Examples of Large Format shots that cannot be done with small

[Tin Can]

Not a drill but a Hot Glue gun. That's a glue drip on nozzle. It is an interesting 'photo' which only proves photography can duplicate computer simulation like the fashion model thread. If that image can be made by 'hand' it can be automated and is everyday becoming better and better by 3D scanning CAP (computer assist photography). I am making up these obvious acronyms.

There's an interesting example on the getdpi forum of great apparent depth of field at close range, using a digital tllt/shift lens and focus-stacking (and who knows what else).

See http://www.getdpi.com/forum/canon/57698-two-canon-tilt-shift-images.html

According to the author of the post, the photo of the drill required "59 slices of focus and about thirty hours of post" [processing]. Heavens !

[goamules]

There's an interesting example on the getdpi forum of great apparent depth of field at close range, using a digital tllt/shift lens and focus-stacking (and who knows what else).

See http://www.getdpi.com/forum/canon/57698-two-canon-tilt-shift-images.html

According to the author of the post, the photo of the drill required "59 slices of focus and about thirty hours of post" [processing]. Heavens !

Yep, that's a lot easier than LF! A lot of people are offering the hard, expensive, or silly ways to replicate a simple LF shot.

[neil poulsen]

Historically, Edward Weston found large format cameras were much more effective for talking attractive young women out of their clothes. Quite a bit of contemporary work, some even on this forum, validates this...

I wonder if he ever photographed the clothes, laying there on the floor?

[David_Senesac]

A few posts have mentioned issues using tilt-shift lenses (apparently meaning on DSLRs) versus the way we view camera users do so more easily by front standard movements. Obviously a few here are not aware of the following.

Of course with an expensive medium format digital back on a view camera tis the same. But in this era there has been a paradigm change that makes using those expensive tilt-shift DSLR lenses of less value.

One can digitally focus stack blend images in Photoshop. Although it is digital macro nature photography users that have been making the most noise with those techniques the last few years, the same strategy can be used for landscapes. Combining that with multi column-row stitch blending given a good manual verniered panoramic head and software like Kolor Autopano and one can capture high resolution sharp edge to edge processed images as long as a subject is somewhat static. Most interest with stitch blending that has been around for several years now involves single shot frames without focus stacking at the extreme end with a robotic head. However the below technique combines the two differently so this is what I am referring to as the paradigm shift.

By using one of the center optically better apertures like F8 to F11 to shoot all the focus stack shots the resulting quality can be sharp edge to edge on a frame regardless of how 3-dimensional a subject is as long as there are no near elements that cause paralax concerns. Lots of such images on the below link. Several of my landscapes last week in Death Valley flower fields using a 24mp mirrorless camera were 4 column by 2 row, 8 image blends with each frame using from 5 to 10 focus stack images. Thus some processed images the result of combining 40 to 50 individual images. One image I processed last night ended up 18,700 by 9000 pixels or enough at 300ppi for a 62.3x30 inch print that has more detail than any of my older 4x5 work.

http://www.davidsenesac.com/2015_Tri...onicles-0.html

David

[Harold_4074]

I may be misunderstanding, but all of my large format lenses have front elements smaller in diameter than the distance between my pupils, which I measure at about 70mm. My old 360mm f/6.5 would have passed this test though.

My 85mm f/1.4, and 300mm f/2.8, for DSLR, however, both exceed 70mm, at 77mm and 105mm, respectively.

Okay, compose an 8x10 head shot with your 360 f/6.5, stop your 300mm f/2.8 down to f/6.5, and make the same picture full-frame in the DSLR. I expect that the angle subtended by the DSLR lens will be considerably smaller than that subtended by the LF lens, so the perspective and illumination quality will not be the same. I think that this is close to what the text I remember was describing, since at the time the book was written the corresponding comparison might have been between a 6x6 TLR or a 35mm SLR and something like the 405mm Kodak Portrait. If the authors were correct, the format does matter because the other imaging parameters are not completely independent of the film (or, today, sensor) size.

