Film Negative Invert and Processing in L* Gamma FAQ

[digitaldog]

I'm not suggesting it. It's written plain as day in Wikipedia's specification and in the pdf file linked on the ICC's page on ROMM RGB.

What is the gamma of ProPhoto RGB?

What's this?

http://digitaldog.net/files/ProPhotoGamma.jpg

 

And a comment about the gamma from another actual color geek from the ColorSync list, YEARS ago and a useful comment on Lstar to boot!

 

On Jul 21, 2007, at 11:20 AM, Chris Murphy <lists@colorremedies.com> wrote:

Anyone know the rational why Kodak used the same working space gamma?

 

ROMM (ProPhoto) has an 8bpc and 16bpc implementation, and with the limited precision offered with 8bpc with such a huge space, going from XYZ to RGB back to XYZ could produce significant errors in blacks. The steeper the TRC (tone response curve), the bigger the reversal error. So a TRC defined with gamma 1.8 was chose, largely for reversibility. It's a useful characteristic because, of course, we convert captures into that space for editing and then we convert them out of the space for output.

 

This is why I think it was a mistake for ECI to move to an L* based TRC with ECI-RGB 2.0, as it still is advocated for use with 8bpc workflows. ECI-RGB 1.0's TRC is gamma 1.8 based, and thus I prefer it over Adobe RGB (1998), which has a similar TRC, defined as gamma 2.2.

 

Chris Murphy

Color Remedies

New York, NY

----------------------------------------------------------------------

Co-author "Real World Color Management, 2nd Ed"

Edited April 26, 2021 by digitaldog

[shijan]

And another little FAQ about using Iridient Developer to export camera scanned film negative to L* gamma for further invert and processing.

 

- Switch to LittleCMS and Highest Quality in preferences.

 

- Open RAW file in Iridient Developer.

- "Histogram, clip warnings, preview color space": Set ICC profile to the same ICC profile into which you expect to export final image. (In this guide i use custom-made ICC profile ProPhotoRGB with L* gamma)

- "Exp" -> "Edit": click "Reset All". For some RAW file types you also may need to uncheck "Use Embedded Curve" This will reset custom contrast curve and will allow to export original "flat" uncorrected image, as it "seen" by digital sensor.

 

- "Detail" tab: Reset adjustments or use lowest possible Anti Aliasing and Chroma Noise Reduction settings. When it goes to film negative scans, these settings need subjective tests for each specific camera. AA and CNR and Sharpness have strong affect to film negative appearance during processing.

 

- "Lens" tab: Keep lens correction disabled until you really feel that you need it.

 

- "Out" tab: Select ICC profile to export (ProPhotoRGB with L* gamma)

 

--------

 

Some advanced options:

- For one click batch processing you need to save settings as preset and use "Batch Window" instead of main UI.

 

- "Color" tab: "Input Profile". This allow you to set your own custom-made ICC profile instead of camera default profile as input. This is useful if you calibrate your system based on IT8.7 target and generate your own custom-made input ICC profile.

 

- "Out" tab: "Disable Color Management Processing" This will globally disable color management and also ignore output ICC profile settings selected in "Out" tab. This mode required when you export scanned IT8.7 target image that will be used to generate custom-made input ICC profile in other apps ( for example in Rough Profiler ArgyllCMS GUI, software perfiles color ICC IT8 colorchecker, fotografia cientifica - Jose Pereira )

[Ed_Ingold]

Is it so complicated to shoot 2-3 frames, process them in raw editor without additional noise reduction

Not hard, simply redundant and unnecessary.

[shijan]

Or simply too lazy

[digitaldog]

Not hard, simply redundant and unnecessary.

Like the idea of attempting creation of an ICC profile for a color neg; already correctly posted as not possible and like your comments here, to be ignored.

Edited April 26, 2021 by digitaldog

[shijan]

And here we go again....

I never suggest to create ICC profile specially for a color neg.

I suggest to create ICC profile based on IT8.7 slide film for a camera/scanner. It is a normal calibration of hardware. We calibrate camera + light source system based on IT8.7 (or any other possible) target and measured reference data provided for that target.

