#1070. Large format, Cinematic … Getting the Looks (1/3)

By pascaljappy | How-To

Dec 14

An abundance of articles mention those idealized styles and the gear aimed at creating them “on the cheap”. My experience in that respect has been … underwhelming πŸ˜‰ But getting to the roots of those looks can help us recreate them with common gear!

Slightly medium format look

A lot has been written about the cinematic look and the large format look.

And, a few years ago, when Sony took the world by storm with their fantastic new mirrorless cameras that could use lenses from almost any brand so long as they covered the sensor, a range of accessories flared up and promised to transform those cameras into large format monsters and cine cams. They probably existed before and could probably be used with other cameras, they just seemed to become more famous as Sony stirred the soup with their extra marital lens matings.

There was the Mirex, which I owned and reviewed and sold immediately after πŸ˜‰ Superb build quality, superb functionality, lazy user. And there were others, the name of which now elude me. All allowed the diligent user to scan the rear of a larger format lens and stitch an image for more pixels and a large format look.

Slight diffusion

Nowadays, all the craze is in high resolution video. God help the manufacturer who releases a 500 bucks camera that doesn’t do 16K in 240 fps from an 80Mp sensor. Might be a good time to invest in Sandisk, Western Digital and Samsung stock.

Because slow-mo, you know. That cinematic look.

Except it ain’t that simple, really. Because there are multiple cinematic looks, not one, to consider. And because profound technical evolutions in cinema cameras is mixing up the looks even more.

Two trees

So I’ll try to explain my personal take on those looks and how to reproduce them, based on a semi-technical approach to the subject. This will take several posts, so this first one will focus on large format photography. Cine looks will come in later instalments.

There are no really large format digital sensors around these days. That would be a lovely idea, but so far, nada. The current dogma is one of slicing and dicing in as many pieces as possible rather than offering a generous uninterrupted surface. A few years ago, you could buy a scanning back that covered a huge surface compared to our modern offerings, but that required a perfectly steady mount and a steady subject, a bit like pixel-shift today. Let’s ignore those early curiosities for this discussion.

Since technology and economics make smaller sensors more viable, the idea behind the Mirex and others seems very valid : grab a large format lens, scan the image circle piece by piece with a smaller sensor, and presto. And it kinda worked, though it was a little cumbersome.

Square stitch 1

An easier alternative is to stitch multiple frames using your everyday camera and home for similar results. Sadly, it’s not that simple.

The photos above and below are both panoramas of two X1D frames. That’s often my goto method to create a square frame without cropping into that sensor. If the subject allows it, it works just fine. But it doesn’t look like anything else than a Hassy X1D photograph. Just like stitches of 4 or more frames with an M43 camera will look like a larger photograph made by an M43 camera, not an Ansel Adam’s 8×10 contact print.

The Mirex had one thing correct in that it required the use of a medium format lens. In my case, Mamiya 645 lenses (which I am now kicking myself for selling πŸ˜‰ ) What my Sony A7r2 was sampling, with the help of the Mirex, was the look of those lenses. A very different look from the Sony Zeiss I was using at the time. And one step in the right direction.

Square stitch 2

But only one step, and not a big enough one for the large format look. So what’s really required?

  1. An actual large format lens
  2. A film look from a digital sensor and digital PP
  3. The proper aperture, focal length and magnification

This may seem obvious, but the Mirex solution only offers us a sliver of steps 1 and 3 and nothing of step 2. Let’s examine those in turn.

Gentle rolloff

There are stunningly sharp large format lenses for sale today. Rodenstock, Nikon, Schneider (and probably others) all sell modern offerings optimized for digital and with MTF curves that put our favourite glass to shame. But, when were talking about the large format look, those are not what we are referring to. The term, while not properly defined, stirs up images of Sally Mann or Brett Weston. Old large format lenses, that it.

And those were not sharp. Not by any stretch of the imagination. In fact, few old large format negatives stand up to much magnification in print. Being so large they don’t need to, even for a decent sized print. In fact, a large part of the large format look hinges on that combination of a not very sharp image that’s not enlarged very much.

