Recent Posts

I suggest you get the client to print a test chart using Adobe Colour Print Utility which can then be sent to you by snail mail.

Jonathan

FWIW, the last few versions of macOS, most certainly Tahoe, have broken ACPU. The likelihood that Adobe will fix it seems slim.

I suggest you get the client to print a test chart using Adobe Colour Print Utility which can then be sent to you by snail mail.

Jonathan

No LR cannot print without color management.

Hello!
I have a client who has asked me as a favour to profile a paper for their new Canon Pro-1100. He is using macOS and printing out of Lightroom.

Generally I don't create profiles for other people, but I do make all of my own. I have only ever used Epson. I use i1Profiler to make targets, print from Mirage which allows for 'no colour management/profile printing', then scan with my i1Pro2 and i1Profiler. I have a good handle on my own workflow, but have never made a profile on another set-up?

Can Lightroom print the target with "no colour management"?
Is there anything different about making a profile for Canon?
Theoretically I can just send him a target and give instructions to print untagged with no colour management and he sends printed target back to me and I scan it? Right?

With thanks
Neil
 

I was having a chat this morning with an ex-photographer, and he raised the point about how durable tiny sensors are or are not.

His observation was that, over time, photographs from these devices begin to look less sharp. It wasn’t about comparisons of cellphones with traditional cameras, but simply a comparison of early photos using a particular cellphone, and those from the very same cellphone a few years later on.

I'm no expert on smartphone light sensors, or smartphone electronics in general, but my experience is that after microprocessing devices survive an "infant mortality" period of a few months, they rarely degrade functionally until they fail completely—if they ever do fail completely.

My guess is that if this phenomenon can be proved empirically, it's the software (firmware) that's to blame.  Smartphone image-processing software tends to involve very aggressive noise reduction, regardless of the ambient light, to compensate for the small sensor size—and, as everybody on this forum undoubtedly is aware, the penalty for noise reduction is typically a loss of edge acuity.  And when the smartphone's post-processing software employs generative AI, it replaces visual elements in the raw image with supposedly improved data generated by a neural network, either for purposes of noise reduction or to eliminate other deficiencies in the image capture; these techniques get more powerful as the software is updated and I wouldn't be surprised if some of them sacrifice edges within the image in order to provide other improvements (or, should I say, "improvements").

There's a special wrinkle for generic smartphones—basically all of them except Apple iPhones (Apple manufactures the hardware and provides all the software)—because the manufacturer (Samsung, etc.) needs to code an "abstraction layer" to provide a connection between its hardware and the interface offered by the Android operating system, which is a specialized version of a more general open-source operating system (Linux).  The creation of this layer may be split between the manufacturer of the chipset that powers the smartphone (e.g., Qualcomm) and the manufacturer of the smartphone proper, which makes the software design challenge even more complex.  (I'm under the impression Samsung is the only manufacturer of both a generic phone and its chipset.  But Samsung still has to write code to make its hardware work with the Android operating system.)

So as software is upgraded over time, there may be regressions in some of its features, either because of deliberate design trade-offs or inadvertence.  I've definitely seen that happen in other contexts.

Elegant and very beautiful.

Thank You.

DxO PhotoLab 10 is out.
Terry McDonald has taken a look at the new version and whether it’s worth making the jump from PhotoLab 9.

Read it here: https://luminous-landscape.com/dxo-photolab-10-worth-the-upgrade/

Elegant and very beautiful.

Claire Rosen’s Birds of a Feather started with a simple idea: photographing birds against ornate wallpaper. Thirteen years later, it became a remarkable body of work built around letting the birds set the pace. https://luminous-landscape.com/claire-rosens-birds-of-a-feather/

This article was fascinating, but immediately raised a slew of questions:

What is the process for obtaining a digital file for making the LVT transparency?
Why make a LVT transparency and then do color separations on that?
What is the process for making the separations (guessing optical filters to create separation negatives)?

Why not make the separations earlier in the process and produce them as 3 LVT monochromes, saving the steps of color LVT transparency and subsequent separation? It seems like these extra steps are not needed to preserve the historical qualities of the dye transfer process.
I understand that the subsequent highlight and shadow manipulation by photographic masks are required to obtain the historic qualities of dye transfer prints, but it seems the intermediate color transparency and separation introduces a place for things to go wrong with no benefit. I would appreciate some enlightment on this!

Hi @old_tv_nut

Thanks for the thoughtful questions. Jean has provided answers, so I’m sharing them here:

What is the process for obtaining a digital file for making the LVT transparency?
For these two projects The Vertigo Project and GodardBardot the digital files were obtained from scans of the Technicolor reels.

Why make a LVT transparency and then do color separations on that?
Because the transparency is the source material for making the separations to make the print.

What is the process for making the separations (guessing optical filters to create separation negatives)?
Yes, Kodak Wratten filters - Red 29, Green 61 and Blue 47B

Why not make the separations earlier in the process and produce them as 3 LVT monochromes, saving the steps of color LVT transparency and separation? It seems like these extra steps are not needed to preserve the historical qualities of the dye transfer process.
The short answer is, that while generating three monochrome LVT separations directly from the digital file may appear to eliminate a step, it introduces a much more difficult calibration problem.

With a transparency the separations are made from a known photographic object rather than three independently generated monochrome records.

You must take into account that the separations are film which are then being exposed onto the matrix, which is also film. The matrix is affected not only by the density of the separations but also by the spectral characteristics of the separation filters, the spectral characteristics of the film used to make the separations, how both the separations and the matrix have been developed - everything must line up so that the matrices can be dyed and layered onto one another to reproduce a colour balanced print with clean, crisp, bright tones.

If you eliminate the transparency and create three separations from an LVT, these separations will not be calibrated to dye transfer and when the matrices are dyed and layered, they will not produce a linear image with the correct densities.