Theoretical digital sensor equivalent to 8x10? [Archive]

Andrew Clearfield

18-Dec-2012, 20:39

I am curious to calculate how large an array of sensors would be needed to be able to record as much information as there is on a well-made and properly processed 8x10 negative.

We all know that a digital sensor may have an image-capturing capability greater than that of a traditional, chemically-based negative of the same area; or accordingly, may be (and usually is) smaller than the traditional film-based format of a comparably-sized camera. The question is, what size, and pixel-capturing capability would be necessary to approximate the amount of information one gets from an 8x10 camera?

Here are my assumptions, to use as a basis for a comparison:

Obviously, such a calculation would be a theoretical limit, based upon an ideal contrasty subject in bright light, with the finest grain/developer combination customarily used in ordinary pictorial photography; since the standard for most digital sensors is equivalent to ISO 100, let us say that the analog-based benchmark would be a sheet of FP4 Plus processed in a metol-hydroquinone developer such as ID-11 (D-76). I prefer PMK Pyro, but the numbers used to be reported for more conventional developers. If memory serves me rightly, such a combination would have a theoretical maximum resolution of about 125 lp/mm, well beyond any lens.

The best general-purpose lenses in use for large-format photography are supposed to have a maximum resolving power equivalent to, say 40 lp/mm in the central area, and perhaps 20 out in the periphery. (I'm using Schneider's figures for their old Super-Symmar series, but I'm assuming my Commercial Ektar and Fujinon-W lenses are not that far off from this. Yes, there are many other factors which go into the design of a lens, but I'm trying to calculate maximum data-collecting capability here.)

An 8x10 negative or transparency has an area of slightly more than 50,000 sq. mm (51,562 to be more precise, but there are the unexposed borders of the film, and anyway, these calculations are going to have to be very approximate, anyway.) Even at a very high 40 lp/mm, this implies the film would have registered a little over 4,000,000 data points (intersections) from a perfect grid inscribed with all lines at the minimum distance resolvable, again under these ideal, high-contrast conditions. More likely, there would be between 1 and 2,000,000 such points recorded, as the resolving power falls off away from the central axis and under more normal light conditions. (Of course, if the image produced has this capability, the number of useful data points at the classical photographer's disposal would be the same in any ordinary photograph, it simply wouldn't be as easy to measure them.)

Now, assuming we are concerned with instantaneous image capture here—no scanning and re-sampling of the data, because the 8x10 camera is not itself limited to perfectly static subjects—how many pixels are necessary to approximate those 1 to 4 million usable grains in the image, and how large a sensor does one need to record the same amount of information? I assume one gets one bit of data (i.e., 1 or 0) from each cell of a sensor, and at 8 bits per byte, if (and this is a big part of my question) it takes 1 byte to contribute the equivalent information of that one exposed and developed grain of silver, (i.e., one pixel) it would seem that a sensor capable of generating 32 Mp would be capable of giving the same amount of data as a black and white photograph from an 8x10 camera, and, since digital sensors are always capable of recording color, the sensor would have to be 4x that size, or 128 Megapixels to have the same capabilities. Now, no one has yet made a sensor that large to my knowledge, but if they could produce one at a cost someone could afford, would that do the trick? Or have I omitted some factors? Equivalently, there is a great deal of software which reads and interprets the image before it is transmitted to a recording medium, even in Raw format. This should reduce the number of Mp, and therefore, of cells necessary to give the same result.

I also appreciate that the physical size of each cell makes some difference. Nonetheless, does a Leaf 80 Mp 40mm X 54mm sensor capture the same amount of information as, say, a 5"x7" film camera? Would it, if you made it physically twice (or four or six times) the size in order to have larger individual cells?

I understand that such an image would still not have the same character or look as the chemically-produced image. But is such a calculation as I have made fundamentally faulty, and if so, why?

Leigh

18-Dec-2012, 21:09

We all know that a digital sensor may have an image-capturing capability greater than that of a traditional, chemically-based negative of the same area;
Huh? Don't know where you got that. Sounds like advertising hype.

A pixel sensor is a physical structure that includes a lens, the active sensor element, and electrical connections.

An image forming element in film is a cluster of molecules.

------

I typically scan my 8x10 film at 2400 ppi, so the 8" side is 8 * 2400 = 19,200 pixels, and the 10" side is 24,000 pixels.

The image area is therefore 19,200 * 24,000 = 460,800,000 (461 Megapixels).

A good drum scan @ 9600 ppi = 8 * 9600 * 10 * 9600 = 7,373 Megapixels = 7.4 Gigapixels.

- Leigh

welly

18-Dec-2012, 21:43

This has been discussed a few times:

http://www.largeformatphotography.info/forum/showthread.php?79815-8x10-digital-capture-back&highlight=8x10+sensor

http://www.aphotoeditor.com/2011/08/23/mitchell-feinbergs-8x10-digital-capture-back/

http://www.largeformatphotography.info/forum/showthread.php?94077-Don-t-think-they-will-ever-make-a-4-x-5-digital&highlight=8x10+sensor

http://www.largeformatphotography.info/forum/showthread.php?70392-8x10-back-in-my-lifetime&highlight=8x10+sensor

paulr

18-Dec-2012, 22:45

timparkin

19-Dec-2012, 03:30

If you're curious about equivalence, read this white paper (http://www.falklumo.com/lumolabs/articles/equivalence/index.html) very carefully.

You'll see why it's actually quite easy to equal 8x10 in situations where you need a lot of depth of field, and almost impossible to do so in situations where you don't.

This theory is born out precisely in tests like the one discussed here (http://www.onlandscape.co.uk/2011/10/the-perils-of-testing/).

