When will JPEG2000 (JPF) support be added

[DeepDesertPhoto]

15 hours ago, R C-R said:

OK, but are you sure all those extra bits actually are used to match the original's 32 bits worth of color variations, or maybe just truncated or rounded to the nearest 8 bit values?

Also, what is the maximum bits per channel your scanner supports without interpolation? Does the color quality of the original photos in their presumably currently aged state really justify even 16 bit depths?

All I am suggesting is it might be worth running some tests before deciding what the best way to go really is.

Sorry for the late reply.
I just got back home from a 7 mile hike at the Grand Canyon.

My scanner is an Epson V-300 flatbed.
Maximum resolution is 9600 PPI without interpolation.
I've used it to scan 2x3 wallet photos and turn them into 8x10s.
The scanner supports RGB 16 bit TIFF.
Unless the photo is a black and white I always scan in 16 bit for all color photos.
Since the photos often need editing, such as sharpening, dust removal, and even repair of tears, I save it in LAB color after editing is complete.
As I mentioned before I find that Lab reproduces the colors better when I make prints.
According to my research LAB is a 3 dimensional color model that combines RGB and CMYK.

My mistake was saving them in JPF.
All photos I still have the original prints for will be rescanned.
NEF files will be reprocessed into LAB 16 bit TIFF.

The OpenXLR 32 bit export I did was more of an experiment. I don't know if I will use it yet.
I will have to see how many JPFs I have than cannot be rescanned or reprocessed from original NEF files.

[R C-R]

59 minutes ago, DeepDesertPhoto said:

According to my research LAB is a 3 dimensional color model that combines RGB and CMYK.

That is not true. LAB is a 3 dimensional color model based on the opponent theory of human color perception. RGB is another 3 dimensional color model, an additive one based on the concept of primary colors. CMY is a 3 dimensional subtractive model, based on the absorption of inks or dyes.

All these color models can be represented as 3 dimensional color spaces; the horseshoe shaped charts you often see depicting color gamuts are really 2 dimensional slices of the 3D space. To fully understand color management & color space conversions it is important to understand the differences among the terms color models, color spaces, & color profiles.

[DeepDesertPhoto]

7 hours ago, R C-R said:

That is not true. LAB is a 3 dimensional color model based on the opponent theory of human color perception. RGB is another 3 dimensional color model, an additive one based on the concept of primary colors. CMY is a 3 dimensional subtractive model, based on the absorption of inks or dyes.

All these color models can be represented as 3 dimensional color spaces; the horseshoe shaped charts you often see depicting color gamuts are really 2 dimensional slices of the 3D space. To fully understand color management & color space conversions it is important to understand the differences among the terms color models, color spaces, & color profiles.

I've been using LAB color for 17 years. I find it better than straight RGB.
Here is an article explaining LAB. http://geraldbakker.nl/psnumbers/lab-explained.html

The biggest advantages of LAB are that the Luminance is separate from the colors and LAB has a wider color gamut than either RGB or CMYK.
I can adjust colors using LAB in ways that just don't work with RGB.
And for me the proof is that way the prints turn out.

You can also make adjustments using LAB that you cannot do with RGB.
Here is a video about making adjustments using LAB in Affinity Photo.

 

[v_kyr]

Yes, using the Lab color space often has it's advantages, since it's first of all perception oriented (so not device specific) and thus offers a much larger color gamut than every RGB based color space.

Quote

CIELAB color space

In the color model of the International Commission on Illumination, the CIELAB model of 1976, the XYZ coordinates undergo a computational transformation to normalize regions of similar color differences around each color space coordinate. The so-called MacAdam ellipses, as areas of equally perceived color contrasts in the XYZ space, were irregularly shaped. Thus, the Euclidean distances of equivalent differences in color sensation should also be better represented mathematically.

CIELAB describes our color sensation better than the tristimulus model (CIE-XYZ) because of the aim of equality. Due to the inclusion of the perceptual component (the normal observer) it describes the color sensation better than RGB, CMYK, HSB or other common mathematical color models. Unlike these, it describes all colors, not just the gamuts that can be represented in a particular technique. As a mathematical color model CIELAB is public domain and freely usable in all respects.

