[why are pantone conversions based on SRGB RGB instead of L*a*b*?]

19 hours ago, Hangman said:

It seems to me that one of their main philosophies is to provide a comprehensive alternative to all the other colour systems out there and to provide their colour system free of charge to end users so you don't end up with third-party ownership of individual colours

I now had a closer look on this, and indeed their goal is to compete with other global color reference systems. What you mentioned above about possible monetary kickback for Adobe because of change of PANTONE policy does not sound such a crazy idea since I suppose Adobe could well have included HLC Colour Atlas (if it is really free) as an alternative color book for PANTONE (or at least as an .ASE file) in their apps. It would certainly be a good idea to have it included in Serif apps.

So it appears that the system will (or perhaps already does) include also instructions for mixing inks. For the moment I suppose giving printers something like this would cause some confusion, though:

The freedom speech aside (since if someone is going to use this as a replacement of Munsell, NCS or PANTONE, it is going to cost a bit, too), this could well become a universal color system one day!

[why are pantone conversions based on SRGB RGB instead of L*a*b*?]

34 minutes ago, Hangman said:

They are all created by freieFarbe e.V., this absolutely isn't a collection of resources gathered from other locations...

Sorry, I was inaccurate. I did not mean the samples and file generation but information itself, so e.g. Munsell ASE file is not naturally intellectual property or creation of their own even if the file is generated by them and color samples carefully produced. But what you mention here is certainly good to know.

But I also asked because I noticed that they urge visitors (or just specifically color system manufacturers?) to upload palette files, and that they also seem to state a typical open-source disclaimer saying something like "if you notice that information shared here violates someone's copyrights, please let us know".

But I'll go and view more carefully what is available and what is stated and stop making more wrong and hasty assumptions! Thanks for sharing this resource!

 

[Missing Font OCRB Barcode Error]

When using embedded fonts, Affinity apps basically behave in three ways:

1) When the font is embedded in a PDF file and the file is placed to be passed through:

• The embedded font will be passed through whether the font is available on the system where export happens or not, unless there is a PDF compatibility issue
• If there is a compatibility issue (e.g. the placed PDF uses PDF version 1.7 and the file is exported using a lower version, like when exporting using PDF/X-based methods),  the conflicting parts will be rasterized and color values reinterpreted and/or the embedded font will be replaced.

2) When the font is embedded in a PDF file and the file is placed as interpreted:

• If the embedded font is installed on the system and the font is correctly mapped, the parts using the embedded font will be correctly rendered and exported.
• If the embedded font is not installed or not correctly mapped, a replacing font will be used.

3) When the font is embedded in an EPS file:

• The font will be converted to curves whether the embedded font is installed/available or not.

When generating things like barcodes (where not only the text parts can be rendered with a font that is embedded in the exported file, but also the barcode itself can be created using glyphs of a special font), it is safest to export using EPS format, if available. If possible, the color format should be for print purposes K100 (no C, M or Y). When generating PDF format barcodes, the most compatible format is non-PDF/X-based one that uses lowest possible version number (1.3; as for Affinity apps, 1.4 will also work in all situations) and that is placed to be passed through.

The safest choice, if available, is to specify that all font based elements are converted to curves.

An example of a PDF where the embedded font was not available, but where there were no compatibility issues in context of PDF placement and export:

barcodefonts_missingfont_nocompatibilityproblem.pdf

An example of a PDF whether the embedded font was not available and additionally there was a PDF compatibility issue (placed PDF was v. 1.7 and PDF version used when exporting was 1.4):

barcodefonts_missingfont_compatibilityproblem.pdf 

In both export PDFs the EPS file works without problems.

[why are pantone conversions based on SRGB RGB instead of L*a*b*?]

Ok, sorry for jumping to false conclusions. Are there any yet available that are creations of their own, or is this merely a collection of resources gathered from the Internet and posted by users? Anyway, a nice resource, and useful even without an exhaustive, global and free alternative color system ever being devised and developed.

My point was more or less that IMO there is no need for such a system, as alternatives are so widely available, many of which free of charge. On the other hand, I think that printed color books like those of Munsell, NCS, and PANTONE, and associated digital references with tools and features making a specific color system still more useful, continue to be needed in certain workflows and will continue to be something that costs. But in general use they have become more or less obsolete since color management went mainstream.

[Pixels and Images and Alphas, Oh My!]

In short "Image" layers are basically what Smart Objects are in Photoshop, to give you a more familiar concept:

https://helpx.adobe.com/photoshop/using/create-smart-objects.html

Channel editing is pretty much different than in PS so that just needs to be learned.

[why are pantone conversions based on SRGB RGB instead of L*a*b*?]