Paper looks interesting - must look :-)

I worked out the possible resolution for different apertures with 10x8 on Delta 100 and I got the following

f/22 = 600Mp
f/32 = 450Mp
f/45 = 300Mp
f/64 = 130Mp

However, this is resolution which is different to how the average punter assesses sharpness. Give your average punter a pictorial image blown up to 40"x50" and we reckon the following are the results (given comparitive analysis vs IQ180 on Alpa)

f/22 = 250Mp
f/32 = 180Mp
f/45 = 100Mp

The results aren't quite comparative because as you increase print size, the IQ180 overtakes the 8x10 for a short period at around 30x40 / 40x50 and then drops back again as you get up to 50x70 etc - this is because the acute bayer pixel sharpening effects peak at those print sizes whereas you can't sharpen the film quite as much without grain/noise getting in the way. Ask the viewers which they prefer though and they invariably prefer the 8x10 :-)

Andrew Clearfield

19-Dec-2012, 20:17

I typically scan my 8x10 film at 2400 ppi, so the 8" side is 8 * 2400 = 19,200 pixels, and the 10" side is 24,000 pixels.

The image area is therefore 19,200 * 24,000 = 460,800,000 (461 Megapixels).

A good drum scan @ 9600 ppi = 8 * 9600 * 10 * 9600 = 7,373 Megapixels = 7.4 Gigapixels.

- Leigh

And that means if I scan a 120-year-old 8x10 thick-emulsion, coarse-grained negative made with a soft-focus lens on your drum scanner @9600 ppi I've improved the resolution of the original camera????

In the process of scanning and photoshopping a picture, many millions or even billions of pixels can be created. Sometimes, interpolation can improve the apparent sharpness of the image. That doesn't mean any more data was captured in the original exposure. I'm talking about the amount of real-world information captured and stored on a reproducible medium (a piece of film, a chip) at the moment the exposure was made. That's very different.

Andrew Clearfield

19-Dec-2012, 20:24

@timparkin:
These numbers seem more likely. How did you arrive at them?

Sharpening effects probably shouldn't count, because you're not capturing any further information, you're just exaggerating what is already there. (Of course, if the net result is that the final print of 30"x40" looks sharper to trained eyes, and without having the flattened, artificial look of an over-sharpened print, then you've accomplished part of the objective of making a photograph in the first place. Naturally, there are a multitude of other factors than just sharpness—but I'm surprised that a 40x54mm 80 mp sensor could do this well, in an objective comparison.)

Leigh

19-Dec-2012, 20:42

And that means if I scan a 120-year-old 8x10 thick-emulsion...
I wasn't talking about scanning junk. If you wish to waste your time doing so, by all means have at it.

- Leigh

Nathan Potter

19-Dec-2012, 21:51

I'm inclined to think that the Ops comparison is meant to simply compare the image capturing ability of an 8X10 film to that of the top digital sensor available (or if it were available in 8X10 format).

There is a great difference in the physics of image capture between a digital sensor and film so even just a comparison at the film plane is difficult let alone including lens, f/stop, DOF etc.

If we look at only the film plane and consider the best digital sensor of 6 or 7 µm pitch (say about 180 pixels per mm). That's about 25 MP/ sq. in. so for 80 sq. in. that would be around 2000 MP. The caveat in this is that currently such sensors are Bayer designs where a color is reconstructed from a reading of adjacent pixels or B&W is determined from mainly the green cells. The actual size of a color pixel is really determined from the particular demosaicing algorithm used - also the case for a B&W image. Thus a color pixel is more like an 18 to 20 µm pixel, with B&W possibly less, depending on the specific algorithm used. This bayer demosaicing then reduces the effective sensor resolving capability by up to a factor of 10 for color and somewhat less for B&W. If the demosaicing algorithm is very sophisticated one can toggle every RGB pixel one by one and use the surrounding pixels to extract a color from every one of the 2000 MP but even with such a technique the color at each pixel is still a synthesis and not a full disclosure of what was at that particular pixel. This uncertainty scales with any size sensor one would choose.

Film on the other hand is a fundamentally different form of image capture as Leigh has suggested above, a "cluster of molecules". Fine grain film would typically have silver halide grains of about 0.8 µm diameter with a range of say 0.5 to maybe even 1.5µm. When irradiated by photons say as few as 1 or as many as 100s' the latent image increases in area by several times and in some extreme cases a darkening may be visible (printing out so called). Under low light conditions the latent image may still be close to the silver halide dimensions - obviously very high resolution and yielding what many traditionalists call pleasing microcontrast after development with sometimes marvelous shadow detail in a final print. At the other end of the density range, heavy illumination from the scene highlights, the larger latent image grows even larger upon development forming silver clumps up to say 20 µm in diameter. So the resolving power of film is a sliding scale depending on the amount of original exposure and modulated by the development sequence.

These are just the variables when considering the sensor/film part of the photo process. When we add the variables of lens resolution, f/stop, DOF, etc. we add a myriad of additional complications to the resolution performance of the system.

I dare say that is why many here would prefer to study the qualitative aspects of an actual negative or print as a basis of comparison.

Nate Potter, Austin TX.

rdenney

19-Dec-2012, 22:25

It should be noted that there are other effects of the format change than just resolution. For one thing, depth of field is considerably different. Another is the required capabilities of the lens, particularly at both wide and the smallest apertures. Issues of resolution ignored, it's very difficult to create an 8x10 image using a much smaller digital sensor, because the lenses are not fast enough to provide the same depth of field effect, or sharp enough at those wide apertures to perform well enough. Thus, many of the images possible with the 8x10 are really challenging with something much smaller (and vice versa).

Rick "especially if fast lenses are used for extreme selective focus on the 8x10" Denney

Andrew Clearfield

19-Dec-2012, 23:28

@leigh:
You totally missed my point: the number of pixels downloaded to your computer file has NO simple relationship to the resolution of the original image. Get it now?

Andrew Clearfield

20-Dec-2012, 00:07

@Nathan Potter:
Obviously, the illumination conditions, choice of film, developer (including dilution), length of development, as well as the many camera variables, including the lens, focal length, distance from the film plane, and so forth will all have an effect upon the final result. I've been doing traditional photography for long enough (over fifty years) to have a pretty good idea of all the variables involved, and the ultimate nature of the image formed especially in terms of granularity. The hypothetical conditions mentioned were those used for maximum resolution tests of standard lens/film/developer combinations, so many of these variables don't apply here. The major manufacturers—of film and of lenses both—used to provide various data to make comparisons between different film and lens combinations, latterly in term of MTFs, and these were used as well to indicate the maximum resolution one could reasonably expect from a given camera/lens combination, and of course, a given format.

