Sharpness of Sheet Film vs Roll Film

[Christopher Perez]

This is a rather interesting conversation.

Film is the resolution limiting factor. Period.

I'm thinking the trollup who looked at the 8x10 images didn't know her head from a hole in the ground. But... I suppose that early chrome material might not be as sharp as newer films. Still, same emulsions and all that for what's availble today, regardless of the format or base thickness..

Regarding lenses, I see no difference on film between the various formats. Sure, a couple of my 120 format cameras can lay down 120+ l/mm to film and give outstanding contrast in the process (to give that enhanced "apparent" sharpness "look" and "feel"). I'm sure that well manufactured 35mm glass can do the same. But by the time an image is enlarged to 11x14, grain starts showing up. Even when using TMax100. At 11x14 enlargement and greater LF wins the day in terms of resolution, "apparent" sharpness, and lack of grain.

[Ralph]

Very interesting discussion.

1) I think Christopher Perez's tests (on his website) show that the best smaller-format lenses almost without exception have a higher resolving power than the best larger-format lenses (Christopher was able to reach 120lp/mm with the Mamiya 7 80mm, and I think the best he could do with a large-format lens was in the mid-70s). None of this comes as a surprise to 35mm shooters, Leica and otherwise, who know that attaching a medium-format lens to their camera rarely if ever results in the same sharpness as a dedicated, sharp 35mm prime does. If it did, there'd be a run on used Hasselblad and Rollei lenses by full-frame digital SLR users. (True, in large format, certain govt.-spec lenses and some of the new small-image-circle digital-intended lenses are exceptions to the automatically-lower-resolution rule, but neither is in common use, for example, by 8x10 photographers.)

2. My reading of the original post was not "Which will produce a better 16x20 print?" -- of course large-format will beat 35mm every time -- but "Which is sharper PER SQUARE INCH of film?"

That's what the "trollop" saw on the light table, and I think it's a valid question: if you took a 24x36mm section of 8x10 film and enlarged it, would it outresolve a similarly enlarged full frame of 35mm film shot using a prime lens? I can't think of any circumstances in which it would.

Like the original poster, I've been struck by the "softness" of large format sheet film, despite doing everything possible to ensure flatness and sharpness: Rodenstock-Apo-S lenses, Quickloads with even pressure plate to prevent film sag, focusing bracketing, and all; none seem to result in anywhere near the resolution *per square inch of film* that a good 35mm or medium-format combination does.

It is only because we don't usually enlarge as much as 35mm-users do (for example, a 16x20 from 35mm is equivalent to enlarging 8x10 film to something like 11-feet x 14 feet) that we don't see the softness.

[Christopher Perez]

Furthering the interesting discussion...

I think it would be a valid test to confirm for our selves that thickbased film can resolve what the thin-based films can. In my tests it does. For a std. 6:1 contrast scene using the finest grained B&W or color films (all modern) the limit is 140 l/mm. Per square inch, film resolution is the same regardless of base thickness.

Now lets consider what a 60 l/mm LF lens will do compared with a 120 l/mm MF optic: In my experience there is NO WAY that a trollop with a 8x or 10x loupe would be able to tell the difference between the two. If said trollop could, she must have truly incredible non-human eyes. Or other contributing assets.

In my situation, I made absolutely certain that the LF groundglass and film holders are in spec. In fact, my 4x5 test system is within +/- 0.004 mm's. It helps. I couldn't believe how out of spec some of my cameras were out of the factory.

In my limited case I really can't tell the difference between my MF and LF images on film. It takes at least 20x for me to even _begin_ to tell a difference. Enlarged print? That's another matter and LF wins the day due in large part to surface area.

[paulr]

"Film is the resolution limiting factor. Period"

This makes no sense, especially in large format or in black and white. Go to the Kodak website and download MTF charts for tmax films. Their mtf response, into the high frequencies, goes way beyond what any photographic lens can approach. A summary of Tmax 100: MTF 100% at 50lp/mm; MTF 70% at 100lp/mm; MTF 50% at 120lp/mm.

In fact, the idea of limiting factors is conceptually incorrect. We don't work with equipment and materials that act as low-pass filters (cutting off resolution past a certain point). They act as contrast softeners, exerting greater influence as frequencies get higher. It's never a weakest-link-in-the-chain proposition. Every link in the chain degrades the image somewhat. The total performance of an optical system is determined by multiplying the amount of degradation of every link in the chain. If we are looking for a resolution limit, it would be the point where contrast of fine detail becomes indistinguishable from the contrast of the noise (including grain). But how high this frequency is tells us next to nothing about the perceived quality of an image. For that we need to look at the contrast of image detail at key frequencies on the print, which are significantly lower than the highest frequencies our eyes are capable of detecting.

"Regarding lenses, I see no difference on film between the various formats."