And yes, that target is printed on slide film, because there is no any other option. And yes, that slide film may have some tiny color casts. And it is normal, because that slide film target was measured by manufacturer and provided reference data for this particular target includes all these color casts and variations.

 

That colourphil.co.uk website is down, so here is a backup for that article Scanner Profiling Theory

"Colour Negatives

There are no IT8 targets available for these, for the simple reason that profiling negatives doesn't work! This is partly due to the vast array of negative film types with differing 'orange masks', together with inconsistent lighting and processing. One reason for scanning negatives is to archive old film. The fact that it is 'old' means that film stock is simply no longer available to photograph IT8 colour tagets."

 

If someone posted somewhere on internet that "No IT8 targets printed on negatives available - so profiling negatives doesn't work", it doesn't automatically should be a pure true.

 

I agree that No IT8 targets printed on negatives available.

But it doesn't mean that we can't use IT8.7 printed on slide film to camera/scanner, and then scan any film negatives on that scanner that will use that custom-made input profile.

In any case final correction for negative (or formally profiling) will be done by RGB AutoLevels (or by Manual RGB Levels adjustment).

 

It is also not too clear what really means there "profiling negatives doesn't work"

We can shoot on negative film the photo of ColorChecker or reflective printed IT8.7 target and use it as calibration for same film roll instead of doing AutoLevels for each frame. Negatives in film industry processed in this way, because for film it is near impossible to manually process each frame, same as we do with photos.

[digitaldog]

https://www.colourphil.co.uk/scanner-camera-profiling.shtml opens a page on this end and for the rest here who wish to read and then hopefully agree with the third actually knowledgeable color expert posting thus far in the thread: ...profiling negatives doesn't work!

And more:

Profiling a scanner with an IT-8 target: Color calibration in SilverFast Ai with film scanners and flat bed scanners for image quality improvement

The IT8 standard does not allow for calibration of negatives. Due to the various orange masks of negatives with the resulting higher contrast it makes no sense to calibrate a scanner for negative material. The IT8 calibration only works for positives.

 

"I honestly believe it is better to know nothing than to know what ain't so." -Josh Billings

Edited April 26, 2021 by digitaldog

[shijan]

So again. Just someone's opinion in advertisement article. Dudes quickly review that crappy SilverFast software for repayment. Discover AutoLevels-like tool specially developed for Negatives in that software. Made conclusion that there is no need to calibrate scanner because AutoLevels can fix it all.

By the way, i read that article long time ago (it is not too many articles to read about IT8.7 film targets), but it doesn't stop me to do my own test and try to calibrate my scanner. And i was really happy with results i got from film negatives with scanner calibrated in that way.

 

Seems your opinion never based on personal tests or works. You even speak instantly with other people's worlds.

Did you personally scan film on both camera and scanner?

Did you you personally do some tests to confirm or deny any ideas shared in this thread?

Did you personally try to calibrate scanner IT8/7 slide film target and compare?

Do you personally somehow related to film scanning and processing at all?

No you don't.

No one have right to say "it is wrong" based on random article that have no real tests or examples.

So why i talking with at all?

[samstevens]

So why i talking with at all?

Only you can answer that question!

[digitaldog]

Did you personally scan film on both camera and scanner?

Did you you personally do some tests to confirm or deny any ideas shared in this thread?

Did you personally try to calibrate scanner IT8/7 slide film target and compare?

Do you personally somehow related to film scanning and processing at all?

Yes to all, starting 3 decades ago.

You really should considering examining MY bkgnd, how many scanner reviews I've produced for PEI and PPA mag (and the technical editor for both), that I was an instructor for Imacon, that I was a member of the ICC photography group, etc.

 

So why i talking with at all?

Need for attention if you must force ME to make assumptions. I hate doing so like another here but, you asked a question (not that you can ever accept anyone's answers that don't fit your predetermined opinions).

Maybe the folks on DP Review forums will like your writings better, the folks on LuLa would laugh you out of the color management forum, and you're doing pretty poorly here among the audience.