What I was trying to do, if possible, was to arrive at some indication of what sort of sensor would be needed to give one an equivalent resolution to that of a silver image. Practical field comparisons under various conditions represent one such basis for comparison, but involve many variables; there is also the question of the intervening effects of the software used to record the digital image on a storage medium, as well as that employed to turn it into the final print. Similarly, in classical photography, the process of turning the negative image into the finished print also added many variables. Such a test may be the most important from a practical point of view, but it is also the most subjective. At the other end of any such comparison is that of the resolution chart, photographed at some optimal distance, at various illumination levels, different apertures, etc. between the digital capture before being manipulated, and the silver halide film, under analogous controlled conditions, but normally using the standard sorts of film/developer/processing which would be employed by actual serious photographers. Here, questions like the average grain size seem reasonable: the subject is flatly lit, so many of the variables you mention do not apply.

Your discussion of the size of the usual array, and of the typical size of the granular silver configuration after exposure to light and development is directly applicable, and I'm wondering if we can continue the analysis in this way. That's all.

gary mulder

20-Dec-2012, 01:03

Much moor interesting than comparing the amount of information stored is to compare the artistic exploits of both systems. You can make a decent print from both. Having both to my disposal I discovered that my output is different depending on which I use. For instants doing landscape using tilt en swing on 8 x 10 is much easier than with the small screen of a digital view camera. And constraining to just a few shots makes you think harder before taking one.
On the other hand in the studio digital gives back the instant feedback that polaroid once provided.

Steve Smith

20-Dec-2012, 02:52

And that means if I scan a 120-year-old 8x10 thick-emulsion, coarse-grained negative made with a soft-focus lens on your drum scanner @9600 ppi I've improved the resolution of the original camera????

No. The scan will never have the resolution of the original but the more detailed the scan, the better chance you have of getting the most from it.

Steve.

Doug Howk

20-Dec-2012, 04:28

Film of course is not the only traditional method for image capture when comparing to digital sensors. From an article on Cincinnati Waterfront Panorama Daguerreotype (http://www.rochester.edu/news/photos/daguerreotype.html)

The clarity of each of its eight copper panels is equivalent to a 140,000-megapixel digital image, thousands of times better than high-end commercially available digital cameras today.
What would the sensor size be that would produce a 140 gig pixel image?

ChuckT

20-Dec-2012, 09:32

I think the only answer that will satisfy this question would be 1 sensor = 1 silver grain.

The math makes my head hurt:)

When the "mavens" accomplish this, I'll be happy.

cvt

rdenney

20-Dec-2012, 10:05

I think the only answer that will satisfy this question would be 1 sensor = 1 silver grain.

The math makes my head hurt:)

When the "mavens" accomplish this, I'll be happy.

cvt

That depends. Do you want to scan silver grains so that their shape can be sharply delineated, as is possible (to some extent) when working with the negative? Then you'll need more many pixels than one per grain. It's possible to simulate a random arrangement with an orderly array, but the spatial frequency of the array has to be much higher.

But the question wasn't how to make as perfect as possible a digital scan of a negative.

The original question, however, can go in any of several directions. If one wants to make an 8x10 image, then only 8x10 will do. But how small do the pixels have to be on an 8x10 sensor to achieve some given standard of performance? If that standard is defined by the lens, we could speculate something like 100 pixels/mm (for lenses that resolve 50 line paris/mm--a high standard for 8x10 applications). That's about 500MP. But that exceeds the requirements of most photographers (and, frankly, most lenses used on 8x10 cameras) by a good ways.

If the standard is defined by the print size, then one should start with something like 5 to 8 pixels/mm on the print, rendered with good MTF, and do the math from there. A 10x enlargement would therefore require 50 pixels/mm at the image plane, or 125MP. That's more consistent with other tests.

Note that those 125MP on a smaller format will require different lenses and get different results than when using 8x10 because of the other differences I mentioned previously.

By the way, OP, people will interpret your question in different ways. If that's not the direction important to you, it's not their fault.

Rick "who'd be pleased to own a 50 pixel/mm 4x5 digital back, if such existed" Denney

EH21

20-Dec-2012, 10:06

Wow, I jumped into a lively debate here, but what I was hoping for was some discussion about the 8x10 digital camera sensor that NYC photographer had built for himself to replace the polaroids he had been using prior to shooting the film. Forget his name but it started with an F... Feldman? Fieldman? hmm maybe one of you would know. There are sensors built that large for military and astronomy. I think the only way to answer the resolution test is for one of you guys to have one of these cameras built. ;-)

EH21

20-Dec-2012, 10:19

I forgot to add that so many tests like this concentrate on the part where the scanned file is compared to the digital file or sometimes even better, large prints made from each are compared, however in practice its much harder to get a good capture with the 8x10. I mean there is focus issues, wind and shake and just more things to go wrong with the film shot in a big camera that you don't have with the MF cameras or tech cameras and digital backs. But mentioning that I fall into the detail trap, cause the resolution is just but one facet of the image. The LF film look it just totally different even if not sharp.

Andrew Clearfield

20-Dec-2012, 10:51

But how small do the pixels have to be on an 8x10 sensor to achieve some given standard of performance? If that standard is defined by the lens, we could speculate something like 100 pixels/mm (for lenses that resolve 50 line paris/mm--a high standard for 8x10 applications). That's about 500MP. But that exceeds the requirements of most photographers (and, frankly, most lenses used on 8x10 cameras) by a good ways.

Thanks for your most helpful reply. I just wanted to point out that, while it may be theoretical overkill, most quality print aficionados see an 8x10 contact print (or whatever size negative you have)—in a blind test—as having both better definition and a longer scale than either an optical enlargement of no matter what standard, or a digitally-produced image. So part of what is behind my question is, What is the scientific basis for this? I've always speculated that, as a scanning and integrating "device," human vision is able to integrate more data than it appears to be able to perceive in any instantaneous test, which might explain the phenomenon.

BetterSense

20-Dec-2012, 11:09

If you're curious about equivalence, read this white paper very carefully.