18 hours ago, Hangman said:

who have developed what is effectively a licence free alternative colour system and one that would/could ultimately compete with Pantone...

I could not access the video earlier but watched it now. It is not clear but it seems that they do ultimately aim at custom ink based color system similar to PANTONE and if so then what I mentioned above is pretty much off the topic.

However when I tried to find out about the development after 2019 I could not find much anything than a database and conversion utility covering over 300 color systems at https://freiefarbe.de/en/thema-farbe/farbdatenbank/. As a graphic designer I am not sure if I want to know more of any of these industrial color systems most of which are unknown to me but if there is ever need, I'll remember this site.... But presently, if there is any need for special inks and the free sRGB based PANTONE color data and representation is not sufficient, I rather "rent" PANTONE for the time I really need to access full color data and add the cost to the invoice.

Perhaps it is the big industry, fashion, interior design etc. that benefit from "open source" color, but for the moment there does not seem to be much in the interest of independent graphic designers.

EDIT: If freieFarbe intends to enter in the world of physical colors, it is not PANTONE they are competing with but the parent enterprise Danaher Corporation who owns X-Rite, PANTONE and Munsell. Yes, please give us free color calibrators, ColorChecker Passport equivalents, printed reference books, universal L*a*b based digital color books, etc. I believe immediately when I see! But if all that is provided is a database of color values of single swatches, it is not nearly enough, because it is not just about communication of color information, but integration with design tools -- apps and devices -- that counts. E.g. their database includes Lab-based color values of Munsell color system, but not e.g. a digital palette, which can be purchased separately from e.g. Adobe Stock at EUR 9.55 as an .AI file that can be exported to an .ASE palette, but then again, you need a CC based AI to be able to make the conversion. See, they are all connected and nothing is given free, but it also does not cost awfully lot assuming that you do this to make a living.

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

24 minutes ago, MxHeppa said:

such uses is ones what i also thinked. totally creative ways use image colours for abstract art.

Yes, manipulating decomposed grayscale channels or interchanging them definitely allows all kinds of color experiments, but in some cases are also a means to get improvements that are difficult or impossible to achieve by other means. 

[why are pantone conversions based on SRGB RGB instead of L*a*b*?]

13 hours ago, Hangman said:

My interpretation, rightly or wrongly, in terms of what freieFarbe e.V. are setting out to achieve is to create an industry colour standard that is free for all in the same way that water should be free for all. Colour should not be something that can be copyrighted or trademarked (in my opinion at least).

But is it a custom ink system that controls how print inks are prepared, or simply just a (digital) color reference? If former I am not sure if they can be free and even competitive (considering creating and testing ink mixes and need for having globally available printed reference books for colors that cannot be represented trustworthy digitally within e.g. sRGB); and if latter, just collections of digital color libraries with profile or L*a*b based reference, then there does not seem to be anything new.

Many people no doubt feel it necessary to pick a color from a predefined and freely distributable color collection, or prefer to deliver a color by some registered name, code or reference (controlled but distributed free of charge by some instance), but I cannot see any issue in current policies. There are free services, and paid services, who already do that. I sincerely welcome any addition freieFarbe can provide, but in a way I see this as a kind of a "counter measure" against some silly thing like PANTONE color of the year: storm in a PANTONE color mug ;-).

I can fully understand that PANTONE charges for their special inks. The point is that most people do not need them. Why should they feel offended if someone needing them a lot has to pay a bit for ease of use (well, they do not get that with any money at the moment 🙂. Charging for color reference does not mean that "this blue within the gamut of human vision is no longer free".

13 hours ago, Hangman said:

is there any likelihood that they may have to adopt the same stance as Adobe by removing the PANTONE libraries

A good question. In the long run perhaps. But I think that PANTONE is primarily interested in protecting L*a*b based references distributed as database-kind of collection, whether legacy .acb books, or some other method. Serif .CSVs distribute sRGB data, and that is freely available also via PANTONE Connect that can be accessed via web (so no plug-in is required), and without cost. I think it is likely that in the future PANTONE colors need to be accessed like this. But it is practically impossible to control the legacy use so they are probably focusing on "new" colors and trends. In a word, professional, industrial use.

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

13 minutes ago, MxHeppa said:

now i wonder how earth is possible but rgb->hsl conversion what i think part of procedural texture becouse formula what i found such have parts what i cannot put formulas.

Are you referring to the live filter that @NotMyFault had created and which he today updated? It should be available in an .aphoto attachment I placed above in one of my posts, and I think that also in the original post of @NotMyFault which he today has updated. Hopefully it should also be integrated in the first macro I posted above. This is a handy filter because it avoids the tedious process of needing to use a third-party app or plug-in to create the the grayscale channel composites (which additionally need to be loaded from a file system because no direct layer support).

