the DMAX of B&W compared to digital I would appreciate it greatly. Many Thanks Paul

 

[Ed Sukach]

It is a Distagon HTF 40/40 coated lens in comparison to the Super Angulon uncoated lens (probably 40 years old). The Super Angulon was far better in the case of landscape photography. I never used the Distagon for landscape photography anymore. However the Distagon was good for other applications.

I've been trying to obtain the MTF charts. I'd really be interested in comparing those two lenses, "MTF - wise", to determine the particular information leading to the conclusion that the "Super Angulon" was - so obviously - superior. A Google search of Schneider doesn't list a "Super Angulon 90mm wide angle" lens, and Zeiss doesn't list a 40/40 anything. Were these lenses used on a Hasselblad/ or Zeiss 6x6 SLR? .. Seems strange that a 90mm in 120 format should be characterized as a "wide angle" lens...?

Can you help me with this search?

 

[eddym]

Ed, the Schneider 90mm Super Angulon is a wide angle lens for 4x5 format. it comes in two versions, one f8 maximum aperture, and the other f5.6.
I'm not sure I get the point of comparing a medium format lens to a large format lens, but there you are.
The technical data of the 90mm f5.6 lens is available here:
http://www.schneideroptics.com/ecommerce/CatalogItemDetail.aspx?CID=168&IID=1840

 

[Jed Freudenthal]

I've been trying to obtain the MTF charts. I'd really be interested in comparing those two lenses, "MTF - wise", to determine the particular information leading to the conclusion that the "Super Angulon" was - so obviously - superior. A Google search of Schneider doesn't list a "Super Angulon 90mm wide angle" lens, and Zeiss doesn't list a 40/40 anything. Were these lenses used on a Hasselblad/ or Zeiss 6x6 SLR? .. Seems strange that a 90mm in 120 format should be characterized as a "wide angle" lens...?

Can you help me with this search?

The comparison between the Distagon and Super Angulon has been done, I guess around 1996. I am in the process to retrieve the MTF information (on paper). Right now I will give further details: The Distagon has been used at thet time on a Rollei SLX, at an optmum diafragma around 8-11. The Super Angulon has been used at f=16. , both on sheetfilm and with a 120 roll film adapter. It is true that one is comparing 6 cm with 10 cm. However, the difference in image quality( with landscape photography) is so much, the difference in negative size cannot explain the difference in image quality.
The MTF of the super angulon is a 'straight' line at around MTF=92% all across the full image (center to edge). There are no off-axis effects, and this results in a fine landscape photography from the center to the corners ( off-axis effects, varying MTF values from center to corner are experienced al poor image effects). Now the Distagon HFT 40/40. The MTF's of these lenses are lower than 92%. But the off-axis effects really show up. In fashion/reporting photography these off-axis effects are hardly noticed. Therfore, I take different lenses/cameras in different situations. And, when I put the roll filmadapter (120 film) on my 4x5" camera; it is in fact a medium format camera.
I understand: your question is,is it allowed to compare different negative sizes? Yes, it is one of the primary objectives of the MTF to compare different things, like lenses, films, negative size etc. For example, one can choose a system where the quality of the print becomes better with larger print size, and one can choose system where the print quality will be lesser with large print sizes. MTF is a tool to predict print quality, and can be considered to be a selection tool. During your photographic activities behind the camera, don't think about MTF's.
I recommend to do the comparative test yourself with the lenses (with MTF data) you have.

Jed

 

[Ed Sukach]

The comparison between the Distagon and Super Angulon has been done, I guess around 1996. I am in the process to retrieve the MTF information (on paper).

I have the information of the Super Angullon XL in my hands right now - MANY thanks to eddym.

Right now I will give further details: The Distagon has been used at thet time on a Rollei SLX, at an optmum diafragma around 8-11. The Super Angulon has been used at f=16. , both on sheetfilm and with a 120 roll film adapter. It is true that one is comparing 6 cm with 10 cm. However, the difference in image quality( with landscape photography) is so much, the difference in negative size cannot explain the difference in image qual
ity.