You'll see why it's actually quite easy to equal 8x10 in situations where you need a lot of depth of field, and almost impossible to do so in situations where you don't.

This theory is born out precisely in tests like the one discussed here.

The test shown actually disproves your point. They used a different aperture on the large-format camera, and then said that the large-format camera had less DOF.

So here are the aperture settings that should have been.

f/11 on IQ180 should be f/53 (or f/45⅔)
f/16 on IQ180 should be f/77 (or f/64⅔)

But we actually used f/32 on the 8×10 and so this has a large impact on depth of field.

Depth of field is only different for different formats if you open the lens more (which is commonly done, because bigger formats are slower).

Of course, in practical photography, speed is an issue. But instead of saying "smaller formats have more DOF" you should really say "smaller formats are faster".

Your other link gets it right

For this reason, it is impossible to tell the images apart. Which in turn means that sensor size is not an image property...

A camera setting is fully described by either of the following parameter sets (including exposure time t):

t, d, FoV, #MP (4 independent parameters)

Bruce Watson

20-Dec-2012, 12:03

I am curious to calculate how large an array of sensors would be needed to be able to record as much information as there is on a well-made and properly processed 8x10 negative.

It would have to be... 10x8 inches, now wouldn't it? There's a lot more to an image than simple resolution. If you want to capture the tonality, the DOF, all that stuff that makes a 10x8 image a 10x8 image, then you're going to have to have a 10x8 sensor. I would have thought that was painfully obvious. Maybe that's just me doing LF for so long, IDK.

Terry Christian

20-Dec-2012, 12:19

A "pixel" has no direct analog equivalent. What we think of as a "grain" of silver is actually a cluster of silver molecules that can vary in size and particular sensitivity. The "1 pixel = 1 silver grain" idea doesn't quite hold up.

Comparing digital to analog media is like comparing apples to oranges. How big does a digital 8x10 equivalent sensor have to be? As big as possible. ;-). Possibly of the same importance is the question of digital achieving the same dynamic range as analog. As it is today, digital is most like slide film in its narrow range.

rdenney

20-Dec-2012, 12:37

I suspect the tonal smoothness is more the cause of the contact-print effect than resolving power--though the usual assumptions about the sharpest thing people can see are generalities that don't encompass all the subtleties. Ctein insists that people can see the effects of high printer dot density without a microscope when comparing prints side by side, but I suspect that if they were separated by more than a few inches that would be less the case.

It's a false choice, though. All digital files have to be printed by a digital printer (or a laser lambda) which is subject to its own resolution limits, while a contact print is only limited by the resolution of the paper emulsion itself. Thus, it may be impossible to compare that digital file with a contact print meaningfully.

I find it interesting that other tests have shown that differences in actual prints do not vary linearly with size. The explanation that the orderly array of pixels has resonances that differ in the frequency domain from the random array of silver grains is something that seems obvious now but I hadn't thought about it before.

Rick "suspecting nobody designing digital equipment will ever define requirements necessary to match a contact print" Denney

BetterSense

20-Dec-2012, 14:00

If you want to capture the tonality, the DOF, all that stuff that makes a 10x8 image a 10x8 image, then you're going to have to have a 10x8 sensor.

This is actually incorrect, but you are free to think what you will.

Andrew Clearfield

20-Dec-2012, 15:09

A "pixel" has no direct analog equivalent. What we think of as a "grain" of silver is actually a cluster of silver molecules that can vary in size and particular sensitivity. The "1 pixel = 1 silver grain" idea doesn't quite hold up.

Duh. Anyone who has ever worked with silver images at significant degrees of enlargement knows that all too well. The question is, under specified conditions, what average size the actual grain of silver would be, and how that would compare to the unmodified pixel as captured by the sensor. Because in both cases these will be the absolute limit of resolution.

Possibly of the same importance is the question of digital achieving the same dynamic range as analog. As it is today, digital is most like slide film in its narrow range.
Now this is indeed important. However, I'm not sure that high-quality digital is as limited as you state: I seem to remember slide films having dynamic ranges more in the 5 to 7-stop range at the maximum (the sharper, more contrasty slower films are less, which is why they have so little exposure latitude) while the manufacturers of top-of-the-line sensors claim they have a range of up to 12 3/4 stops. As we all know from learning the Zone System (?? I hope!) b&w film has a total usable range of about 14 - 15 stops (pyrocatechin, anyone?) but as a practical matter, no more than nine stops in the final print. (Zone 0 means that no density was recorded).

Andrew Clearfield

20-Dec-2012, 15:58

Thus a color pixel is more like an 18 to 20 µm pixel, with B&W possibly less, depending on the specific algorithm used. This bayer demosaicing then reduces the effective sensor resolving capability by up to a factor of 10 for color and somewhat less for B&W. . . .

Fine grain film would typically have silver halide grains of about 0.8 µm diameter with a range of say 0.5 to maybe even 1.5µm. When irradiated by photons say as few as 1 or as many as 100s' the latent image increases in area by several times . . . Under low light conditions the latent image may still be close to the silver halide dimensions - obviously very high resolution . . . At the other end of the density range, heavy illumination from the scene highlights, the larger latent image grows even larger upon development forming silver clumps up to say 20 µm in diameter. So the resolving power of film is a sliding scale depending on the amount of original exposure and modulated by the development sequence.

Does this mean we can take the typical granularity of fine-grain film as ranging from, say 1/8 the size of a pixel (four silver halide molecules of 0.8μm) to 20μm, which is the same size as a color pixel, or slightly larger than a b&w one? Wouldn't this be the theoretical limit?

Of course, then the magic of computer algorithms intervenes, and we end up with something which looks considerably sharper than the grainiest part of a silver image. But it sounds like it is still nowhere near the maximum resolution obtainable in shadow detail—which would explain the much-vaunted ability of the very large negative (especially when contact printed or scanned at very high resolution) to "see into the shadows."

paulr

20-Dec-2012, 22:40

Does this mean we can take the typical granularity of fine-grain film as ranging from, say 1/8 the size of a pixel (four silver halide molecules of 0.8μm) to 20μm, which is the same size as a color pixel, or slightly larger than a b&w one? Wouldn't this be the theoretical limit?