Think of the best images produced by the iPhone 12 Pro Max. Well, the exact opposite to that πŸ˜‰ The iPhone produces tack sharp images on minute pixels that must be enlarged a lot too produce a 20″ print. Around 70 times. Compared to 2 times from a 8×10 negative. Leaving resolution aside, the final aesthetics from a greatly enlarged tack-sharp jpeg and a only slightly enlarged slightly sharp negative are completely different.

Nikon D80

Also, large format implies film. Hence gentle highlight rolloff. And that’s another aspect of modern cameras that’s radically different from large format films. Film had a very long shoulder to its tone curves meaning that the transition between very light and completely blown-out zones was very subtle, almost imperceptible. Whereas, in their haste to pretend they are more sensitive than quantum physics allow, digital cameras tend to cut off very brutally. Which is synonymous with hideously.

More than any other factor, my love of the Hasselblad X1D files comes from the fact that its highlight rolloff is far nicer than the one of the Sony A7R2 it replaces. See the cloud photographs, two photographs above. And I believe the CCD image above (from an old Nikon D80) is also far nicer in the highlights than what most high-end CMOS cameras deliver today, in spite of its (theoretically) much lower dynamic range. Even with the glorious Sony A7R4, it is wise to underexpose (sometimes massively) to keep colours in check in light areas of the frame. And, as nice as it is, the X1D still falls a long way short of well processed B&W film from a large format camera.

So, how can you recreate that gentle rolloff? Through HDR and post-processing. Not the mad tone mapping HDR we have all come to know and hate, but one that gently blends well exposed shadows with underexposed highlights. This is where the Zone System still comes in handy. You can even bracket, overlay the two frames in PS layers and simply blend one into the other using good ol’ dodging and burning. Of course a bit of curve massaging for gentle slopes at either ends probably helps as well.

Good highlights and a Black Sun!

Then, there’s aperture. The f/64 group didn’t get that name from using f/1.2 lenses πŸ˜‰ In order to obtain sharp images from relatively close up to infinity, large format photographers needed to close down the aperture to such extremes: f/16 to f/64 was the norm.

And there’s focal length. If you accept the “natural” looking lens has a focal length roughly equal to the diagonal of the film/sensor, that’s 40-ish mm for full frame, 10-ish mm for the iPhone and 300-ish mm for 8×10. So you could grab an old medium format 300mm close it down as far as it goes (f/32) and start stitching for vaguely similar results to 8×10.

If it’s sounding like a lot of work, it’s because it is (why do you think you’re not seeing 20 large-format-ish pics on this page? πŸ˜‰ ) But Leonard has recently show us you can hand hold a 90 frame panorama with astounding results. So, it’s doable.

Sony A7r2 single frame

Don’t think you can skip the long lens part of the deal πŸ˜‰ Even with very soft light and gentle highlights, the brutal contrast between foreground and background, above, shouts “small” (full frame) sensor.

The local length, aperture and optical design give you the look and perspective of large format. And it’s through stitching of those tiny “large format” post stamps that you broaden the angle of view. Sorry πŸ˜‰

Now, my longest “old” lens is a 90mm Elmarit-M lens that looks absolutely gorgeous in the right conditions.

The sublime Mandler look (compare to photos above and below)
Wishing well, Sony A7r2 and Loxia 25mm

Finally, there’s the actual chemical process vs the inkjet or screen, to consider. Should I even bother describing the difference between a retina screen and rice paper with a gelatin coating? πŸ˜‰

Here, I prefer to diverge from the traditionalists and their chemical-stained borders. Simulating those in a digital process is a step too far for me. It is faking it rather than recreating it. In fact, I very much admire the work of Christopher Thomas, who photographs on large polaroid negatives and then scans for a digital output. But, even though the stains are genuine, seeing them on a digital output is off-putting enough to keep me away from his otherwise glorious prints. I find a good piezo print on art paper really lovely and vintage enough for me. And a carbon print even is better still, if your budget stretches ten fold. Still, that’s just me and everyone’s subjective mileage is legitimate at this point πŸ™‚

So, this, above, is a partial attempt at putting my money where my mouth is. It is a 9-frame stitch made with a 50mm C-Sonnar at f/4, for a roughly 7x7cm sensor size (still a long way from true large format). To me, it still looks a tad “sharp” and modern, but the tonal range is spot on. For the next post, I’ll try a longer focal length and smaller aperture. More important : What do you think? Does this look more like a Linhof or a Smarpthone?

