smadell

Walt...
I've attached an image with i) a circular image of a flower on top; ii) a series of black lines beneath it; and iii) a mask on the flower layer. This seems to work the way it should. I'll note that I started with a canvas measuring 1024x1024 and tried to resize it to half-size, and the mask got royally screwed up. I haven't quite figured out why. But, I deleted the mask and created a new one on the smaller document, and this works well. Was this a "lock children" kind of issue? a bug? I don't yet know. Anyway, consider the attached a file as a kind of "proof of concept" in that it can be done; implementation was a bit clunky, though.
Photo with Radial Mask.afphoto

TONYGAMBLE!!!  DON'T DELETE ANYTHING!!!  (unless you already have...)
Before you delete and re-install...
The tools you are looking at are NOT from the Develop persona. You are looking at the tools from the TONE MAPPING persona. No wonder it didn't work!
In the list of Persona icons at the top left, there is the (i) Photo persona; (ii) Liquify persona; (iii) Develop persona - shaped like a hexagon; (iv) the Tone Mapping person - shaped like converging vertical bars; and (v) the Export persona. When you open a Raw file, it will open in the Develop persona. If you click on Develop, you enter the Photo persona. To re-enter the Develop persona, click on the THIRD icon in the row; you have entered the Tone Mapping persona, where you will not find what you want.
Also, from what I've seen, when you're in the Develop persona working on a Raw file, the White Balance tool will adjust not only the Blue-Yellow axis, but also the Magenta-Green axis. Remember that, if you develop a Raw file and then re-enter the Develop persona at some later time, you are NO LONGER working on a Raw file.

Assuming that you’re in Photo, attach a Mask to the layer with the image that should be transparent in the center. With the mask layer selected, put a Radial gradient into the mask - black in the middle (to make the center of the image transparent) extending to white at the periphery (to make the edges of the image opaque). Slide the center point of the gradient back and forth to change how quickly the transparency gives way to opacity.
If you put the second photo underneath, it should show through in those areas where the top image’s mask is black.

If you are doing the watercolor directly in Affinity Photo, then the suggestion by DWright would seem to be the best one. However, if you are trying to get rid of an "almost white" background in an existing image, your best option might be using Blend Ranges. (1) Open the image in Photo; (2) Put a Fill Layer below it – I've used Red for best visualization; your choice of color will depend on the colors already in your image; (3) Open the Blend Ranges for the image (not the fill layer) and set them to something like what's shown in the screenshot below. I've set everything to be visible on my painting layer, up to a luminosity of 98%. Above that luminosity level, the painting layer goes from completely visible (at 98% luminosity) to completely transparent (at 100% luminosity). Sliding the node at 98% back and forth will let you see what parts of the painting are being affected, since the Fill Layer will show from below in those areas; (4) Delete, or Hide, the Fill Layer. Or, change the Fill color to pure white.
Then, you can print onto your paper and the areas that were originally "almost white" should now be transparent (or pure white, if you changed the Fill color), and should transition from opaque to transparent/white smoothly rather than abruptly.

I am attaching an Affinity Photo macro that turns a photo into a “Paint by Numbers” image. The attached file is a macros category, and should be imported through the “hamburger” menu at the top right of the Library panel. The category can also be imported into the iPad version of Affinity Photo, although there is one important limitation (more on this later).
When you use the macro, it creates a Group called “Paint by Numbers Effect.” All of the changes are inside of this group, so you can turn it on and off simply. Once you look inside the group, you will see multiple layers. From bottom to top, these are:
1) Original Image - Merge Visible
This is a “merge visible” version of your photo. It includes all of the editing you may have done up until that point.
2) Posterization Adjustment
This is a Posterize adjustment layer, and is meant to reduce the number of colors/tones used in the effect.
3) Outlines
This is a separate layer, created (in part) by using a Detect Edges filter. It provides outlines for the areas of color. This mimics the outlines that were present on the Paint by Numbers boards we used as kids.
4) Normalize Colors
This is a copy of the original Merge Visible image, and has its blend mode set to Color. This is used to reset the posterized colors to more natural ones.
5) Adjust Brightness & Contrast
This is a finishing adjustment, and can provide a better final effect.
* * * * * * * * * * * * * * * * * * * *
Once you click the macro, you will be presented with a default version of the effect. A dialog box allows you to set a number of parameters. As you change each of these parameters, the Paint by Numbers effect is updated. You are asked:
Posterize - How Many Colors?
The default value is 4, but values between 3 and 6 generally give good results. If your image is a portrait, judging the final result by looking at what the different values do to the subject’s face is a good idea.
Outlines - Adjust the Opacity
The default is 50%. Adjust this upward to make the outlines more prominent; adjust it downward to make the outlines less obvious. Set to 0 to make them go away entirely.
Finish - Adjust the Brightness
Finish - Adjust the Contrast
The default is 10% brightness, and 20% contrast. Adjust these up or down to give you final effect the desired finish.
When you click Apply, the effect is finalized. Obviously, you can manually change any of the settings after the fact. However, you should know that while the number of posterization levels for the image is non-destructive, the originally chosen number is also used to create the outlines, and this is a destructive change. Although you can change the posterization level after the fact, it is not advisable to do this. The outlines might no longer line up with the individual areas of color.
Also, for some reason, the iPad version of Affinity Photo handles the macro pretty well but will not allow you to change the number of posterization levels before finalizing the effect. It is baked in at 4 levels. You can change this after the fact, but (as above) the number of levels in your posterized image may not match your outlines very well.
Here is the macro in action. The original image is top left; the parameters dialog is shown in its default state, and then changed during the course of the macro; the final effect is shown top right

