If LF looks so great on web, why can't it be done with 35mm or a DSLR?

[marshallarts]

I tried to sum it up in the title. An ignorant question at first glance, but hear me out. I've only dabbled in LF with my Graflex, but I absolutely love the images. The crispness and falloff towards out of focus is what really strikes me. I've been trying to find equivalent techniques with 35mm to achieve the same look but it just can't bring that LF look to the table.

I can see how a contact print from a 4x5 would be something of unparalleled resolution in comparison to an enlarged 35mm print, but what confounds me is how come a low res scan of such a print yield that same effect vs the 35mm scan. That is obvious but wouldn't that knock down the resolution theory as to what give LF that feel?

I've never been much of a printer or scanner so this very well could be a simple answer. Is there something else I'm not taking into account? I'm tempted to get very detailed into my question but feel it would sidetrack from this broader question of what makes LF a magical format.

[Ken Lee]

Can you show us some example images which illustrate the look you're referring to ? Otherwise it's just speculation on our part.

4x5 is at the small end of Large Format: try looking at an 8x10 or 11x14 contact print - or larger - to see how lovely a photograph can be.

Large Format represents the latest in 19th century technology: not only resolution, but exquisite tonal separation and smoothness. Once you enlarge past 2 or 3 times, these tend to diminish. Nowadays, people prefer large sensors for the same reason. To quote Ken Rockwell (http://www.kenrockwell.com): "This $300 used 4 x 5 is sharper than a new $3000 Hasselblad and worlds beyond a $5000 Leica, Contax, Canon or Nikon.".

Another term for it is over-sampling (http://en.wikipedia.org/wiki/Oversampling).

If you make a 3x enlargement with 35mm, you can get the same basic image quality - but a rather small print. Small cameras are ideal for certain applications, but they are not the Summum Bonum of image quality. Speed and convenience perhaps, but not image quality.

There is also the issue of focal length and depth of field or blur: while a normal lens on 35mm is 50mm, it's 150mm on 4x5, and 300mm on 8x10. A 400mm lens at f/16 has the same depth of field as a 200mm lens at f/8, and a 100mm lens at f/2. If you want to make a portrait with a blurred background, it's very easy with Large Format. With formats like 35mm or smaller, it can be a challenge.

Then there's perspective control and view camera movements which allow you to get virtually unlimited depth of field...

These issues had already been figured out in the 1800s.

Speaking of the 19th century, here's a site (http://www.piercevaubel.com/cam/) that shows many nice portable Large Format cameras going back to that time, made in the USA. There were many others made elsewhere too. Here's (http://www.largeformatphotography.info/forum/showthread.php?t=75204) a page that shows some forum members with their cameras. And finally, here's (http://www.largeformatphotography.info/forum/showthread.php?t=45775) a page that shows some LF cameras which people have built themselves.

[Greg Lockrey]

Yeah.... a low res scan is a low res scan. It doesn't matter if it's from a scanner or a camera......

[cjbroadbent]

I've wasted a good deal of time shooting the same subject in multiple formats (https://picasaweb.google.com/cjbroadbent/Formats?feat=directlink). At print level or on the printed page there is no contest. It is due to what Jim Galli calls "peaches and cream" (see his posts and Ken Lee's 5x7s).
The recognizable edge that large format has at web level is clearly due to the gentle way it drops focus. You can try with a Summilux or whatever wide open, but the focus transition never matches up to a big old piece of glass. I don't think resolution has much to do with it.

[marshallarts]

Thank you so much for your responses. Ken, I appreciate your lengthy response. My question was intentionally vague and broad because of the number of variables involved. Though I'm not as experienced with LF I have a firm understanding with the smaller formats and principles of photography and optics.

cjbroadbent, I think you honed into my question addressing the focus drop off. I can achieve a similar DoF (within limits) shooting wide open on a FF 35mm camera but even then the transition isn't the same as with LF. This has been what I've been struggling with. Trying to achieve that focus transition.

But as you said resolution doesn't seem to be the issue here since at the web level it's effectively low res.

Any other thoughts?

[Sirius Glass]

I have had printed 24"x36" C-41 prints from mulitple 35mm negatives. The right exposure ... et al and it looks great. Yes, from six inches I can see grain in some places.

However for serious [sirius] work I use the Hasselblad 6x6.

