A "medium format plug-in"

[frozenintime]

i'm fairly new to a lot of this, and come from the music/audio world.

it seems that there is a certain tonal quality to medium format (and other high-end/large negative equipment) that often makes it instantly recognizable as such.  

wouldn't it seem within the realm of computers and modeling these days to build a plug-in that attempts to adjust a digitally-captured image to reflect the particular way that these film cameras naturally capture light and color and contrast?

the idea comes from the digital audio world, where a great many plug-ins and programs attempt, sometimes with a shocking degree of success, to digitally duplicate the particular processes (and thus sound) of expensive analog gear.  the result is that you can record a voice digitally and add the "sound of tape" (as if you had recorded it to 2" tape) or digital versions of famous compressors or reverb chambers from abbey road.  respected engineers have confessed to me that they often can't tell the difference between the two.

is there a reason that this kind of thing would not work with photography?

[jjlphoto]

Well- you can't add quality to a 35mm DSLR file so it looks like a medium format file, but you can make fake Hasselblad borders (with that twin "v" notch on the frame edge) and do a cut and paste job with them. Or there are fake 4x5 polaroid transfer borders, fake 4x5 T-55 polaroid negative borders, etc. There are also scads of actions that mimic "film" looks like velvia actions, cross processing looks, kodachrome looks, etc. Just Google "film effects from photoshop" and you will get lots of sites.

[BernardLanguillier]

wouldn't it seem within the realm of computers and modeling these days to build a plug-in that attempts to adjust a digitally-captured image to reflect the particular way that these film cameras naturally capture light and color and contrast?
[a href=\"index.php?act=findpost&pid=89449\"][{POST_SNAPBACK}][/a]

I don't know if the list is comprehensive or not, but to me the difference between MF/LF and 35 mm results from:

- more "pixels" giving the impression of more continuous color transitions at a given enlargement,
- less grain giving access to finer details.

These are difficult to simulate if the information is not there.

Regards,
Bernard

[ddolde]

I don't there is a characteristic look to medium format really.  Better resolution and tonality than 35mm film of course but 4x5 and larger is even more so.

You're dreaming in my opinion.

[Jonathan Wienke]

While the resolution cannot be duplicated (detail must be captured, not calculated), things like tonality, color "look", and film grain structure can be simulated quite accurately if one is willing to take the time and trouble to do so. Any film's color and tonal response can be duplicated quite accurately if one takes the time to build a sufficiently accurate profile of the film's color response. Then applying a particular film look is as simple as converting an RGB image from ProPhoto to the film profile, then re-applying the ProPhoto profile. The grain structure of a given film can be duplicated by scanning an appropriately-exposed OOF shot of an evenly-lit white wall, then layering that over the image with the appropriate blending settings.

[PeterLange]

Given that some 35mm DSLR’s can compete with the resolution of classic MF film, it should be more a question of how to emulate the so-called ‘film-look’.  There are many respective tools and Actions available which attempt to realize this with digital capture (see e.g. below links).

Without having yet tried them all, it’s on my agenda to check and find something which allows to start ab-initio with ‘native’ Raw data (already demoisaiced, Exposure and WB adjusted, but without ANY further enhancement of tone, saturation and hue). Thus, having a clear, well-defined starting point, rather than an already half processed image.  If someone around here knows about a respective tool / approach, I’d be pleased to learn…

http://www.atncentral.com/download.htm
http://www.alienskin.com/exposure/index.html
http://www.fredmiranda.com/VelviaVision/
http://www.thelightsrightstudio.com/photoshop-tools.htm

Peter

--

[:Ollivr]

SLightly off-topic since its not about a plugin:

A common characteristic of film would be described as "creamy highlights" and good detail in the shadows. Unfortunately digital cameras often lack in these respects. On the luminance graph, the highlights fall off rather steep (not creamy at all) and due to the low data entropy in the shadows there is too little detail.

WHat I do is exposing to the right, of course, plus modeling the highlight and shadow slopes in my raw converter. This differs from what I used to do earlier - brutally spoken filling up the luminance space with as much peak data as possible (as the "auto" mode of the raw converter might).
Now, I maintain at least 1/3 for the highlight slope and 1/3 for the shadow slope. I do so by altering contrast and shadow/highlight/exposure relations. WHat I get is sort of a "bell curve" and an image which may seem a bit dark and lack contrast. The 16bit output goes to Photoshop, where auto curves take care of brightness/contrast. This gives me quite a different image than I would obtain with my other approach, and one looking a tad more filmish to me.