That raises other questions: Are you comparing enlargements from the Rollei with enlargements from the -- what was it -- 4x5 camera? Was the same enlarging lens used (ands how did the MTF of the enlarging lens affect the image?) Was the same film, same developer, same processing used? I find it difficult to understand how a lens designed for use with the 4x5 format - with a much larger "circle of confusion" built in to the design specifications could - other than by random good fortune, offer more resolution that one designed for 6x6.

The MTF of the super angulon is a 'straight' line at around MTF=92% all across the full image (center to edge). There are no off-axis effects, and this results in a fine landscape photography from the center to the corners ( off-axis effects, varying MTF values from center to corner are experienced al poor image effects)

What aperture are you observing? In the MTFs I have here printed out, there is NO such straight line.

Now the Distagon HFT 40/40.

I cannot find information about a "Distagon HFT 40/40." Can you guide me to it?

... it is one of the primary objectives of the MTF to compare different things, like lenses, films, negative size etc. For example, one can choose a system where the quality of the print becomes better with larger print size, and one can choose system where the print quality will be lesser with large print sizes.

Prepare for an intelligent reaction .. (blank look) .. Huh? You have completely lost me there.

MTF is a tool to predict print quality, and can be considered to be a selection tool. During your photographic activities behind the camera, don't think about MTF's.

I don't.

I recommend to do the comparative test yourself with the lenses (with MTF data) you have.

Exactly what I am trying to do... and what I will continue to do, as soon as I straighten out this "Distagon HFT 40/40" thing.

I will say, in comparing the Super Angulon to the Distagon CF f/4 50mm (FLE) - they seem to be remarkably close.

 

[kunihiko]

Ed,

I guess he means Distagon HFT 4.0/40 for Rollei SLR. I'm not sure it's identical to Distagon T* 4/40 IF CFE.

For a fair comparison, when he shoot 6x6 with Super Angulon XL 90mm he only uses very center of the image circle. 40mm from center is enough. It can be said that MTF curve of SA90mm is almost straight at f22 from the date sheet which linked above. It's not that much straight for entire image circle, but straight enough for 40mm from the center.
If he shoots 4x5 with SA XL90mm he needs 75mm from the center. It still can be said it's straight enough if he doesn't use any movement.

Anyway, as you have already pointed out, comparing Distagon 40mm and Super Angulon XL 90mm isn't fair. 40mm is 40mm, 90mm is 90mm.
If someone compares the Zeiss lenses for 6x6 format with Super Angulon XLs it should be SA XL38mm with Biogon 38mm or SA XL90mm with Planar 100mm.
All of them is looking good from the data sheets, but I don't think I could shoot SA XL38mm on a SLR because my Hassy needs some space for the mirror. I would not shoot Distagon 40mm on my 4x5 because it doesn't have enough image circle for 4x5. Even if I shoot 6x6 with my Toyo I will never use retro focus design like Distagon because I don't have to have that huge and heavy lens. Distagon is better at relative illuminance, though.

Even though the MTF curve of SA XL90mm looks bit better than Distagon 40mm, they are in different leagues. One is for MF SLR, other is for LF view camera.

 

[Ray Heath]

why don't you guys just stop now, we can all see how technically adept you are, give it a rest

 

[Ed Sukach]

why don't you guys just stop now, we can all see how technically adept you are, give it a rest

I am not, and I do not think the others here, discussing either Dmax or MTF are trying to exhibit Technical Prowess. The subject of this thread is "The Math(ematic)s of a Fine B&W Print", and that is what we are doing.

If you feel "offended" (???) think of pages in a Photography magazine - you are not OBLIGATED to read those, and you are not obligated to read the "pages" here. Certainly, there is no one with a pry bar forcing my eyes - I am only trying to understand what is being written here - with the hope that I may learn something.

Would you rip pages out of someone else's magazine?

BTW - I noticed that there is selection here called "Thread Tools". Selecting that, you have an option: "Ignore This Thread". You mighr find that useful.