Quote

Conversion from RGB to Lab

Color spaces are complex mathematical bodies and the conversion is system-related. The RGB color space is always device-specific and Lab is perception-oriented. In addition, the L * a * b * color space has a larger color gamut than any (technically tangible) RGB color space. For example, in Lab-> RGB calculations it must be clarified how the colors outside the (RGB) target color space are projected in these (gamut mapping).

Though AFAIK, Affinity Photo doesn't offer or support to use LAB in it's RAW (Develop Persona) processing module, instead just after RAW -> RGB conversion inside Photo Persona ...

• RAW to LAB?

... where other RAW converters (like RawTherapee for example) do offer LAB based manipulations inside their RAW develop mode (RAW -> LAB) and thus not from a predefined color limited RGB color space. 

[DeepDesertPhoto]

14 minutes ago, v_kyr said:

Yes, using the Lab color space often has it's advantages, since it's first of all perception oriented (so not device specific) and thus offers a much larger color gamut than every RGB based color space.

Though AFAIK, Affinity Photo doesn't offer or support to use LAB in it's RAW (Develop Persona) processing module, instead just after RAW -> RGB conversion inside Photo Persona ...

• RAW to LAB?

... where other RAW converters (like RawTherapee for example) do offer LAB based manipulations inside their RAW develop mode (RAW -> LAB) and thus not from a predefined color limited RGB color space. 

Yes, that is true. APh does not have LAB for processing NEF or other RAW camera formats.
When I process my NEF files I have APh set to open the NEFs in 32 bit RGBA Wide Gamut.
After editing I then develop it and convert to LAB 16 bit for any final editing. I then save the final edited image as a TIFF in LAB 16 bit.
I use that TIFF version as my master copy for making any prints and RGB 8 bit JPEGs that I upload to the art sites and stock agencies I deal with.
I would send them the TIFF but most of the websites I deal with only accept JPGs.
 

[v_kyr]

9 minutes ago, DeepDesertPhoto said:

When I process my NEF files I have APh set to open the NEFs in 32 bit RGBA Wide Gamut.

If I remember correctly APh Develop Persona uses a ROMM RGB like mathematical color space when processing RAW files, though some former forum member then showed/proved in the past, that only the sRGB gamut seems to have been used always instead. But I don't know if in the meantime things have been fixed or changed here in this regard during APh version updates. - However, since RAW images do not have color info at pixels and so are not in any color-space (instead they have color primaries corresponding to the wavelengths that the Bayer filter on each pixel defines), using an RGB based color space (even a wider gamut one) in/during RAW development limits the overall possible and theoretical available color spectrum. Thus RAW -> LAB should offer more possibilities than RAW -> RGB -> LAB here from the beginning.

[walt.farrell]

1 hour ago, DeepDesertPhoto said:

I use that TIFF version as my master copy for making any prints

I'm curious why LAB offers any advantages when making prints. (And I'll freely admit that I'm a complete novice about LAB.)

First, I'm not aware of any printers (home printers, at least) that use LAB, so there must be a conversion somewhere from LAB to either RGB or CMYK (or some other combination of inks) depending on the printer.

Second, in my experience (largely limited to EPSON Photo printers of various makes) the colorspace of the image is not the limiting factor. The limiting factor is the gamut of the inks and paper that I'm using.

[R C-R]

4 minutes ago, walt.farrell said:

I'm curious why LAB offers any advantages when making prints.

So am I. As you said, gamut is determined by the inks & paper, not the color space or color model.

[v_kyr]

1 hour ago, walt.farrell said:

I'm curious why LAB offers any advantages when making prints. (And I'll freely admit that I'm a complete novice about LAB.)

First, I'm not aware of any printers (home printers, at least) that use LAB, so there must be a conversion somewhere from LAB to either RGB or CMYK (or some other combination of inks) depending on the printer.