My guess is that you're comparing using resolution numbers, which are a contrivance that tell little about how a lens performs in real life. If you compare MTF charts of small, medium, and large format optics, you will see a significant increase in performance as the format size gets smaller. It's just much, much easier to design a lens for a 40mm image circle than for a 150mm image circle. In the end it doesn't matter much ... a 30% loss in modulation is insignificant compared with a 300% reduction in enlargement factor.

To see for yourself, go to the Schneider site and compare the Apo Symmar L series with their digital (medium format) series. You can also go to the Zeiss site and compare hasselblad lenses to contax lenses. Same phenomenon in both cases. It's NOT that the companies are trying less hard with the larger formats. They're doing the best they can within the design constraints of each format.

In the larger scheme of things, I suspect that film flatness and diffraction make a bigger difference than lens performance, but it's all a factor.

[Mike Butler]

I agree with Christopher.

I'll never forget the first time I saw an 8x10 transparency of a food shot on the light table when I started in magazine editorial 18 years ago..

But improvements in color emulsions (grain, contrast, saturation, sharpness) probably did more than anything to displace large format in favor of medium format. The quality of the film simply made the smaller film size good enough for 4-color separation (unless you needed the creative perspective control of LF, of course.)

Viewing distance is another factor in sharpness. That art director looking at the 8x10 through a loupe might have been akin to viewing a billboard from a couple feet away.

Subject matter can also affect the perception of sharpness. Even the highest-resolving films might look mushy if the subject doesn't offer a lot of texture or fine detail.

Best,

[Herb Cunningham]

I gotta get in this.
When i was a part time pro, in college, i used to shoot weddings with a speed graphic and #2 flashbulbs. That was with the stock graphic lens, I think a 150 f 4.5. Never any complaints about sharpness.

I also did the occasional portrait or bride's shots with a dagor, I think was 240mm, anyway it was all the bellows could do to get a good focus, I had to lock down the rail really tight so it would not move due to bellows tension.

I ran across an 8x10 from that era, the 1950's, that was enlarged on a condenser besler, a month ago. The print is so sharp that you can see the threads in the knots on the bride's veil. Any sharper than that, who needs it for ordinary photo work, i.e non scientific or military.

Many times I see people who know little about photography and less about art looking at an 8x10 or god help us a 16x20 at a distance of a few inches. Does not make sense. I think the human eye can only resolve ten or so lines per mm right?

[paulr]

"Many times I see people who know little about photography and less about art looking at an 8x10 or god help us a 16x20 at a distance of a few inches. Does not make sense. I think the human eye can only resolve ten or so lines per mm right?"

Just remember that sharpness and maximum resolution have almost nothing to do with each other. You could look at two prints, one that resolves 30 lp/mm (measured with a microscope) and another that resolves only 5 lp/mm, and the lower resolution one might look much sharper, if it preserves more contrast in the 1 to 5 lp/mm range.

It's also possible for negative A to look sharper than negative B in a contact print, but for B to look sharper than A in a 5X enlargement.

I'd also add that some people like to examine a print from a few inches away. Not just out of technical geekery, but out of fascination with
the sense of detail that goes on for ever. Makes me feel like a voyeur.

[Oren Grad]

Many times I see people who know little about photography and less about art looking at an 8x10 or god help us a 16x20 at a distance of a few inches. Does not make sense.

Remember that people have different purposes for their pictures. With rare exceptions, I don't make prints to hang on a wall. I make them to hold in my hand, and enjoy looking at subtleties in the way the image is rendered and grooving on, as Paulr put it, "the sense of detail that goes on for ever". "Good enough for normal viewing distances" isn't good enough for my normal viewing distances.

I think the human eye can only resolve ten or so lines per mm right?

The appearance of fine detail in a print is affected by MTF at frequencies quite a bit higher than 10 lp/mm. Even if you can't resolve separate lines, the information still makes a difference in the appearance of edges and of fine detail.

[Mark Sawyer]

Comparing 35mm to, say, 8x10, I would think the increased distance from the lens to the film would play a factor; less than 2" on 35mm turns into 12" on 8x10.

I doubt the quality of glass in 35mm lenses is better than in large format. If anything, the lower production numbers and increased cost of lf lenses, along with the more demanding nature of the people using them, would lead manufacturers to put their best glass into lf lenses.

I don't think the film base would have much to do with it, either; it's not between the emulsions of the film and paper, (or in the film/loupe/eye circuit). It might cause a slight effect of diffusion in illumination, like between prints from a diffusion enlarger compared to a condenser enlarger, but to a much lesser degree.

Diffraction, too, probably plays a part at small apertures...

As a b/w person, I prefer lf for the detail, but also for the tonal scale, control, and the working process.

[John_4185]

Grain can enhance perceived sharpness.