 

[Ray Heath]

BTW - I noticed that there is selection here called "Thread Tools". Selecting that, you have an option: "Ignore This Thread". You mighr find that useful.

i don't want to ignore anything but i would like to see it move on

 

[Photo Engineer]

Moving right along:

Ed, MTF is a very slippery thing and is a complex function of the lens, film and way both are used such as format and f stop as shown in your posts and Kunihiko's. It is also a function of development, printing and viewing conditions.

That said, in summary, I go back to my comments that it is nearly impossible to come up with maths of photographic systems. After the fact, if we had a beautiful picture, it was often possible to reconstruct the variables, but that too often did not work to 'prove' what gave a picture its certain qualities.

So, in the end, the picture is the sum of a variety of math functions, none of which are sufficiently understood or under the control of the photographer, but are used as an 'art' to come up with the beauty we know as photography.

PE

 

[Ed Sukach]

Ed, MTF is a very slippery thing and is a complex function of the lens, film and way both are used such as format and f stop as shown in your posts and Kunihiko's. It is also a function of development, printing and viewing conditions.

Yes - I agree.

So, in the end, the picture is the sum of a variety of math functions, none of which are sufficiently understood or under the control of the photographer, but are used as an 'art' to come up with the beauty we know as photography.

Considering everything, ... I disagree, - sort of.

What you say is true, at least partially; I think the "mathematics" are of interest/ some degree of "importance" ... but the greatest part of "the image" is in its esthetics (note 1).

Note 1. I'm using "esthetics" instead of "aesthetics". If that spelling was good enough for Clement Greeberg, it certainly is good enough for me.

 

[Photo Engineer]

Ed, without the math we could not design a good lens, or a good film, it is just that the photographer need not worry about it too much when out there taking pictures with a good lens, camera and film and when printing and processing and printing with good chemistry and equipment.

That is what I'm trying to express.

And, trying to explain such a mix of math functions to the average person is nearly futile unless they have a math and physics background.

PE

 

[Jed Freudenthal]

Ed,

I guess he means Distagon HFT 4.0/40 for Rollei SLR. I'm not sure it's identical to Distagon T* 4/40 IF CFE.

For a fair comparison, when he shoot 6x6 with Super Angulon XL 90mm he only uses very center of the image circle. 40mm from center is enough. It can be said that MTF curve of SA90mm is almost straight at f22 from the date sheet which linked above. It's not that much straight for entire image circle, but straight enough for 40mm from the center.
If he shoots 4x5 with SA XL90mm he needs 75mm from the center. It still can be said it's straight enough if he doesn't use any movement.

Anyway, as you have already pointed out, comparing Distagon 40mm and Super Angulon XL 90mm isn't fair. 40mm is 40mm, 90mm is 90mm.
If someone compares the Zeiss lenses for 6x6 format with Super Angulon XLs it should be SA XL38mm with Biogon 38mm or SA XL90mm with Planar 100mm.
All of them is looking good from the data sheets, but I don't think I could shoot SA XL38mm on a SLR because my Hassy needs some space for the mirror. I would not shoot Distagon 40mm on my 4x5 because it doesn't have enough image circle for 4x5. Even if I shoot 6x6 with my Toyo I will never use retro focus design like Distagon because I don't have to have that huge and heavy lens. Distagon is better at relative illuminance, though.

Even though the MTF curve of SA XL90mm looks bit better than Distagon 40mm, they are in different leagues. One is for MF SLR, other is for LF view camera.