But by all means, continue the CWOBaT. :)

Edited April 27, 2021 by digitaldog

[digitaldog]

After all these pages, time for words of wisdom from my dear friend and business partner, the last Bruce Fraser. It puts post #1 and so many following posts into perspective:

 

"You can do all sorts of things that are fiendishly clever, then fall

in love with them because they're fiendishly clever, while

overlooking the fact that they take a great deal more work to obtain

results that stupid people get in half the time. As someone who has

created a lot of fiendishly clever but ultimately useless techniques

in his day, I'd say this sounds like an example." -Bruce Fraser

 

Yes, it does sound like an example, to nearly, if not everyone who's replied. :(

[shijan]

Very nice, but from start i don't read those "lyrics" written by some random other people and that you periodically take from your "quotes list" and repeat.

Looking at your background now it is easy to understand the source of all your strange behave here.

As i mentioned, i learned from few of your videos 10 years ago or so. It helped me to understand difference between "Convert to Profile" and "Assign Profile" and how color spaces transformed one into another during color management. It was not something special, just one of the many videos made by many other people i watched that evening. I just remember that low-fi userpic with dog and monitor :) All those basic tutorials and info where useful for start, but i moved further, done a lot of own research and tests, discover a lot of interesting things...

I just thought by mistake that you are more open minded in real life discussions, but not the one who deny things without any reason instead of critical thinking, and speaks like he represent some collective mind opinion of all users on this forum and Luminous Landscape as well.

 

So during 3 decades you never ever think try calibrate scanner with IT8.7 target and see it it helps with negative scans, only because someone wrote in some software review that "it will not work"? Great!

Guess all this time you also only used scanner software to invert negatives because in manual it was written that this is the only way how it should be done...

But ok, instead you can measure something with Delta E, and explain in your video lessons that it is very important to do every day.

[digitaldog]

So during 3 decades you never ever think try calibrate scanner with IT8.7 target and see it it helps with negative scans, only because someone wrote in some software review that "it will not work"?

 

Still unaware of the need and a basic understanding of reference data from targets to build ICC profiles. Pages ago, confused about color accuracy and reference data, confusion about target reference data too. Your hole gets deeper after each post you write. Glutton for punishment.

 

What is your day job sir?

Edited April 27, 2021 by digitaldog

[Ed_Ingold]

I have used several generations of IT8 transparencies to calibrate my Nikon scanners. However that has no bearing on scanning nor converting color negatives. For that you need to photograph a test chart under the same lighting and with the same exposure used for the subject. There are several programs and Photoshop plugins which will create an ICC profile from that test exposure. Unfortunately the profile is nearly useless if either the lighting, exposure or film type is changed. The process is not difficult, but unless you're doing product photography, it's not worth the time and effort. Besides, why would I use film for a paid job? That ship has sailed. My interest is in archiving film I shot 20 years or more ago.

[shijan]

I didn't work with Nikon scanners, so i can't deny or confirm. It may illustrate that Nikon scan have quality enough factory input ICC profiles. Who knows, maybe factory ICC profile where simply generated by Nikon based on same IT8.7 slide target.

It is also unknown yet how exactly you calibrate and process images. It always could be hidden user mistake, or mistake caused by some uncommon tech behave provided by scanner manufacturer for unknown reason.

For example there is a huge known problem with original ICC profiles from Nikon scanners. You can read details in this article. Scanning Colour Negative Film Using ICC Profiles

"It occurred to me that Nikon Scan installs a bunch of profiles for the scanner but they don’t show up in Photoshop or other imaging applications. I tracked down why this was. Nikon had altered the metadata in the header to make them not appear to be valid. Specifically, they changed the “type” from “scnr” (a scanner or input profile type) to “nkpf” (a non-standard value). If you take a hex editor (or even vi) you can just find the value “nkpf” in the first few lines of the file and change it to “scnr” and you will get access to use the profile in any application of your choosing."