And, while I’m about it, I tried a film medium format look on the same walk. It is below. Convincing?

Those are starting points. After that, everyone explores a different direction. Care to share your ideas or results?


​Never miss a post

​Like what you are reading? Subscribe below and receive all posts in your inbox as they are published. Join the conversation with thousands of other creative photographers.

  • philberphoto says:

    Another brilliant post! The “Mandler look” image has a background cloud that is so good that it, alone is almost worth the effort… of reading the whole item :-). You reveal and shake up lots of variables, and it stirs my imagination. Which is all to the good, as is anything that forces me away from status quo. Kudos!

  • Andreas Aae says:

    Except for the lack of grain, your pictures are damn close. Can you in fact tweak the highlight rolloff on the hassy?
    When looking for the large format look, the easy remark would be: just shoot film! ( I am contemplating a contribution on this when my moving house is sorted out…).
    But, take a look at what these guys are doing, and start saving: http://largesense.com/
    Secondly: When stitching together a 4×5 with a small sensor, you get a 4×5 cover. However, when simply stitching 2 or more ‘regular’ images, if done correctly by rotating around the lens entrance pupil, you actually get an image equivalent to using a curved sensor/film, something of potentially very high quality…

    • pascaljappy says:

      Oh, my word !!! Thank you for that link, Andreas. What gorgeous videos! I had no idea those guys existed. I’ll link to them in the next post
      Yes, the rotation is like approximating a curved sensor, with all the advantages for that, you are right to point it out.

      If you ever want to write something about this, or film, you are most welcome! That would be a fascinating topic to cover.

    • PaulB says:


      I will second their thank you for the link.

      Though it could be trouble for my retirement fund.


  • jean pierre (pete) guaron says:

    My favourite is gentle rolloff. WOOF!

    I can’t get this stuff about cine. Whenever I’ve done cine, I did it with a cine camera. Cine is different from stills.

    While it might be “possible” to do both with the same camera, is it appropriate?

    Comparison – would you prepared to attach a towbar to your Bugatti Veyron, and use it to haul a plough around a paddock? – or drop a V-6 2000cc with down-draft Webers, into a 2CV, to drift down to the local store for the latest newspapers?

    Cine requires totally different lenses, and a totally different approach.

    “Horses for courses”. Hybrids will ALWAYS be a compromise. Which, come to think of it, probably fuels my disinterest in cell-phonograms.

    • pascaljappy says:

      I think vloggers are finding a lot of value in hybrids. Stills + video + fancy moves (slow mo …) in one affordable package. Not my goal, but I get it πŸ™‚ And, to be honest, given the price of cine cameras, easily 10 times the price of photo cameras, the purist approach is out of reach for most πŸ˜‰

  • Frank Field says:

    Pascal —

    An excellent post!

    Our materials have come a long way since the days of Edward Weston using a three-way convertible lens on his 8×10 camera for all of his non-portrait work! Films availabe in those days were anything but panchromatic. I suspect that it is basically impossible to fully emulate the materials of the past with today’s gear.

    I’d hope that we would celebrate the benefits of today’s materials and processes and use them to deliver results we could not produce in the past.

    One of the things I try to convey in images is the texture of the scene. Your square stitch 2 and next to last images just pop out of the computer monitor for me. Wonderful tonality, rich texture that is clearly visible even at screen resolutions.

    Today’s processes allow us to do things that prior generations could not even imagine. Our ability to stitch to produce a truly high res image as you have done is one such process.

    I never did any monochrome photography in the days of film yet have come to enjoy it with today’s gear. I much appreciate the ability to apply a literally infinite range of color filters a posteriori to achieve desirable tonal separation in monochrome. I can not imagine myself having to guess at the right color filter and being limited to a finite choice of filters to use a priori! (I do admit to a certain limitation in my color vision that is not uncommon among males.)