Would be interested in how that Fred Miranda plugin performs...

O.

[John Sheehy]

Without having yet tried them all, it’s on my agenda to check and find something which allows to start ab-initio with ‘native’ Raw data (already demoisaiced, Exposure and WB adjusted, but without ANY further enhancement of tone, saturation and hue). Thus, having a clear, well-defined starting point, rather than an already half processed image.  If someone around here knows about a respective tool / approach, I’d be pleased to learn…[a href=\"index.php?act=findpost&pid=89529\"][{POST_SNAPBACK}][/a]

Images Plus, MaximDL, IRIS (free), DCRAW (free).

[Hermie]

http://www.atncentral.com/download.htm
http://www.alienskin.com/exposure/index.html
http://www.fredmiranda.com/VelviaVision/
http://www.thelightsrightstudio.com/photoshop-tools.htm

Peter

--
[{POST_SNAPBACK}][/a]

... and  Imagenomic's (Noiseware) new plugin 'RealGrain' :
[a href=\"http://www.imagenomic.com/rg.aspx]http://www.imagenomic.com/rg.aspx[/url]

[James Godman]

One of the biggest factors that distinguishes medium format / large format is being able to achieve a shallow depth of focus.  Too many of the available digital tools make it difficult to get anything out of focus.  I would also argue that the bokeh produced by many medium and large format lenses is more pleasing than most small format lenses.

[pss]

apart from the optical differences (depth of field, perspective) there is simply more information with MF....just like the best filter in the world can make a 128kb mp3 sound like a uncompressed AIFF...there are things that can be done and of course the smaller the print, the easier to fudge things (kinda like playing tunes over a pocketradio...harder to tell the difference)
just like in music, no filter can add missing information....of course with music, you can add layers of "fake"? sounds...fake trees will look like fake trees in the background on a 16x20.....

[PeterLange]

Images Plus, MaximDL, IRIS (free), DCRAW (free).
[a href=\"index.php?act=findpost&pid=89548\"][{POST_SNAPBACK}][/a]

John,

I recognize that my post wasn’t clear enough - sorry.

What I meant is a tool to accomplish the second step, the main transform from a linear & somewhat accurate state to a preferred film look.  So that you could come from Iris, Dcraw, etc. (also would mention ACR under some provisions) in order to apply this xy tool which then would do all the tone, saturation and hue shifts required.

As far as I understand all existing tools try to create the film look based on a kind of re-rendering.  It is assumed that some processing was done before. If this starting image looks lifeless and flat, then a Velvia Action can of course do a great job. But if the starting image is already contrasty & well saturated, then same Velvia Actions might do more damage than good.  The entry state is undefined which in my opinion leads to confusion about which enhancements to apply before or via such software. Everything gets a case by case decision.

What I’m looking for is stand-alone Provia-tool which accepts somewhat ‘native’ data (already demoisaiced, Exposure and WB adjusted, profiled and converted to a standard encoding like ProPhoto RGB, BUT so far without any further enhancement of tone, saturation and hue). Then, it would create the desired Privia look on this basis.

Peter

--

[jjlphoto]

Phase One CO Pro has some Color Styles you can select during the RAW process such as '70s look, porcelain look, cross processing, Agfa and Ektachrome, etc. As far if there is any identifying characteristic that 2 1/4 film has that 35 or a DSLR does not have, there really isn't one that stands out. The only really noticeable thing that a medium format system does different (aside from increased file size) than a DSLR system is that focus fall-off occurs faster, but that effect can be mimic'd with a DSLR buy shooting wider open, or with post processing effects. There is a post effect that re-creates actual "out-of-focusness" complete with aperture shaped highlights called Depth of Field Generator Pro.

[pss]

alienskin has a plug in that creates film looks....tons of pre-sets, unlimited adjustments (grain,...)

[frozenintime]

i am coming from a pretty naive and unexperienced view in saying that high-end film cameras do seem to share some kind of unique capture of color and tonality that i don't see in many digital photographs. now, this could be a function of specific film, or even the general aesthetic sense of well-known art photographers whose monographs we all have seen.  

there is a smoothness and richness that i don't see in digital, in my limited experience.  if this is a function of pixels, then i guess the inquiry ends there for now. but jonathan wienke is speaking in a language i understand when he talks about 'modelling' a given film's color and tonal response.  this is how famous musical equipment is being "modelled" digitally, often with fantastic results.  couldn't the way 8x10 responds to a given scenario be understand mathematically to the point that it could be rendered into a formula of some kind?

anyway, just some food for thought.
interesting comments from all of you!