When I have become more consistant with 4"x5", that will be come the camera of choice for stationary or slow moving subjects. I just have to become "one" with each of the LF cameras.

Steve

[jnantz]

you can do the same thing with 35mm or mf
you just need a 35mm or mf roll film back
and put it on the back of your 4x5 camera.
i used to have a 35mm recommar kodachrome back
that i used for this exact purpose - and i used it for a while
with brass lenses and a graflex slr/ speed graphic ...

you can also do lf looking things with a holga+35mm
adapter ... none of these things cost much ...

good luck !
john

[Ken Lee]

Some lenses have more exaggerated dropoff than others. This will be true in the 35mm realm too. Lenses with fewer elements will generally give very pleasing results in this regard. For example, the 3-element Apotar in my 6x9 Agfa folder, gives lovely blur rendition. Tessars are 4 elements. I love their blur rendition. Heliars have 5 elements, and are designed to give exaggerated blur at wide apertures. These lenses are available for 35mm cameras.

The longer the lens, the greater the blur. If you shoot a 300mm lens at f/8 on a 35mm camera, it will have the same blur as when shot at f/8 on 8x10. The only difference will be that on a 35mm camera, it will have a very narrow field of view because of the small size of the film.

To get equivalent blur on a small film, you need a fast lens. 300mm at f/4.5 = 150mm at f/2 = 75mm at f/1 = 50mm at f/0.75 There aren't many lenses that wide for 35mm or medium format, are there ? There are lots of 300mm lenses for Large Format which open to f/4.5. You can purchase one for less than the cost of an entry level DSLR.

Very few people shoot sporting events with an 8x10 view camera. Why should we expect a 35mm landscape photograph to compete with an 8x10 ?


http://www.kenleegallery.com/images/tech/aatruck2.jpg
"When asked what camera I use, I reply 'The heaviest one I can carry'." - Ansel Adams

[marshallarts]

Some lenses have more exaggerated dropoff than others. This will be true in the 35mm realm too. Lenses with fewer elements will generally give very pleasing results in this regard. For example, the 3-element Apotar in my 6x9 Agfa folder, gives lovely blur rendition. Tessars are 4 elements. I love their blur rendition. Heliars have 5 elements, and are designed to give exaggerated blur at wide apertures. These lenses are available for 35mm cameras.

Great info! I wasn't aware of that.

It's hard to not think my question was a novice one. But I have a very solid understanding of DoF and formats. (It's hard to avoid going to these obvious answers). The fastest lenses I have are my 50mm f/1.4 on a 5DmkII and a 25mm f/0.9 on my 4/3rds camera (with a 2x crop factor). They have the same equivalent focal length but the 50mm f/1.4 offers shallower DoF on my full frame 35mm DSLR.

Doing the best estimate I can with a DoF calculator and equivalent FoV, wide open the closest match for 4x5 would be a 180mm f/5.6. I would think it would be possible, but it still doesn't look the same.

Ken, your explanation about lens elements may be the missing link. I would love to learn more about this. If you have any additional info or suggestions for web pages, I'd love to know.

Also.... I intentionally started this thread asking "why" and not "how" to avoid the obvious retort, "shoot LF if you want a LF look". I'm interested in the theory of understanding why.

[Oren Grad]

DoF formulas and tables depend on certain simplifying assumptions about optics. One complication that is not accounted for is that the subjective appearance of focus depth depends on the way the lens renders the transition from in-focus to out-of-focus. To oversimplify a bit, the smoother the transition, the greater the apparent DoF, even for lenses with the same FL set to the same aperture.

Also, beware of comparing subtle aspects of optical character between film or analog prints on the one hand and digital files (original capture or scan) on the other. Post-processing of digital files - especially sharpening algorithms - can mess up OOF transitions.

[marshallarts]

... One complication that is not accounted for is that the subjective appearance of focus depth depends on the way the lens renders the transition from in-focus to out-of-focus.....

This appears to be the information I am missing out on and would greatly like to learn more about. Can someone please direct me to a website or someplace where I may learn more about this!?

Thanks!

[Ken Lee]

There's theory, and then there's empirical results. Google the term "bokeh" and be prepared to do a lot of clicking. Everyone has their own favorite lenses in this respect.

...and then there are "portrait" lenses, purposely non-corrected, to give blur, halo, coma, etc. Many of these lenses go back to the 19th Century, and are mounted "in barrel", IE no shutter. They go back to the days when people controlled exposure with a hat.