I admit to having phrased the original argument (dating from probably the 1960s) to rule out things like multiple digital images being merged or assembled to give the effect of a large front element without perspective problems; this is partly because I spent some time trying to figure out how to do exactly that, and concluded that only a mannikin head (are you reading, this, Randy Moe?) would have the patience and ability to hold still long enough. I gave this up when I realized that translating the small-format lens radially takes the optical axis with it, so that combining the array of images would give a third type of image, not a simulation of either the large, fast SLR or the large, slow LF lenses.

[Jody_S]

Okay, compose an 8x10 head shot with your 360 f/6.5, stop your 300mm f/2.8 down to f/6.5, and make the same picture full-frame in the DSLR. I expect that the angle subtended by the DSLR lens will be considerably smaller than that subtended by the LF lens, so the perspective and illumination quality will not be the same.

I think you'll find that if you back up your 8x10 and shoot side-by-side with your dSLR, so the images on film/sensor are the same size, the perspective and illumination will be identical (crop your 8x10 and check).

[Harold_4074]

I think you'll find that if you back up your 8x10 and shoot side-by-side with your dSLR, so the images on film/sensor are the same size, the perspective and illumination will be identical (crop your 8x10 and check).

Correct. But the thread title is "large format shots that cannot be done with small format", not the other way around. I asked for a head shot filling the frame of the 8x10, not an image 1/80 that size taken from a position where the subtended angle of the lens equals that of a DSLR with a 300 mm lens.

And in any case, the information content of the square inch of film will not equal the information content of an 8x10 negative---no matter how you chose to estimate the performance of the DSLR.

If I've misinterpreted the intent of the thread, I presume that goamules will chime in and correct me

[Jim Jones]

Another example where large format is better is in pinhole photography, where theory proves that LF has superior image quality. In pinhole photography the light sensitive material need not be flat. It can be curved and even spherical. For example, http://www.largeformatphotography.in...7&d=1351947177, (post 15 on Post yer pinholes. This would have been difficult to achieve with roll film.

[Tin Can]

Okay, compose an 8x10 head shot with your 360 f/6.5, stop your 300mm f/2.8 down to f/6.5, and make the same picture full-frame in the DSLR. I expect that the angle subtended by the DSLR lens will be considerably smaller than that subtended by the LF lens, so the perspective and illumination quality will not be the same. I think that this is close to what the text I remember was describing, since at the time the book was written the corresponding comparison might have been between a 6x6 TLR or a 35mm SLR and something like the 405mm Kodak Portrait. If the authors were correct, the format does matter because the other imaging parameters are not completely independent of the film (or, today, sensor) size.

I admit to having phrased the original argument (dating from probably the 1960s) to rule out things like multiple digital images being merged or assembled to give the effect of a large front element without perspective problems; this is partly because I spent some time trying to figure out how to do exactly that, and concluded that only a mannikin head (are you reading, this, Randy Moe?) would have the patience and ability to hold still long enough. I gave this up when I realized that translating the small-format lens radially takes the optical axis with it, so that combining the array of images would give a third type of image, not a simulation of either the large, fast SLR or the large, slow LF lenses.

Yes, I am reading this and I read almost every word written by all posters to this thread, most of time.

I study their websites, much to learn, not much time...

[Jody_S]

Correct. But the thread title is "large format shots that cannot be done with small format", not the other way around. I asked for a head shot filling the frame of the 8x10, not an image 1/80 that size taken from a position where the subtended angle of the lens equals that of a DSLR with a 300 mm lens.

And in any case, the information content of the square inch of film will not equal the information content of an 8x10 negative---no matter how you chose to estimate the performance of the DSLR.

If I've misinterpreted the intent of the thread, I presume that goamules will chime in and correct me

The classic 35mm portrait lenses, like Canon's 85mm f1.2 or the Russian 85mm f1.5, will give a very similar result as a 360mm on 8x10. That's what they were made for. Wide-open, they even glow. The only real technical advantage to the 8x10, as others have pointed out, is the ability to choose your plane of focus at that shallow depth of field.