[why are pantone conversions based on SRGB RGB instead of L*a*b*?]

I think that there is very little need for any "alternative" trademark or authoritative (even free, community based) color books or libraries. Custom palettes (whether L*a*b or sRGB-based) are very easy to create and share and there are such online palette-providers already plenty some free some subscription based or otherwise charging something for their services.

Since color management is supported so widely (including web) there has not practically been much need for specifically "matching" colors against printed references or a specific color system for well over a decade. One gets more predictable and trustworthy colors with a calibrated display (with wide enough gamut to cover printing or audience needs) and ISO color profiles than with any such color reference system.

PANTONE and similar systems are still useful when there is need for custom inks, and then a printed reference is still the best guide. The change in PANTONE policy can be a bit of a nuisance for people who constantly need these colors in their workflows, but they are professional designers and the cost is trivial, and creating e.g. spot swatches is pretty easy even without linked color books and even if the current PANTONE Connect plug-in at its current stage is a poor legacy component and its UI is awkward and usability very poor.

The impact to most users is mostly unnoticeable because disruptive behavior has been diminished, and workarounds (e.g. to assign an sRGB representation free of charge in situations a color book reference is broken) pretty easy. For those users who are accustomed to use trademark color books as a collection of conveniently accessible and somehow meaningfully grouped colors there are plenty of other options available, directly based on shareable, accurate, profile-based color values. PANTONE is by no means needed for that. Personally I'd prefer a practice where color values are always just given in context of profiles.

EDIT: One further point: instead of using fixed predefined color palettes, there is more practical need for creating custom palettes based on some criteria, e.g. on accessibility, color harmony, etc., but typically something that can be created on principle of color styles, and which can then easily be switched so that alternative designs with different palettes can easily be generated. Such services are widely available on cloud based template based designs, and such palettes can subsequently be shared not just with team members and designers using the same service but with anyone involved.

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

2 hours ago, NotMyFault said:

You can find import/export under the burger menu, in the category level entries.

Thanks, got it.

Here's an example of using two macros for making editing less clunky.

2 hours ago, NotMyFault said:

You can do it that way, converted (destructively) to pixel layers, and it certainly has its advantages.

Yes, these kinds of files just become easily quite huge so it might be a good idea just to delete the intermediate stages once finished.

Below are the two macros that I use on the video. I have not created many .afmacros so I am not sure if they should always be initiated somehow to ensure proper behavior. Well, it is not difficult to re-record, hope these give an idea how to do it. 

RGB2HSL_PrepareEdit.afmacro

RGB2HSL_FinishEdit.afmacro

EDIT: I have no idea if the live filters are actually saved within a macro (as I have saved them as presets), could someone confirm!

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

Here is demo of full roundtrip using live filters by @NotMyFault for RGB2HSL and HSR2RGB conversions. Spare channels are used to create grayscale layers that are used for editing and then, after editing, another set of spare channels are created to recompose the HSL image with live filter to convert to RGB.

One macro could be created to apply the HSL live filter on original RGB, creating spare channels and grayscale images and turn the image to grayscale mode, and then another macro that creates spare channels of edited grayscales, converts the image to RGB and then recomposes HSL and applies HSL2RGB live filter.

A resized smaller version of the image used in the clip attached below, hope nothing went wrong along size reduction:

RGBHSLRGB_Roundtrip_small.afphoto 

NOTE: This image has already @NotMyFault's RGB2HSL live filter fixed.

Photo: John-Mark Smith | Pexels

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

1 hour ago, MxHeppa said:

but now i wait recompostion solution

It is because of after decomposition the channels are typically wanted in grayscale for editing, and then being quickly recomposited for preview, that using an app like GIMP might be the best choice. But creating macros to do these conversions within APhoto might effective, too. The process could also be made automated for decomposition once there is a working live filter, meaning that it would be possible to start with an RGB image on one layer, having it split to specific channels using a macro that applies a live filter, then creates spare channels and grayscale images for editing, and finally allows recomposing running the process in the other direction.

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

Here is a clip where grayscale decompositions made by G'Mic plug-in are opened in APhoto (v2) into separate layers and then each layer is saved as a spare channel. The spare channels are then loaded into R, G and B channels of a new layer, and the live filter by @NotMyFault from the post linked above (also included in my .aphoto roundtrip file in one of the posts above) is applied on the recomposed image to get the HSL values mapped to RGB.