 

Lucky there is a download link for fixed (hex-edited) Nikon Scan input ICC profiles: Dropbox - Nikon Scan 4.0.3 Original Hex Edited Profiles - Simplify your life

 

I already post side be side examples and tests that illustrate difference between camera original input profile and custom calibrated camera in post #104 Film Negative Invert and Processing in L* Gamma FAQ , but same as all other useful info it disappeared in this "discussion".

It also was well visible difference in images made with original and custom-made input profile from Minolta 5400II scanner.

[Ed_Ingold]

I used Silverfast, which can both create ICC profiles from an IT8 transparency, and use them when scanning slides. Nikonscan, the oem application, uses only built-in profiles. However you can use Photoshop to Assign a profile to a scan, then Convert to a standard version, including AdobeRGB. I can also create a profile from an image of a test chart, film or digital, and use it in the same manner. It doesn't matter if the original already has an embedded profile. The creation process uses the colors as presented, Assigning changes those colors to conform to the new profile, and Converting allows others to see the corrected colors even if they don't yave the new ICC profile.

 

The same process can be used to correct negative scans, but in a very limited fashion, as described previosly.

 

Sorry, but I've been doing this since you were probably in nappies.

[samstevens]

Sorry, but I've been doing this since you were probably in nappies.

People really say this stuff! It must provide such a sense of self satisfaction and superiority ... or maybe screams just the opposite..

[digitaldog]

People really say this stuff! It must provide such a sense of self satisfaction and superiority ... or maybe screams just the opposite..

Maybe but neither changes the facts ignored here or the facts presented.

 

What's yours in terms of scanning and CM and when might your audience read them? Be unusual for you to post on this actual topic:

Film Negative Invert and Processing....

[samstevens]

What's yours in terms of scanning and CM and when might your audience read them?

Just reading along here. Don’t know enough about the topic. That’s why I’m paying attention: to learn. Since so much that’s personal has been said or implied back and forth by a few of you, I didn’t think my interjection would raise such an eyebrow ... or two ... :rolleyes:

[shijan]

Sorry, i don't have any further interest to proof or deny any things here as well as respond to rodeo_joe|1's and digitaldog's provocations.

I shared my workflow in details. It was developed during many years and tested enough. If you have other workflow and if it works for you - feel free to use your own workflow and be happy.

 

This workflow FAQ (shared here and on many other forums) was a part of my upcoming modular camera scanning system hardware project. It will be manufactured in small batches same as my other existing hardware, and will be available to anyone who interested in fast and quality 35mm camera film scanning.

Stay tuned...

[Ed_Ingold]

I would be more interested were it not for so many fundamental errors and misappropriations. I see a lot of hand-waving and poor craftsmanship in preparing examples. Most of all, there is nothing presented which is as good or better than existing techniques, even though the process is far more involved. This is not a better mousetrap (or wheel).

[digitaldog]

Just reading along here. Don’t know enough about the topic. :rolleyes:

Thanks for admitting that.

Those of us with actual experience in the topic (some decades) appreciate you're in the stands watching the game and have never played.

“Experience is a wonderful thing, it enables you to recognize a mistake when you make it again.“ -Source Unknown

[digitaldog]

I would be more interested were it not for so many fundamental errors and misappropriations. I see a lot of hand-waving and poor craftsmanship in preparing examples. Most of all, there is nothing presented which is as good or better than existing techniques, even though the process is far more involved. This is not a better mousetrap (or wheel).

+1

Peer review is a bitch. :D

[samstevens]

Thanks for admitting that.

Interesting you consider wanting to learn an admission. I guess that goes along with the superiority issue I mentioned.

Those of us with actual experience in the topic (some decades) appreciate you're in the stands watching the game and have never played.

More of the attitude that typifies misery.

Experience is a wonderful thing, it enables you to recognize a mistake when you make it again.

We each have various experiences and come here to share them ... in some cases not so much share as insist. Since I haven’t offered opinions or facts on the topic you’re more experienced in, I haven’t made the kind of mistake more experience in these matters would avoid, so your quote, as is so often the case, is a non sequitur.

[samstevens]

Peer review is a bitch.

So are some peers.