[James Godman]

Hello Frozenintime-

Another thing to consider, with people photography at least, is that working with large format is a very different process than working with a dslr.  This process effects the aesthetics of the image created, along with the technical factors.  This way of working will also effect the subject's expression, body language and so on, to the photographer's advantage or not.  And I don't know of a filter created that will emulate these factors

[Hermie]

Given that some 35mm DSLR’s can compete with the resolution of classic MF film, it should be more a question of how to emulate the so-called ‘film-look’.  There are many respective tools and Actions available which attempt to realize this with digital capture (see e.g. below links).

Without having yet tried them all, it’s on my agenda to check and find something which allows to start ab-initio with ‘native’ Raw data (already demoisaiced, Exposure and WB adjusted, but without ANY further enhancement of tone, saturation and hue). Thus, having a clear, well-defined starting point, rather than an already half processed image.  If someone around here knows about a respective tool / approach, I’d be pleased to learn…

http://www.atncentral.com/download.htm
http://www.alienskin.com/exposure/index.html
http://www.fredmiranda.com/VelviaVision/
http://www.thelightsrightstudio.com/photoshop-tools.htm

Peter

--
[{POST_SNAPBACK}][/a]

Peter,

When upgrading to DxO 4.1, I noticed the new DxO FilmPack:
[a href=\"http://www.dxo.com/intl/photo/dxo_optics_pro/overview/dxo_filmpack/dxo_labs_scientific_approach]http://www.dxo.com/intl/photo/dxo_optics_p...ntific_approach[/url]

Herman

[aperture12]

Hello frozenintime,

I don't consider myself to be knowledgeable enough about photography and/or audio to give a fundamental 'right' answer to your question, but I know something about both digital image processing and digital audio processing... so I'll give it a try.

In audio, the current state of technology makes it possible to capture the full range of audible frequencies with incredible detail (24-bit/96-kHz recording). High-end hardware imparts a (relative) very small 'fingerprint' on the recorded material: the result is extremely 'clean' source material. This material may not always be pleasing to our ears, so we process it with very advanced algorithms like concolution-reverbs, etc. This way it sounds much more like the great music halls or the abbey road studio: sound environments we have learned to appreciate. But strictly technically speaking, these algorithms deteriorate the pristine, clean charasteristics of the source material. Once the sound file is processed, it is almost impossible to revert it to a clean state.

Current potography is another story: today's commercial camera's and lenses are not able to capture the full spectrum of what the human eye can see, and most camera's use 12- or 14-bit recording which may result in 'banding' issues. (HDR-techniques try to overcome the range problem, but it differs enormously from the way our eyes deal with different light intensities). All camera-lens-sensor/film combinations impart a HUGE fingerprint on the source material: type of lens, sensor, etc. Trying to convert a 35mm image to the looks of a MF image, is trying to convert highly deteriorated source material to another state of deterioration. This could only be achieved by reverse engineering from the 35mm image to a 'clean' intermediate state. And as mentioned earlier: image detail has to be captured, it can not be calculated.

In the end there seems to be only one satisfying approach: don't try to mimic, try to make the most of it.

-- Martijn

[PeterLange]

Peter,

When upgrading to DxO 4.1, I noticed the new DxO FilmPack:
http://www.dxo.com/intl/photo/dxo_optics_p...ntific_approach

Herman
[a href=\"index.php?act=findpost&pid=90005\"][{POST_SNAPBACK}][/a]

Thank you, Hermie.

I’ve noticed this option with DxO and I must say that it looks very interesting (just referring to their website; don’t have DxO yet).

Wouldn’t it be great to have such a feature with Camera Raw. At least, this idea ‘bothers’ me since some time because it should be possible in principle.  However, that’s ongoing work – quite a tough one I have to say.  As it has been shown sufficiently, ACR can be well enough linearized simply be setting the main tonal controls to zero, thus, providing a kind of virgin rendition (ugly though) but suited to work with.  The rest is about finding all the nice transforms i.e. brightness, saturation and hue tweaks which make the preferred rendition of a film look…  That is why I'm looking left and right, trying to collect all related information.

Anyway, it might not be wrong to bring this subject also to the attention of the Adobe Camera Raw forum.

Peter

--