It would probably be easiest to just use GIMP and decompose and recompose there. It would also be easy to do the decomposition by using the Python script that I provided, because it makes directly an RGB file with H, S and L mapped to R, G and B of an 8-bit bitmap, which could then be opened in APhoto and just have the live HSL2RGB filter applied on it. But that of course requires Python 3 to be installed with the required libraries, which would be an additional effort if not already done.

NOTE: When you open the G'Mic created decompositions, they are in Grayscale mode. Make sure you have the images as pixel layers. You can have them in layers in a grayscale image and then save the spare channels from Intensity channels, or you can convert the image directly to RGB and then spare the H, S, and L values from any of the RGB channels (it does not matter from which because they all have identical intensity values). But when you create the RGB with HSL values to recompose, you need to have RGB color mode to have the HSL spare channels loaded into R, G and B channels.

EDIT: You can use the same methods to decompose and recompose L*a*b and CMYK images. If you have a specific workflow, you can save a lots of time by creating macros.

Photo: John-Mark Smith | Pexels

[Missing Font OCRB Barcode Error]

9 hours ago, Row13 said:

Does anybody know how I can clear this Missing Font error? Thanks in advance

If the placed file is an EPS, you can open it and render the embedded font as curves.

If the placed file is a PDF, place it for "Passthrough" (= default) so that the embedded font is used when you export. If you place a PDF as interpreted file, you need to have the embedded font installed.

Certain apps like Adobe Acrobat Pro and macOS Preview, can also flatten the fonts and convert an embedded font to outlines.

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

On Windows, G'Mic plugin could be used to decompose layers into many different channels:

It is not possible to output split channels onto layers when using Affinity Photo, but the files are saved onto disk and then opened for manipulation. Recomposition could not be done via G'Mic, because multiple layer input cannot be used, but the recomposition can then be done e.g. by composing an RGB image from spare channels created from grayscale separations and then using the live filter.

[Issue with colors when opening a PNG or JPG file]

Could it be that on the second mac the image has embedded Display profile, while on the first one it has either sRGB embedded or no profile at all (in which case sRGB will be assigned as a default working profile)? When you create screenshot on macOS and image is saved on disk, the Display profile will automatically be embedded. When you open such file, the embedded profile will be used by default. 

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

On 1/31/2023 at 11:35 PM, NotMyFault said:

A long time ago I made procedural texture filters to convert from RGB to HSL and vice versa.

Thanks! The HSL to RGB filter works nicely, but I cannot get the RGB to HSL work properly to use it for HSL separation (and subsequent saving to grayscale channels), so I have used the following modified Python script (picked from codedrom.com) to first create HSL values separated and combined into R, G and B channels.

import colorsys
from PIL import Image
import numpy


def main():

    print("-------------------------")
    print("| codedrome.com         |")
    print("| RGB to HSL Conversion |")
    print("-------------------------\n")

    try:

        image = Image.open("test.png")

        hls_array = create_hls_array(image)

        new_image = image_from_hls_array(hls_array)

        new_image.save("test_hls.png")

    except IOError as e:

        print(e)


def create_hls_array(image):

    """
    Creates a numpy array holding the hue, lightness
    and saturation values for the Pillow image.
    """

    pixels = image.load()

    hls_array = numpy.empty(shape=(image.height, image.width, 3), dtype=float)

    for row in range(0, image.height):

        for column in range(0, image.width):

            rgb = pixels[column, row]

            hls = colorsys.rgb_to_hls(rgb[0]/255, rgb[1]/255, rgb[2]/255)

            hls_array[row, column, 0] = hls[0]
            hls_array[row, column, 1] = hls[1]
            hls_array[row, column, 2] = hls[2]
            #print("Hue", hls[0])
            #print("Lightness", hls[1])
            #print("Saturation", hls[2])

    return hls_array


def image_from_hls_array(hls_array):

    """
    Creates a Pillow image from the HSL array
    generated by the create_hls_array function.
    """

    new_image = Image.new("RGB", (hls_array.shape[1], hls_array.shape[0]))

    for row in range(0, new_image.height):

        for column in range(0, new_image.width):

            rgb = (round(hls_array[row, column, 0]*255), round(hls_array[row, column, 2]*255), round(hls_array[row, column, 1]*255))

            new_image.putpixel((column, row), rgb)

    return new_image


main()

This kind of an RGB could then be opened in APhoto, its HSL values in RGB channels converted to RGB with the live filter (or if necessary, saved into spare channels and then saved in grayscale layers representing Hue, Saturation and Lightness).

rgbhslroundtrip_v01.afphoto

As you mentioned, the HSL live filter does not work at all in APhoto v2 because of some change in reserved words. It would be great to have the code updated and also changed so that it could be directly used as a HSL separator similarly as the Python script (I have just have done something wrongly to not having been able to use it that way).