Cooke, an English optical company, still offers a portrait lens for Large Format. See http://www.cookeoptics.com/cooke.nsf/products/largeformat.html.

Then see Jim Galli's pages (http://tonopahpictures.0catch.com/). He has compared and contrasted a large number of vintage lenses.

[Ivan J. Eberle]

In addition to bokeh, with LF view cameras and tilt/swing moves, there's also the possibility of using these longer focal length lenses and keeping everything near-far in focus. This gives another kind of telephoto presence to a shot that's hard or impossible to pull off with non-view cameras even at the same focal lengths (And then, these shots being oversampled/downsampled, also look better at modest sizes viewed on the web).

[Oren Grad]

This appears to be the information I am missing out on and would greatly like to learn more about. Can someone please direct me to a website or someplace where I may learn more about this!?

Thanks!

You can find a brief description of the problem in section 22.3.1 of the 1988 edition of Sidney Ray's text, "Applied Photographic Optics". I'm sure it's in the later editions too, but since I don't have them I can't give you exact citations.

Beyond that, I don't know of any systematic, informative discussion of this. There's lots of confused and confusing "I like this bokeh, I don't like that bokeh", but AFAIK not really anything of practical use specifically about the perceptual impact on DoF of focus transitions.

If you're digging into matters of bokeh for the first time, Harold Merklinger's article is worth a read:

http://www.luminous-landscape.com/essays/bokeh.shtml

Because it's not well-documented and because there's a substantial subjective element, the best teacher on this topic is just experience of your own, using different lenses.

[Leigh]

LF will look "sharper" at any print size because it is.

Film has an inherent resolution limit. As you enlarge the negative to produce a print, that resolution is reduced by the enlargement factor.

For example, if you had a super-duper film that could resolve 100 lp/mm, and wanted an 8x10 print, you could achieve that resolution in the final image by contact-printing an 8x10 negative.

If you enlarged a 4x5 negative two-fold, your best resolution would only be 50 lp/mm.

If you enlarged a 2x2.5 negative (if such existed), your best resolution would be only 25lp/mm.

If you start with an MF or 35mm negative, your achievable resolution is that much less.

- Leigh

[Richard M. Coda]

What everyone else said. Plus what Ken Lee said... I have calculated an 11x14 contact print at 3.85GB. You can WALK into an 11x14 contact print. DSLRs are no better than 35mm.

http://rcodaphotography.blogspot.com/search?q=eleven%3Afourteen

[paulr]

LF will look "sharper" at any print size because it is.
....

If you enlarged a 2x2.5 negative (if such existed), your best resolution would be only 25lp/mm.

If you start with an MF or 35mm negative, your achievable resolution is that much less.


The OP is talking about images viewed on the web; a pretty high resolution monitor displays around 2 lp/mm. At typical display sizes, a dslr or a 35 mm camera can equal the MTF of a large format camera at these resolutions. Especially after the image has been properly sharpened.

I suspect any differences in the look are about something else. Sometimes there may be no difference, but other times there's a vague something or other that tells you it's a big negative.

The first time I picked up a book of Nick Nixon's photographs, I said to myself, "these look like 8x10 contact prints." I don't know what gave that impression. There was nothing special about the book reproductions. I don't think Nixon did much with selective focus that couldn't be duplicated with medium format. And the pictures were all handheld, so it probably had nothing to do with sharpness. But something made that instant impression.

[Oren Grad]

And the pictures were all handheld

That's an astonishing claim, given that the vast bulk of Nixon's published work has been shot in 8x10 and larger, using an assortment of folding flatbed view cameras. His recent MFA show was primarily 8x10 contact prints, though there were a few 11x14's and enlargements too. So can you cite a source?

[Oren Grad]

Just to tie my comments more properly back to the original question as actually posed: most of the subtlety of a contact print from an LF negative is lost on the web. A little bit of the special flavor survives, to be sure - mostly the part having to do with gross DoF effects of using long focal lengths - but I think that what really happens when we view LF reproductions on the web is that we project on to the reproduction what we know of how LF prints look and feel "in the flesh". In other words, it's mostly in our heads, not on the screen.

[paulr]

So can you cite a source?

No, just something I'd heard more than once; that Nixon used handheld 8x10 for some bodies of work. I couldn't verify it just now with a quick web search, so maybe I'm wrong.