Yip for printing materials LAB can't be used directly, digital images therefor always have to be color mapped/converted here to the color capabilities of the used printing device, meaning the specific printers with their drivers, inks and papers, which then also define the limiting factor and do work instead with the device specific CMYK color space. - See also what common printing companies and services tell about this (for example CMYK • PMS • RGB • LAB: What does it all mean?). - Other than that, a good made short read for LAB is maybe also ...

• What are the pros and cons of using Lab color? - see the LAB PRIMER answer part there, especially related to printing/output also "Coming Down (aka The Cons)"

 

[R C-R]

13 minutes ago, v_kyr said:

See also what common printing companies and services tell about this (for example CMYK • PMS • RGB • LAB: What does it all mean?)

From that source:

Quote

 It has already been established that RGB and LAB are not used for graphics being printed on paper. That leaves two choices: CMYK or PMS color.

 So again, what are the advantages of using LAB when printing?

[v_kyr]

36 minutes ago, R C-R said:

From that source:

 So again, what are the advantages of using LAB when printing?

Quote

With respect to a given white point, the CIELAB model is device-independent—it defines colors independently of how they are created or displayed. The CIELAB color space is typically used when graphics for print have to be converted from RGB to CMYK, as the CIELAB gamut includes both the gamuts of the RGB and CMYK color models.

...

Lab' covers all potential colours, every colour that the human eye is able to see. As such it is an absolute and therefore is not colour managed as there are no degrees of 'Lab' for a colour profile to describe. Because of this Photoshop uses it as its native colour space. So every time an image is converted from RGB to CMYK, or back again, it is passed through 'Lab' in the process.

Mostly these here for color accuracy ...

Quote

What makes Lab more efficient is that it is independent from other factors. For example, you can use the exact same CMYK recipe on 5 different presses and you'll get 5 different results depending on the paper, the lighting in the room, the ink viscosity and many other factors that required adjustments and compensation by the press operator when the color was printed. Because of that, you cannot figure out a precise CMYK recipe simply by analysing the printed result. The same goes for RGB; for example, some screens use a blueish white, even when they are supposed to be neutral and calibrated. Are the lightbulbs in your office perfectly neutral? Probably not, and the amount of sun you get will also affect the colors you see. This makes RGB and CMYK unreliable when trying to reproduce a specific color.

Lab doesn't take external factors into consideration, therefore it gives you a precise color value by finding the attributes of a color. By giving someone else a Lab color number, they know exactly what color you are refering to.

That explains why you are not getting the same color in RGB. Even if you disregard your monitor and just use the numbers, there is a conversion that happens to fit the Lab color in the smaller RGB space, and that conversion is done according to the RGB color profile that you are using and its settings.

What is the purpose of Lab

There are many things that you can do with Lab. Here are but a few:

1. It is the best color space to use when it comes to color conversion as it gives you a precise reference point and it offers the widest range of possibilities.

2. When measuring color with a spectrophotometer you'll usually use Lab, even though most spectrophotometers will also give RGB and CMYK. As I mentionned before, CMYK and RGB are affected by many external factors, which makes them unreliable when you try to reproduce a color.

3. Because of the 3 axis, a Lab color can easily and clearly be represented on a graph.

4. In the printing industry, for example in flexographic printing, Lab is used to measure color variations and to adjust the press/ink accordingly.

As you can see above, it is the best tool for color conversion and for getting a precise color reading when trying to precisely reproduce a color. Because of this, many professionals need Lab.

 

[DeepDesertPhoto]

4 hours ago, v_kyr said:

If I remember correctly APh Develop Persona uses a ROMM RGB like mathematical color space when processing RAW files, though some former forum member then showed/proved in the past, that only the sRGB gamut seems to have been used always instead. But I don't know if in the meantime things have been fixed or changed here in this regard during APh version updates. - However, since RAW images do not have color info at pixels and so are not in any color-space (instead they have color primaries corresponding to the wavelengths that the Bayer filter on each pixel defines), using an RGB based color space (even a wider gamut one) in/during RAW development limits the overall possible and theoretical available color spectrum. Thus RAW -> LAB should offer more possibilities than RAW -> RGB -> LAB here from the beginning.

I set it up that way by going into the preferences.
Here is a screenshot of the preferences I set for color.