To understand the diferences you'd have to compare both MTF curves and S/N ratios between the film and sensor in question. What you commonly find is that the finest grained, highest resolving films have extremely high extinction resolutions, whcih can be difined as the finest detail the film can visibly resolve before its contrast drops below the noise floor. This is what you attempt to measure with old fashioned resolution charts. The trouble is that detail this fine, or anywhere close to it, looks so bad that it's not photographically relevant, unless you're operating a spy satelite. A good digital sensor has a lower extinction resolution, but extremely high contrast and s/n ratio right up to its limit.

This is why it's possible in some cases for a digital sensor to make a better looking medium sized print than a piece of film, but a worse looking mural size print.

The other issue you'll see addressed in the links I posted is diffraction. An 8x10 camera with a good lens can resolve a stupefying amount of detail at high MTF if the lens is used near its optimum aperture (typically around f8 or f11). But it will have extremely shallow depth of field compared with a much smaller format that covers the same field of view. If you stop down to equal the depth of field of a medium format digital camera (with a lens set to its own optimum aperture) you'll be introducing enough difraction to level the playing field.

There are parallel situations where a top end full frame dslr can have advantages over medium format digital.

Martin Courtenay-Blake

21-Dec-2012, 03:03

Getting away from the science and math of this thread a similar type of argument/discussion has broken out in the 35mm arena since the arrival of the full frame sensor Nikon D800/800E. Advocates of the digital form of image making are stating that this camera not only produces superior images than traditional film (including 50ASA velvia), but challenges the superiority of medium and even large format film. Interestingly most of the wags supporting this view have never actually taken a picture with a film camera but are very sure that what the marketing men are saying must be true. Personally I would like to see a direct comparison between a virtually identicle image produced on the D800 and a Velvia (or better still Tech. Pan.) image scanned at the very highest possible resolution (i.e. a resolution theoretically higher than the grain structure of the film), with the same amount of post scan processing, before making a decision as to which is best. My gut feeling is that the digital image would run things pretty close on the detail front whilst there would be subjective differences which would cause one to fall into one or other camp.

The image quality from the highest resolution Hassy backs, 200MP I believe, do make remarkable images and may well be as good or superior to film. I would like to see some direct comparisons. But it is clear that it would take a hell of a lot of pixels to challenge film even on 5X4 let alone 10X8, the suggestion of gigapixels probably being right.

But, is it really just a matter of resolution for resolutions sake. A digital image is going to be different in many, possibly subtle, ways from that produced by film. Everyone has their favourite film (and developer in the case of monochrome) and it's these personal decisions we make when producing an image that makes them different and, in many cases, identifiable to a particular photographer. My concern with camera to print digital photography is that we may end up with vast numbers of extremely high resolution images that simply look the same.

Leigh

21-Dec-2012, 06:27

If you make this assumption and that assumption,

and use this fudge factor and that fudge factor,

and this equivalence and that equivalence,

you can make the numbers come out however you want.

That doesn't make them valid.

Remember... All of these supposed comparisons were developed by folks interested in selling digital imaging.

- Leigh

paulr

21-Dec-2012, 10:45

Getting away from the science and math of this thread a similar type of argument/discussion has broken out in the 35mm arena since the arrival of the full frame sensor Nikon D800/800E. Advocates of the digital form of image making are stating that this camera not only produces superior images than traditional film (including 50ASA velvia), but challenges the superiority of medium and even large format film. Interestingly most of the wags supporting this view have never actually taken a picture with a film camera....

And their claims are being dismissed by people who have never seen the results of a d800. Not too many people have looked side by side.

I can speak somewhat knowledgeably about this particular comparison, having worked exclusively with 4x5 for fifteen years, and getting a d800 five months ago. The comparison actually parallels the one between 8x10 and a Phase One back quite closely.

Short version: I've been making 12"x18" proof prints from d800 files that are to my eye better—by a lot—than anything I've done or seen with 4x5. This is true only for one of the two projects I've been working on. The other is hand-held, in low light, and those prints look every bit like what we expect from small camera pictures.

I can attribute the superiority of the d800 prints primarily to the depth of field / diffraction calculus that I've mentioned before, and to the very high modulation and s/n ratio of the sensor. It also helps that there is no intermediate scanning or enlarging step to degrade the image.

It's possible that 12x18 is close to the sweet spot for this sensor. I would fully expect that if we made comparisons at larger and larger sizes, we'd hit a point where the 4x5 prints start to look better than the d800 prints. I can guess where that point would be, but haven't been able to test it. The biggest print I have for comparison is a 40x50 darkroom print from one of the black and white 4x5 negs. I think this size would be an interesting test for the d800. It would likely look worse than print from 4x5, and would certainly look different.

paulr

21-Dec-2012, 10:47

My concern with camera to print digital photography is that we may end up with vast numbers of extremely high resolution images that simply look the same.

This would be a failure of imagination, not technology. The files produced by these cameras are almost infinitely malleable. Unlike film, it's not possible to attribute a "look" to a digital sensor.

bob carnie

21-Dec-2012, 10:47

I just finished making some 40 x60 inch prints from the D800 that pretty much backs up the above.

rdenney

21-Dec-2012, 11:51

Remember... All of these supposed comparisons were developed by folks interested in selling digital imaging.

Not really true. But in any case, there are as many comparisons made by those trying to refute the validity of digital imaging as by those trying to promote it.

What I rarely see is a comparison based on what the photographer is trying to achieve, rather than what the equipment is capable of achieving.

For me, my digital prints usually come closer to my visualization than do my darkroom prints. Partly this is a limitation on technique. But it is also a limitation on what kinds of infrastructure I can support in my home, and what kind of time I can devote to photography. The compromises on time and infrastructure substantially limited what I could do in a darkroom, and I had more time in those days than I do now. The power tools give me more productivity up to a level of craftsmanship that is pretty good, versus the potential for craftsmanship I could never achieve anyway using non-power tools. In practice, it means I attain a higher level of mediocrity.