For the purposes of this thread, I don't think it makes a difference. The reproductions in the book I mentioned anything that could distinguish between 8x10 and 4x5, in terms of detail.

[rdenney]

I tried to sum it up in the title. An ignorant question at first glance, but hear me out. I've only dabbled in LF with my Graflex, but I absolutely love the images. The crispness and falloff towards out of focus is what really strikes me. I've been trying to find equivalent techniques with 35mm to achieve the same look but it just can't bring that LF look to the table.

Resolution is only part of the difference. Remember that for any given image at a given print size, the larger format will require a longer lens which will reduce depth of field.

Let's compare using a tight facial portrait printed to 8x10. Let's say you use an 8x10 camera to photograph a face. If you use a normal lens of, say, 12 inches, and focus at two feet distance, you'll create a face about five inches tall, or half the long dimension of the frame. For the resulting contact print, the depth of field will be 0.03 feet at f/11.

To get the same composition from 35mm format, you'd need a 50mm lens focused at the same two feet to make the same tight facial portrait. The face would still be about half the long dimension of the frame. The depth of field will be 0.04 feet at f/1.4. To get the same depth of field as you got using f/11 on 8x10, you'd need f/1.2, and even that still provides a hair more depth of field than the 8x10 format example.

If your 12" lens for 8x10 opens up to f/5.6, then you can see that nothing in the smaller format will be able to match the same selective-focus look.

Since you specifically mentioned how the crispness falls off towards the out of focus, the depth of field is a key contributor to what you are seeing.

Other contributors include how large-format lenses are designed. If you are using a tessar design (such as with nearly all 127mm and 135mm press lenses used on Graflexes), it will provide a different rendering of the out-of-focus areas than will the double-gauss normal lenses on nearly all 35mm film and digital cameras. Tessar designs also tend to be sharper in the center than at the edges, especially when used at wider apertures. Tessars in small format do exist, but they tended to be slow, and 35mm photographers wanted faster lenses for a variety of reasons, including hand-holdability and more extreme selective focus. So, they migrated sooner to double-gauss and plasmat designs with the effect that most normals were already there when press cameras were still using tessar-type lenses. My small cameras with double-gauss lenses create a much different--and less smooth--out-of-focus background than do my larger-format lenses. I've often heard this tessar effect described as an "old-fashioned" look.

Then there is the issue of larger formats having more tonal information per square inch when print (or display) sizes are the same.

Rick "noting that web display is generally consistent with the assumption of using the same print size for compared formats" Denney

[jp]

Thin DOF is apparent on the web.

The web doesn't do justice the majority of the time for showing LF images. It's what we got though, for the most part, and it's convenient for sharing imaging basics. The deep shadows, the texture of a tree trunk 30 feet away in a landscape, blades of grass in a field or barnacles on a shore scene are the sort of details dealt injustice by the web.

Aside from details, consider the finished product: the look of a platinum print or double coated cyanotype on arches platine, a dye transfer print, a cibachrome, a karsh portrait on traditional commercial B&W materials, b&w silver prints that are at the same time and buttery smooth tones and full of detail. All things the web fails at.

I'm not damning the web; I'm in the Internet business. But I don't pretend it's well suited to what makes LF a nice part of photography.

[Leigh]

People keep talking about thin DoF, yet...
one of the most common LF styles is using Scheimpflug to achieve virtually "infinite" DoF.

There's a disconnect here somewhere.

- Leigh

[rdenney]

People keep talking about thin DoF, yet...
one of the most common LF styles is using Scheimpflug to achieve virtually "infinite" DoF.

There's a disconnect here somewhere.

Recall what the OP wrote:


I've only dabbled in LF with my Graflex, but I absolutely love the images. The crispness and falloff towards out of focus is what really strikes me. I've been trying to find equivalent techniques with 35mm to achieve the same look but it just can't bring that LF look to the table. I can see how a contact print from a 4x5 would be something of unparalleled resolution in comparison to an enlarged 35mm print, but what confounds me is how come a low res scan of such a print yield that same effect vs the 35mm scan.

His objective is not to achieve overall sharpness on large prints, but to understand why LF has a characteristic look even on low-res scans, particularly in the way they drop out of focus.

Does that close the gap for you?

Rick "who answered his question, not necessarily yours" Denney