BTW: GIMP (and of commercial apps, e.g. CorelPHOTO-PAINT) has a pretty comprehensive channel separation and combination support, so it could be used to do these kinds of task without a fuss:

 

 

I have not been using the procedural texture filter much so I probably failed to find a way to save a user-defined filter calculation for later use. I could create a preset category and import presets, but how to export or save? Being able to do that would make user-defined filters much more effective.

[Is possible combine three black white images way one is L and others are A,B. and similar to CMYK.]

As APhoto macros are kinds of black boxes, here's an .aphoto file with a history starting from a grayscale image that has L, a and b as distinct layers, showing a procedure of converting the image to Lab using spare channels. I think this was pretty much what the macro above did (though it might have been built on a bit different starting point).

Grayscale2LAB.afphoto

[why are pantone conversions based on SRGB RGB instead of L*a*b*?]

Another issue [specifically within Affinity apps] is that even if PANTONE colors were given in L*a*b space they cannot be shown in that color space, unless the gamut of the document color space is also L*a*b (or wide enough to cover PANTONE definitions). In practice spot colors are typically used in CMYK color documents in which case many PANTONE spot colors look quite desaturated and unrealistic.

[Unpopulate/Remove Placed Images Form Selected Picture Frames]

I think that there is a bug (or kind of a silly omission) in the code since if you manually expand the image groups, the Locate in document button becomes available (non-grayed out) and you can then activate the selections on the canvas and delete the images. The Locate button should be available even if the image groups are not expanded because they are actually already selected within the groups. 

I guess this is still a bit quicker than first using the context menu to expand all layers within Layers panel and then selecting image resources from within the Layers panel? I guess it much depends on whether you have the images large enough on the Layers panel so that you can recognize them from the thumbnails (note that you can change the size of the layer thumbnails).

[Boolean operations with letters do not work in Affinity Designer 2.0.3!]

I played a while with v.2.0.4. Divide operation (now mainly on macOS) but still could not get it work consistently. But the feature is broken on both platforms, and now I got different behavior also on macOS depending on font, and also depending on whether text based objects (whether still as frame or artistic text or converted to curves and merged) were involved or not. So I got bored in the middle of trying to produce a similar chart of mere Division operation in different situations (both shapes filled, only top or bottom shape filled, neither shape filled, and then performing Divide simple and Alt+Divide), because getting so varied and inconsistent results.

I am not sure about usefulness of getting different results for Division depending on having the Alt key pressed down or not. I would prefer the operation to behave similarly as Illustrator so the operated shape areas would always be retained after being divided, no matter if filled or unfilled (so no area would ever be lost and could afterwards, after division, be filled if wished). I can see some limited usability in Inkscape kind of Division where the division is confined within the area defined by the top object and the rest is removed (disregarding the fill state), and perhaps this could be a possible functionality for holding down the Alt key during the operation.

[Boolean operations with letters do not work in Affinity Designer 2.0.3!]

9 minutes ago, ,,, said:

Alt/opt-clicking Divide does not produce a Compound. The alt/opt key is a modifier for Divide when one or more unfilled curves are in the collection of operands. The key is used when you want the fragments of unfilled curves to be kept, otherwise they are discarded.

Ok, thanks for clarifying, I need to examine this operation more closely to understand it better. I have previously thought that Boolean operations are pretty similar in different apps, and to certain extent they of course are, but they work differently in Affinity apps and e.g. CorelDRAW, Illustrator, VectorStyler and Inkscape, and also use a bit different terms.

[FAT binaries are wasted disk space ...]

18 hours ago, v_kyr said:

Especially the MAS/Apple Store should IMO be usually able to handle that,

But does it in practice? I meant promotion much in the sense that including Silicon distribution might be a kind of a requirement to be able to deliver via App Store in the first place [and a means to get the developer get their app support as soon as possible the new architecture] -- and then the fat binaries (Universal apps) would basically be a pragmatic requirement for also ease of use? 

[FAT binaries are wasted disk space ...]

33 minutes ago, v_kyr said:

But and that is a big BUT, there was always the possibility during PKG installation, or afterwards, to easily strip out automatically the by your platform unneeded architectures from apps in order to save disk space.

I think that there is a chance that separate deliveries are if not prevented then discouraged in order to promote that latest and greatest, especially if the method of distribution is Apple Store? Just a guess. There are these kinds of restrictions also for delivery via Microsoft Store (I am not sure but I think that if separate 32 and 64-bit packages are wanted to be delivered, it must happen off the MS store...).