As to this issue, the question that I think is relevant is what would it take in a digital system to achieve what one can achieve using an 8x10 camera, and I think trying to answer that in absolute terms is intractable. I agree that to do what an 8x10 camera does, it would take an 8x10 sensor. Once we cross that bridge, we can have conversations about how many pixels that sensor needs to achieve a given outcome, but I sense that not everyone is prepared to cross that bridge.

For me, the point of large-format resolution is to maintain the sense of endless detail. I want the stuff in focus in my prints to give the effect that the closer one looks at the print, the more detail there will be. I'm prepared to refuse them a magnifying glass (beyond the bottom lens in their bifocals), but up to that point, I want it to hold up. When the detail turns fuzzy, it breaks that willingness to believe that the detail is endless.

This is not about sharpness at all. I have compared a fine duotone reproduction in Yosemite and the Range of Light to a Special Edition print of the same topic (Dogwoods...), and the duotone is made apparently sharper by the edges created by the dot pattern. But the Special Edition print retains the sense of endless detail. I suspect Paul is close to the mark with discussions of MTF, signal to noise ratio, and extinction limits, but I also suspect that what achieves the goal is a rather subtle interpretation of these, not microscopic comparisons or mathematical speculations.

Rick "who does the best he can with what he can live with" Denney

Martin Courtenay-Blake

21-Dec-2012, 11:56

I am actually getting the cash together to purchase a D800. I am sure it will be a satisfactory advancement over my current Pentax DSLR, and I'll be able to use my existing AF Nikkors. When I first considered the D800 I had a search on the interweb and found a superb comparison between the D800 and the then current EOS 5Dll. The Nikon seemed to have the edge over the Canon, mainly because of increased resolution. Both were then compared to images from a Pentax D645.....absolutely no contest. The images from the Pentax looked amazing. In pure pixel count the Pentax has only 10% more than the Nikon but it was like comparing a 35mm transparency with 5X4, the difference really was that great. The reviewer put it down to the actual size of the pixels and their ability to record detail more accurately.

If i could justify the price the Pentax would be my camera of choice. I don't know how long it will be but if a company was able to produce a 5X4 digital back using the same principals as Pentax have with their D645, i.e. pixel count per se is not the only way to produce stunning digital images the possibilities could be very enticing.

rdenney

21-Dec-2012, 12:03

If i could justify the price the Pentax would be my camera of choice.

Mine, too. I have at least a couple dozen lenses that could be used on one, including some that work really, really well. And including shift lenses in 45, 55, and 75mm.

But it won't make the same images as my 4x5, let alone an 8x10. None of the lenses has tilt, and none of them have that clean lack of lateral color and distortion that large-format near-symmetrical lenses provide (that's the secret in the sauce for the Mamiya 7).

But I still want one. Can't afford it, of course.

Rick "still buying lottery tickets" Denney

paulr

21-Dec-2012, 12:43

Remember... All of these supposed comparisons were developed by folks interested in selling digital imaging.

I've never had anyone even ask. Oustide these little technical communities, it's hard to find a collector or editor or curator who cares what kind of camera you used, unless there's something especially interesting about it.

paulr

21-Dec-2012, 12:55

rdenney

21-Dec-2012, 13:18

The pentax 645d does produce a higher quality image than the Nikon, but I'd disagree that it's night and day. Comparing files on screen at 100% doesn't always clue you in to how prints are going to look. The extensive comparison I saw between the Nikon and the phase IQ160 showed a clear advantage to the Phase back (as you'd hope, for that kind of money). But the guy conducting the tests said that at print sizes below 30 inches there really wasn't much difference to talk about.

The Nikon can't quite match the resolution of the Pentax but beats it in dynamic range. Which quality is more important depends on what you're up to. For that matter, negative film still beats every digital sensor in dynamic range ... how much dr you need can help with any of these camera decisions.

The test I saw that few people were talking about is the Nikon vs. the pentax 67 and fuji 6x9 film cameras. This was a complete slaughter by the nikon, in terms of resolution, although the person conducting the tests liked working with film enough that he kept the fuji as his main camera.

It is important to note just how demanding that much enlargement is of the lenses, however. I've seen frankly stunning prints made using a D800, but that photographer (who works at the local store and sells the things) also said that there are only two or three lenses in the Nikon lineup that can explore the capabilities of that sensor, when used at optimal aperture and with impeccable technique. I suspect the technique would still be required for the 645D, but the lenses would get by with two-thirds the resolution to get the same result. For the biggest prints I can make, that's the difference between a lens that must resolve 120 line pairs/mm versus a lens that resolves 80, both at good enough MTF to achieve good MTF at 5 line pairs/mm on the print. That difference cuts across some very significant boundaries, in terms of what available lenses are able to do.

But I'll stop there--this if off-topic.

Rick "back to the digital equivalent of 8x10" Denney

paulr

21-Dec-2012, 13:32

I don't think that's off topic at all, Rick. If you're comparing different formats it only makes sense to look at what lenses are available for each.

This point gets lost when people compare various film cameras to the medium format digital backs. The Schneider and Rodenstock lenses available for MF technical cameras are by a large margin the best lenses you buy. They're better than anything that's ever been made for large format, and better than anything that's ever been made for small format.

They're better than what you can get for the pentax 645, too, but those pentax lenses are still excellent.

This may be changing in the small format world, where there's all this renewed r&d energy. Nikon and Canon are both making some spectacular lenses (alongside a lot of not so spectacular lenses). Schneider and Zeiss have just announced small series of lenses for the upcoming year, designed to take live up to the new generation of sensors. They will not be cheap.

I've been using a Schneider PC Super Angulon shift lens on my Nikon. It's better than any other small camera lens that I've used, and better than my Schneider large format glass. But it's nowhere near as good as their current MF digital offerings. The new stuff coming out this year for the small cameras may change this.

Brian C. Miller

21-Dec-2012, 13:53

Current thread summary:

Film and digital are different.
Size matters.
The quality of the lens required increases as the density of the sensitized material increases and the format area decreases.

"But you still have not told me in which direction to point the camera - and this is what matters."
-- Ralph Steiner

(Oh, and as for thread topic, based on my own experimentations, the digital sensor required to equal 8x10 resolution is 320Mp.)

paulr

21-Dec-2012, 14:03

Current thread summary:

Film and digital are different.
Size matters.
The quality of the lens required increases as the density of the sensitized material increases and the format area decreases.