I am talking about the Super Angulon 90/6.8. The design of this lens is over 50 years old. In more recent years, this lens has been covered with coating. And Schneider still sells this lens. Of course, this lens must be a good one, otherwise it has disappeared. I made a comparison between this Super Angulon and the Distagon HFT 40/40, because images are approximately the same. (6x6 vs 4x5"). The difference in image quality was so much; the fact of the difference in negative size cannot explain the difference.
More recently, I compared the normal lens Planar 80 ( of the Rollei SLX) and a APO Sironar -S 150/5,6 for 4x5" . Again a difference that cannot be explained by the negative size. The APO Sironar -S 150/5,6 was far superior. In fact, I have bought the APO Sironar -S 150/5,6 on its MTF properties and the large image circle.
However, I still use the Rollei ( a Rolleiflex 6008 by now) for 'moving objects' and the prints I make from the Rollei negatives are relatively small. The standard size for the 4x5" negatives is 16x20" , but I could easily go to a square meter ( and once I did that). In fact, when the maginfication goes up, not only the size of the print goes up but the quality of the image too. This means that there is detailed information ( high spatal frequencies) in the negative that will not show up at low magnification.
When printing from 35 mm negatives, I usually observe a decease in print image quality for increasing magnification.
These phenomena can be explained using MTF reasoning, including the MTF of the human eye. In fact the properties of our human vision are important here. And that is the central point of the studies, I mentioned before. It is remarkable, that the MTF's of the human eye show so little variation between human beings. And it is remarkable that people not have to 'learn' what good image quality is. In the German study, the people in the panel were not told what good image quality is. Nevertheless have the same idea what good image quality is. These observations led the German research group to the link between MTF and subjective image quality. However, it has been made clear that there are more factors (outside MTF) that can influence image quality ( grain, printing tone, printing contrast, vibration of the camera and enlager, depth of field). Moreover they found that the subject has an optimal MTF. Landscapes require a high MTF, but portraits require a lower MTF. ( a portrait should not be too detailed). Rollei will bring a Rolleiflex with a Xenar as portraitlens on the market this year. And look at the MTF of this lens ( web site of Rollei) it has a relatively low MTF at the lower frequencies. Rollei has good reason to that.

Jed

 

[Ed Sukach]

I made a comparison between this Super Angulon and the Distagon HFT 40/40, because images are approximately the same. (6x6 vs 4x5"). The difference in image quality was so much; the fact of the difference in negative size cannot explain the difference.

The area of a 4(10.16cm) x5(12.7) negative is five - six times greater than a 6cm X 6cm (actually 5cm x 5cm). That, to me is an unreasonable stretch of "approximately".

By that example, 4"x5" would be "approximately" the same as 8"x10" - which I don't think it is... not even close.

 

[Roger Hicks]

The area of a 4(10.16cm) x5(12.7) negative is five - six times greater than a 6cm X 6cm (actually 5cm x 5cm). That, to me is an unreasonable stretch of "approximately".

By that example, 4"x5" would be "approximately" the same as 8"x10" - which I don't think it is... not even close.

Same order of magnitude...

Besides, think of economics. Taking the normal level of accuracy of economic predictions as your standard, 4x5 and 8x10 are very close indeed.

Cheers,

R.

 

[Ray Heath]

I am talking ...................

Jed

thnx Jed, you win, just to see what your on about i went to your gallery, i think it must be broken as their are no images displayed

come on guys, like it or not, photography is an expressive art form, it's not maths, it's feeling, emotion, subject, lighting, artist choice, creativity, artistic intent, meaning, message, interpretation, expression, it's about getting out there and experiencing your world

 

[gainer]

My final word, I promise.
We have been discussing two different uses of the MTF. The general form is the Fourier Transform, which is a statistical function of sorts. The transform is derived from the autocorrelation function. The transform is from a time function to amplitude and phase angle functions of frequency in the transmission of audio or radio information or from a spacial function to a spacial frecuency function. The transform is an important measure of the information content of an audible or visible message stream. When the quantity measured is a ratio of input to output of a message stream, we are interested in the maximum information rate through the transmission system. The modulation transfer function is properly applied to lenses, filters and the like and in photography is properly multidimentional because color as well as spacial frequency must be taken into account. The MTF is a valuable predictor of performance of a linear system.

The information content of a visual art object has no esthetic meaning. One cannot say whether the source of the object was analog or digital, and the information content can only tell one how much information capacity a system must have in order to transmit the object from one place to another with no more than a specified loss of information.