[R C-R]

5 minutes ago, v_kyr said:

Mostly these here for color accuracy ...

OK, but printing is inherently device-dependent so at print time color accuracy & gamut are determined by the conversion from whatever color space & color profile are being used in the document to whatever the printer must use to get the best results from its inks, the paper, etc. ... right? 

[DeepDesertPhoto]

3 hours ago, walt.farrell said:

I'm curious why LAB offers any advantages when making prints. (And I'll freely admit that I'm a complete novice about LAB.)

First, I'm not aware of any printers (home printers, at least) that use LAB, so there must be a conversion somewhere from LAB to either RGB or CMYK (or some other combination of inks) depending on the printer.

Second, in my experience (largely limited to EPSON Photo printers of various makes) the colorspace of the image is not the limiting factor. The limiting factor is the gamut of the inks and paper that I'm using.

I used to use Epson printers but I found certain colors just did not come out right and black and white images always had a greenish tint to them.
I switched to Canon and the prints had more accurate colors and black and white images came out with more accurate grays with Canon.
I am not sure why Canon is better, it might be due to how the printer drivers handle images in LAB color mode.
I set this stuff in the printer settings.
I do know that if you leave the printer on factory default settings it will print using whatever the factory settings are.

I don't think the printer actually prints in LAB color.
I think the driver software takes the LAB profile and configures it as close as possible to whatever color gamut the printer is capable of.

All I know is that when I open an image in LAB color the resulting print colors seem closer to what I see on the screen.
When I try to print images in RGB some color shades of don't look the same as what is displayed on the screen.

[DeepDesertPhoto]

14 minutes ago, R C-R said:

OK, but printing is inherently device-dependent so at print time color accuracy & gamut are determined by the conversion from whatever color space & color profile are being used in the document to whatever the printer must use to get the best results from its inks, the paper, etc. ... right? 

I am not an expert on printers.
What I know is based on experimentation.
I know that printers are inherently CMYK because all printers have at least 4 to 5 ink cartridges.
I use a Canon Printer and it has the standard Cyan, Magenta, Yellow, and Black(K).
But Canon also has a 2nd Black cartridge that is listed as Pigment Black.
All of the other cartridges are made of dye ink.

When I open an image that is already set to LAB color mode the print window that pops up has several setting options for color.
Here are two settings I tweak to make sure my images in LAB get printed correctly.
If I leave these settings in factory default the prints don't turn out right when I try to print LAB color mode images.
So it might not be just the LAB color, but also how the printer is set up to handle that color mode.

 

[R C-R]

10 minutes ago, DeepDesertPhoto said:

I don't think the printer actually prints in LAB color.

A printer can print only using the inks or dyes or whatever other type of colorants the process uses. That is what restricts the gamut, not the document color space. Specifying an appropriate color profile, & soft or hard proofing as needed, is the only way to get accurate prints that I know of.

[DeepDesertPhoto]

1 minute ago, R C-R said:

A printer can print only using the inks or dyes or whatever other type of colorants the process uses. That is what restricts the gamut, not the document color space. Specifying an appropriate color profile, & soft or hard proofing as needed, is the only way to get accurate prints that I know of.

Like I said, I came to my conclusions through experimentation.
When I found a setting in my printer that produced the most accurate colors I saved that setting as a separate preset so that I did not have to adjust it every time I tried to print an image.
Notice in the screenshots of my prior comment that I have a preset called "Photo-Gloss-13x19".
That setting is a preset I created specifically for LAB color mode images being printed on 13x19 inch photo paper.
I have similar presets for different sizes and types of images.

[DeepDesertPhoto]

55 minutes ago, v_kyr said:

What is the purpose of Lab

There are many things that you can do with Lab. Here are but a few:

1. It is the best color space to use when it comes to color conversion as it gives you a precise reference point and it offers the widest range of possibilities.

2. When measuring color with a spectrophotometer you'll usually use Lab, even though most spectrophotometers will also give RGB and CMYK. As I mentionned before, CMYK and RGB are affected by many external factors, which makes them unreliable when you try to reproduce a color.