Yes, but there are many other interrelated factors, including iso (which correlates directly with s/n) and aperture/diffraction.

So in theory, size does not matter ... it can be compensated for by adjusting other factors.
In practice, there's a limit to how far it's possible or practical to adjust those factors, based on what's available, affordable, useable, etc...

It's safe to say size matters, but not in an absolute sense. And the degree to which it matters is changing all the time.

Brian C. Miller

21-Dec-2012, 15:38

Yah, sure, there's interrelated factors. It's just that in any comparison of apples, oranges, bananas, egg plants and potatos, that each should be tested at its optimal sample. So all of them are freshly harvested at their ripest state, and then prepared. Comparing 8x10 pinhole using Delta 3200 to an 80Mp back with Rodenstock Digaron might be valid in some sense, but not for über-resolution tests. The best test that I've seen so far has been Tim Parkin's On Landscape, #28 (http://www.onlandscape.co.uk/issues/lgb-0028/), comparison, but the part of it which was the real eye-opener wasn't the resolution, it was just how difficult the digital camera view camera was to operate, in comparison to the film view camera.

As for the "biggest print," I have a panorama from a 645 that looks good at nearly 4ft across. I have seen a 6x9 (cropped to 6x8) from Delta 3200 that looked totally fabulous at 30x40. And yes, both of those prints are nose-sniffed by the viewers. (I sure got close to the latter print!) With the right algorithms, I really don't know how large a digital print would have to be before the viewer rejects it in favor of an optical film print. That really depends on the subject of the photograph, doesn't it?

Andrew Clearfield

21-Dec-2012, 20:43

On Landscape[/i], #28 (http://www.onlandscape.co.uk/issues/lgb-0028/), comparison, but the part of it which was the real eye-opener wasn't the resolution, it was just how difficult the digital camera view camera was to operate, in comparison to the film view camera.

Much thanks for the link to this article: it brought a lot of different analyses and viewpoints together. (Also thanks to everyone who has been contributing to this thread!)

I'm not about to sell my 8x10 Deardorff or my 4x5 Wisner, either, although I do worry that film, paper, and chemicals may become almost impossible to find (or absurdly expensive, due to lack of demand—as Amidol already is) long before digital technology has advanced to the point where there is absolutely no advantage to the silver image. (Which would be a terrible moment I don't look forward to, in any case, especially since none of these engineers have any concern whatever about archival stability!) I'm not going to throw away my Sony DSLR, either, although I absolutely loathe the obsolescence factor that is inherent in digital technology. Buying a new lens when and if something really superior comes along is one thing; having to buy a new camera, or sometimes a new system, every few years quite another.

(It's not really relevant to a LF forum, but I like the attitude of the folks at Leica: it may not sell many cameras, but I love the fact that I can mount the latest APO-aspheric-whatever on my M3 or even my museum-piece IIIf (or if I want a "vintage camera" look, even a truly ancient Summar or Elmar on an M9—not that I have one); when they finally went digital, they made sure that the flange-film plane distance was still the same, so interconvertibility remains. Recently, I had to get a six-year-old Sony repaired because, of all goddamned things, the plastic On/Off switch broke off. At first, they weren't sure they still had the part for an 'obsolete' camera.)

Greg Miller

22-Dec-2012, 06:12

paulr

22-Dec-2012, 13:04

...just how difficult the digital camera view camera was to operate, in comparison to the film view camera.

Between the bad ergonomics and the price, I think the medium format digital tech cams are still basically a technology demonstration, except with a relatively small number of specialized photographers. I'm certainly not ready to get one. But they represent some impressive writing on the wall.

With the right algorithms, I really don't know how large a digital print would have to be before the viewer rejects it in favor of an optical film print. That really depends on the subject of the photograph, doesn't it?

Sure, it would depend on a lot of things. But I think we'll see it as a general pattern. At some size, all else being equal, a print from film will start to look better than a film from a sensor. Just because film images degrade in a way that most people find less offensive when the degradation gets extreme. This already seems to happen at a pretty big size, if you're using current sensor technology.

BetterSense

22-Dec-2012, 17:50

At some size, all else being equal, a print from film will start to look better than a film from a sensor. Just because film images degrade in a way that most people find less offensive when the degradation gets extreme.

If you are talking about pixelization, that can be cured by up-rezzing (with a good algorithm). You can even add fake film grain if you want. Nobody should ever see pixels in a digital print, no matter how large and how low-resolution.

paulr

22-Dec-2012, 20:42

I've never seen pixelization in a print. When a digital print gets too big, you become aware of the resolution limit but the tones stay smooth. An analog print has a normal MTF curve, where the contrast simply drops off as the resolution gets higher. Fine details get more and more ghost-like, and eventually drop below the noise floor. Digital printing processes work more like a brick-wall low-pass filter. Contrast of the fine details remains high, but there's a hard frequency limit. Combine this with the typical low to extremely low noise floor, there's nothing for the detail to disappear into. When the detail in a print is limited by huge print size to under a couple of line pairs/mm, and there's no noise to fill in the gaps, the image starts to look like plastic. Kind of a Barbie version of the world. An artificial look, not pretty for most purposes. Adding noise can help, but I don't think it solves the problem completely. I have not experimented much with this, because my current digital camera can print up to 40 or 50 inches wide before that plastic degradation becomes visible, and I haven't tried to experiment with anything bigger.

ChuckT

2-Jan-2013, 07:58

I saw pixelization in a print only once - in a demo of the Kodak Profesional Photo-CD system (in 1996?). We made a 4x6 image into a print that was 10" wide by (whatever) high. And brother were those pixels obvious. That Dye tranfer printer was sharp!

Other than that I doubt if ANY printer would do that today. The printer interpolation/smoothing processes simply will not let that happen.

cvt

SergeiR

2-Jan-2013, 08:30

Ah.. the joy of comparing digital to film.. We at it again :)

You cant escape simple physics - in projection - size of receiving media - matters. To get smoother gradients you got to keep increasing mp on 35mm well past point that optics can resolve it even now. But even if it were on par - there always will be certain "something" about film photography. Nostalgic element that would keep pulling people in.