Unfortunately, the MTF is even less useful when there is a non-linear element somewhere in the system. Film is non-linear by its nature. Printing paper is also. So, for that matter, is the human eye. Under certain circumstances, the overall system can be made to approach linearity, but we photographers relish the ability NOT to do so in order to excercise "artistic license.

We can use the MTF of the lens to predict the transform of the scene that will be put on the film provided that the lens is linear enough. We can't do a whole lot of mathematical prediction with an MTF of film because that is a non-linear element. The paper or other printing medium introduces another non-linear element.

We can use the same lens on either analog or digital camera. If we can't use the MTF accurately to predict the next step in the analog system, there is no way to compare mathematically the film image with the digital image that appears on the viewing screen. MTF is not guaranteed to work on non-linear systems. A comparison of MTF's will therefore not give a true basis for judging which system is better.

 

[Jed Freudenthal]

The area of a 4(10.16cm) x5(12.7) negative is five - six times greater than a 6cm X 6cm (actually 5cm x 5cm). That, to me is an unreasonable stretch of "approximately".

By that example, 4"x5" would be "approximately" the same as 8"x10" - which I don't think it is... not even close.

It is a factor of approximately two (linear). between 6x6 cm and 4x5". And 35 mm is another factor of two compared to 6x6cm. MTF is in c/mm and not square mm. The image quality studies are done on all these film sizes, and one can use the data on each size. One of the MTF applications is the comparison of image quality between different negative sizes. Like a comparison between different films-developer combinations ( that is why firms like Gevaert-Agfa, FOMA, Kodak give MTF data for film-developercombinations).
[unfortunately, we do not see any information from Kodak anymore].
Jed

 

[Jed Freudenthal]

My final word, I promise.

Gadget:
Although I am a physicist, I try to clarify the point in non-scientific language. In the past, I was also sceptical using MTF. But, it might have been around 1980 that somebody told me, Rodenstock made its lenses for 4x5" just meet the quality of the human eye. And Rodenstock, Linhof and the German MTF research group were all in Munchen Germany,
In other words: it is not necessary to make 'better' lenses, because the human eye is unable to see it. I found this intriguing, and I checked it. It was true. After that I made many comparisons between negative sizes, lenses, films etc. And, after a while you get a feeling what the MTF language has to tell. But, it applies to image (print) quality only as judged by humans (subjective). And there is much left in ther artistic domain. Don't worry, MTF is just to control a technical part of the photographic process.

You were mentioning the non-linear parts of film and paper. The MTF applies in these areas too. But, there is much more going on. Kodak tried once to understand and control that. They did not succeed, at least not with simple and practical procedures.

Jed

 

[Jed Freudenthal]

Ed, without the math we could not design a good lens, or a good film, it is just that the photographer need not worry about it too much when out there taking pictures with a good lens, camera and film and when printing and processing and printing with good chemistry and equipment.

That is what I'm trying to express.

And, trying to explain such a mix of math functions to the average person is nearly futile unless they have a math and physics background.

PE

I fully agree with you. But why not give it a try. I checked it lately. I explained MTF to non-scientific people. And, within 10 minutes they understood it. Of course, practice is required to get it in a grip. But for me, this experience was a signal that it is possible to get the idea across.

Jed

 

[Ed Sukach]

It is a factor of approximately two (linear). between 6x6 cm and 4x5". And 35 mm is another factor of two compared to 6x6cm. MTF is in c/mm and not square mm.

A "Factor of two" is still describable as "approximate"??

Unfortunately I don't view a photograph as one dimensional (the length of a single line) ... I see an entire area.

Possibly you might help here .. I was about to look this up, but I am pressed for time at the moment: What is the diameter of the "circle of confusion" used in the design of a lens intended for use with the 6 x 6 cm format; and what is it when designed for 4" x 5"?

BTW - What were the dimensions of the finished prints used in you comparisons?

Also .. Try the MTF charts for the Hasselblad - Zeiss Sonnar CF f/4 150mm for its MTF....