3. Because of the 3 axis, a Lab color can easily and clearly be represented on a graph.

4. In the printing industry, for example in flexographic printing, Lab is used to measure color variations and to adjust the press/ink accordingly.

As you can see above, it is the best tool for color conversion and for getting a precise color reading when trying to precisely reproduce a color. Because of this, many professionals need Lab.

 

This is the reason I use LAB.
Color accuracy.
When I print my LAB color images they do come out very close to what I see on the computer screen.
RGB and straight CMYK color modes don't always look right when compared to what I see on the screen.

[v_kyr]

1 hour ago, R C-R said:

OK, but printing is inherently device-dependent so at print time color accuracy & gamut are determined by the conversion from whatever color space & color profile are being used in the document to whatever the printer must use to get the best results from its inks, the paper, etc. ... right? 

Modern prepress printing is much more complex than that, printers there measure density and adjust/controls inks via color measuring technology, they rely on the process color L*a*b* values to calibrate their press and certify their printing process (density, dot gain, color superimposition, etc.). - See and look for example here, or maybe read through a "Handbook of Print Media" from Heidelberg, or talk with a prepress domain professional about how they do get color accuracy.

 

[R C-R]

8 minutes ago, DeepDesertPhoto said:

When I found a setting in my printer that produced the most accurate colors I saved that setting as a separate preset so that I did not have to adjust it every time I tried to print an image.

What are you using as a standard for judging the most accurate colors? It won't be the same for every photo, & unless your monitor is calibrated the on-screen representation could be quite inaccurate. Among other sources, please refer to this article & the similar topics list at the bottom of the page, plus the "Image Content" section of this one.

[DeepDesertPhoto]

5 minutes ago, R C-R said:

What are you using as a standard for judging the most accurate colors? It won't be the same for every photo, & unless your monitor is calibrated the on-screen representation could be quite inaccurate. Among other sources, please refer to this article & the similar topics list at the bottom of the page, plus the "Image Content" section of this one.

If left on factory default you would be correct. The LCD screen of my Mac would not give me color accuracy straight out of the box because the default setting tends to have a bluer tint.
I carefully calibrated my LCD display by running the calibration program built into the Mac.
It actually took 15 minutes to properly calibrate because I had to manually select some of the test patterns according to how I perceived the blacks and grays.
If white objects looked white on the screen after calibration that is how you know if the colors are correct.
If any white objects or white patterns on the screen have any kind of tint then the calibration needs to be redone.
 

[R C-R]

13 minutes ago, DeepDesertPhoto said:

I carefully calibrated my LCD display by running the calibration program built into the Mac.

The builtin calibration app does not produce accurate color calibration, just a rough, OK for casual use, approximation based on subjective estimates that change depending on several factors, including ambient light color, the color of nearby objects, & even time of day & diet.

For accurate calibration you need a hardware device & the appropriate software. See for example this article, one of the linked ones I referred to earlier.

[DeepDesertPhoto]

1 hour ago, R C-R said:

The builtin calibration app does not produce accurate color calibration, just a rough, OK for casual use, approximation based on subjective estimates that change depending on several factors, including ambient light color, the color of nearby objects, & even time of day & diet.

For accurate calibration you need a hardware device & the appropriate software. See for example this article, one of the linked ones I referred to earlier.

I'm using a MacBook Pro Retina Laptop.
I went to a website for testing the accuracy of my LCD screen and even though it is not 100% accurate it is in the "good" range.
The only thing that is visibly out of adjustment is the gamma.
The site also said that all monitors should be calibrated once every month or two depending upon usage.
I checked the date of my last calibration and it is dated January of 2019.
So it is overdue for a calibration.
I ran a builtin utility for the color profiles in my Mac and out of 65 profiles it only found 3 errors.

Since it takes a significant amount of time to do this calibration I will have to get to it another time.

But another way I know that my photo colors are accurate, or at least good enough for commercial use, is through the agencies I sell my photography through.
They are very harsh critics and if something is visibly wrong with the photos I submit they will reject them and tell me what is wrong with them.
I have never gotten any rejects for color problems.