    Thanks for a thought provoking post!


    • pascaljappy says:

      Thank you very much, Frank.

      Yes, panchromatic films are a “problem”. Maybe with a sensor with no IR/UV filter? But those tend to be very weird looking and nip & tuck on the sensor is stretching research further than I’m willing to go πŸ˜‰ πŸ˜‰

      Your photographs certainly convey that sense of texture and I’m happy you see than in the stitched photographs here as well.

      We’re very lucky to have access to so many tools and techniques these days. Everytime a new one comes along, it can get very odd looking and artificial (as early uses of HDR were) but, when experts get hold of them and introduce them into their workflows, the creative envelope stretches out and the process gets integrated in a more “traditional” look. It’s fascinating to observe, isn’t it?

  • Michael says:

    Wow, such a thought provoking topic that just happens to coincide with my current road if investigation. I have found myself trying to figure out the major areas of concentration and you put it all in order for me. Highlight roll off has been my latest bit of investigation and shooting cloud formations has allowed me to work heavily on curve adjustments to the highlight side, shadows are next. Masking two, or three, exposures has shown some good results so far.

    Stitching is something I am just getting started with so you’ve ramped up my motivation to do more than play with it.

    From your first paragraph on I was thinking about Christopher Thomas and was surprised/kind of not surprised by your reference. I find his restaurant/retail works done at night particularly interesting.

    I look forward to your next few articles. Thanks again for continuing to be so topically motivating for me.


    • pascaljappy says:

      That’s wonderful Michael πŸ˜‰ Great timing and great encouragement for me to continue scratching. Christopher Thomas is a brilliant photographer and a very interesting person. I was lucky to see his recent Bittersweet prints and was blown away by some of them. So sensitive and evocative. And his earlier work is even more powerful, sometimes even difficult to watch. Incredible …

      I hope you achieve the results you are looking for, please send a few our way if you feel like it.


  • PaulB says:


    Another good article with images worth of envy.

    Though, I am not sure they really convey the β€œLarge Format” look. I am not sure it is possible to capture the look by building up from smaller components, as β€œbig” has its own virtues.

    I’m not sure I can say why I think the look is missing, since memory is fleeting regarding what the look is. But, if I was to say what factor is missing from these images, it might be focal length, or to be more nebulous, coverage. You are making big images using lenses made for a smaller image plain, rather than a lens made for a larger image plain and the effect may be visible, or at least influencing our subconscious.


    • pascaljappy says:

      Hi Paul, thanks for your comment. Only the last but one photograph is meant to convey the large format look. The others are just illustrative πŸ˜‰ And even that one isn’t really quite there, although it has the pop and tonal values. I’ll be working on one or two more for the next instalment.


  • Paul Perton says:

    I don’t do much commenting, but really wanted to compliment you on some exquisite images in this post. Wonderful, colourful and evocative. Call me envious.

  • NMc says:

    Nice set of photos with an interesting set of words.
    I am a bit conflicted. I do like the many old looks from analogue photography however trying to recreate these in digital risks fakery instead of communicating atmosphere/or whatever.

    Certainly sensors with gentle highlight roll off/compression would be the one technical advance I would like to see. Quantum dot photo diodes were supposed to do this but the development of this technology has been cancelled I think.

    Anyway a link to a tube that discusses the idea of cinematic, https://www.youtube.com/watch?v=fy6b8ZaKAWI . Enjoy, I know I did. Some β€˜cinematic’ films can be lacking in content or narrative, I think that is ok because it is primarily a visual medium (as well as a sound medium).
    Regards Noel

    • pascaljappy says:

      Hi Noel,

      to me, what matters isn’t really replicating the exact look. As you say, that feels fake. Particularly trying to replicate film in a digital era. But the imaging qualities of a large format “sensor” are every bit as appealing today as they were a century ago. The great tonal richness, the gentle rolloff and the superlative 3D. If I can achieve that look via stitching, that’s good enoug for me.

      Thanks for the link πŸ™‚

  • Sean says:

    Hi Pascal,
    What a wonderful, thought provoking and intelligent post, and accompanying images. I’m very partial to the grey scale images shown here – they sing!

  • >