And there is different look from different sensors and CPUs on digital cameras.. Its less now when Sony dominated dSLR sensor world but its still there. More so between Dalsa and Kodak in Leaf / Phase One / Pentax in MFdb world. Could be almost nullified by PPs of course.

Greg Miller

2-Jan-2013, 08:48

... But even if it were on par - there always will be certain "something" about film photography. Nostalgic element that would keep pulling people in.

And there is different look from different sensors and CPUs on digital cameras.. Its less now when Sony dominated dSLR sensor world but its still there. More so between Dalsa and Kodak in Leaf / Phase One / Pentax in MFdb world. Could be almost nullified by PPs of course.

Same as comparing one film to another film. Each film has its own "something". And each person needs to shoot each film to know how they stack up for them personally. Nothing new with digital.

SergeiR

2-Jan-2013, 11:55

Same as comparing one film to another film. Each film has its own "something". And each person needs to shoot each film to know how they stack up for them personally. Nothing new with digital.

Absolutely :) Magic bullet ;)

Sylvester Graham

2-Jan-2013, 18:52

Is 3,200 megapixels enough?

http://www.uni.edu/morgans/astro/course/Notes/section1/lsst.jpg

D. Bryant

2-Jan-2013, 19:00

Huh? Don't know where you got that. Sounds like advertising hype.

A pixel sensor is a physical structure that includes a lens, the active sensor element, and electrical connections.

An image forming element in film is a cluster of molecules.

------

I typically scan my 8x10 film at 2400 ppi, so the 8" side is 8 * 2400 = 19,200 pixels, and the 10" side is 24,000 pixels.

The image area is therefore 19,200 * 24,000 = 460,800,000 (461 Megapixels).

A good drum scan @ 9600 ppi = 8 * 9600 * 10 * 9600 = 7,373 Megapixels = 7.4 Gigapixels.

- Leigh
No one in their right mind sum scans 8x10 firm @ 9600 PPI.

Greg Miller

2-Jan-2013, 19:16

Absolutely :) Magic bullet ;)

Huh?

Vaughn

2-Jan-2013, 19:21

No one in their right mind sum scans 8x10 firm @ 9600 PPI.

No, that would be the left mind that does that. The right mind uses an old thrift store Diana camera. :)

But anyway, it is a question I have gotten asked out in the field with my 8x10 a few times (usually in Yosemite Valley). So I will answer "Between a third and a half gig." May or may not be 'correct', but it will sound impressive!

Bernice Loui

2-Jan-2013, 20:30

There are no technological " Magic bullet (s)"... that can ever replace human creativity, skill of their craft and sense of artistic expression.

There appears to be a belief that more megapixles or larger film size will provide the special magical widget that will create great expressive images... Simply NOT true.

Film will result from a different image than digital, this is built into the foundation of each technology. Regardless of the digital post processing, number crunching or any other means of data alterations.

Accept each for their strengths, problems, limitations and possibilities as they are just another tool in the imaging tool box.

For a very long time there has been a lot of resources and expertise to try and make video look like film. This included highly complex anti aliasing filtering built into the video electronics system and optics (special lenses which are really expensive built by Fujinon, Canon, Angenieux and.. that incorporated part of the anti aliasing filtering) none of these system were able to completely duplicate the look of film.

To this day, the debate of electronic imaging -vs- film rages on in the video and cine world.

Part of the difference comes down to the basic physics of how film -vs- digital image sensors function...

Bernice

Absolutely :) Magic bullet ;)

Greg Miller

2-Jan-2013, 20:40

Well said, Bernice.

Bernice Loui

2-Jan-2013, 20:51

Consider the instant gratification of digital imaging -vs- film... and if the image created does not meet the expectations of it's creator, the image is erased and more taken in an effort to meet the creators expectation.

This could be another reason why so many image creators are lured into the fantasy of using digital / electronic imaging to meet all their imaging needs and every expectation..

Bernice

paulr

2-Jan-2013, 22:35

Ah.. the joy of comparing digital to film.. We at it again :)

If it's an informed discussion, it will be different every year! Digital technology changes at an incredible rate. The pros / cons and cost / benefit of using one medium over the other are constantly shifting.

PhiloFarmer

3-Jan-2013, 13:03

Is 3,200 megapixels enough?

http://www.uni.edu/morgans/astro/course/Notes/section1/lsst.jpg

I'd love to know more about that sensor....can you share??

Sylvester Graham

4-Jan-2013, 00:17

I'd love to know more about that sensor....can you share??

http://www.lsst.org/lsst/gallery/camera

Space: the final frontier.

Personally, I don't care much for pixel counting, and don't count sharpness as the most important quality of a photograph. That being said, I have my doubts about silver halide photographs being able to quantitatively out resolve digital, much less a bank of multiple sensors like the one above. Now, other chemical light recording methods, like photolithography, are a different story I think.

paulr

8-Jan-2013, 17:55

Would it be out of line to get practical? Anyone in NYC using an 8x10? It would be fun to do some side by side comparisons and compare results in prints.

timparkin

9-Jan-2013, 05:29

timparkin

9-Jan-2013, 05:38

Short version: I've been making 12"x18" proof prints from d800 files that are to my eye better—by a lot—than anything I've done or seen with 4x5. This is true only for one of the two projects I've been working on. The other is hand-held, in low light, and those prints look every bit like what we expect from small camera pictures.

That depends on your scanning and sharpening technique. yes the D800 has high acutance at just the right lpmm for a 12x18 print but you can also sharpen a 4x5 scan to give the same acutance but you must have a very smooth scan at a high resolution to avoid creating noise. This is why it's often best to scan 4x5 on a drum scan with a very large aperture because it provides a very smooth result albeit one that loses detail. This then allows you to play with acutance without too many side effects.

I've tested the D800 against a 'well scanned' 4x5 and printed them both and the D800 nowhere near the 4x5 to be honest. And by nowhere near I mean it's like comparing the D800 with an 80-100Mp camera.. And if we're testing black and white then the gap is even larger.

Tim