As always, I am one person with one computer and have not tested this in every possible scenario. Try it and, if you like it, keep it and enjoy it. This forum has provided me with so many good ideas and answers to questions; this macro is my attempt to “pay it forward.”
[Note: Credit where credit is due. I am indebted to Dave Straker, whose recent YouTube video gave me some excellent ideas for this macro. Dave’s channel is called “InAffinity,” and is a steady source of helpful information. Thanks, Dave!]
 
Paint by Numbers.afmacros

In the past few years, I have posted a number of Macros for anyone who wants them. Some have been more widely downloaded than others. With the release of version 1.7, however, I have found that a number of them are using Adjustment Layers that have been updated by Serif. The macros now contain a step invoking the "legacy" version of the adjustment which , once the macro completes, is then automatically updated to the new version.
So far, so good.
But, version 1.7 puts up a notification stating that the adjustment has been updated, and this notification appears each and every time the macro is invoked. Even though the notification eventually goes away on its own, it's certainly an eyesore if you use the macros more than sporadically.
Bummer.
So, I am attaching updated versions of the macros that have been affected by this change. These include the Infrared Simulation Effect as well as the more-recently posted Saturation Mask macros. They are labelled as "v1.1" macros, since they have been re-coded with the new versions of the adjustments. Additionally, minor changes were made to the Infrared macro such that all the added layers are contained within a Group (which can be easily turned on or off with a single click).
Both of these are attached to this post, and can be downloaded here. They are macro categories and therefore should be imported into the Library panel (not directly into the Macros panel). You can feel free to delete the older versions, although those older macros will still work - they will just give you the "nag screen" each time they are invoked.
For reference, here are links to the original posts for both sets of macros:
Infrared Simulation: https://forum.affinity.serif.com/index.php?/topic/40759-infrared-simulation/
Saturation Masks: https://forum.affinity.serif.com/index.php?/topic/76900-saturation-masks-a-simple-method/
Please enjoy the updated macros.
Infrared Simulation v1.1.afmacros
Saturation Masks v1.1.afmacros

If you are doing the watercolor directly in Affinity Photo, then the suggestion by DWright would seem to be the best one. However, if you are trying to get rid of an "almost white" background in an existing image, your best option might be using Blend Ranges. (1) Open the image in Photo; (2) Put a Fill Layer below it – I've used Red for best visualization; your choice of color will depend on the colors already in your image; (3) Open the Blend Ranges for the image (not the fill layer) and set them to something like what's shown in the screenshot below. I've set everything to be visible on my painting layer, up to a luminosity of 98%. Above that luminosity level, the painting layer goes from completely visible (at 98% luminosity) to completely transparent (at 100% luminosity). Sliding the node at 98% back and forth will let you see what parts of the painting are being affected, since the Fill Layer will show from below in those areas; (4) Delete, or Hide, the Fill Layer. Or, change the Fill color to pure white.
Then, you can print onto your paper and the areas that were originally "almost white" should now be transparent (or pure white, if you changed the Fill color), and should transition from opaque to transparent/white smoothly rather than abruptly.

If you are doing the watercolor directly in Affinity Photo, then the suggestion by DWright would seem to be the best one. However, if you are trying to get rid of an "almost white" background in an existing image, your best option might be using Blend Ranges. (1) Open the image in Photo; (2) Put a Fill Layer below it – I've used Red for best visualization; your choice of color will depend on the colors already in your image; (3) Open the Blend Ranges for the image (not the fill layer) and set them to something like what's shown in the screenshot below. I've set everything to be visible on my painting layer, up to a luminosity of 98%. Above that luminosity level, the painting layer goes from completely visible (at 98% luminosity) to completely transparent (at 100% luminosity). Sliding the node at 98% back and forth will let you see what parts of the painting are being affected, since the Fill Layer will show from below in those areas; (4) Delete, or Hide, the Fill Layer. Or, change the Fill color to pure white.
Then, you can print onto your paper and the areas that were originally "almost white" should now be transparent (or pure white, if you changed the Fill color), and should transition from opaque to transparent/white smoothly rather than abruptly.

If you are doing the watercolor directly in Affinity Photo, then the suggestion by DWright would seem to be the best one. However, if you are trying to get rid of an "almost white" background in an existing image, your best option might be using Blend Ranges. (1) Open the image in Photo; (2) Put a Fill Layer below it – I've used Red for best visualization; your choice of color will depend on the colors already in your image; (3) Open the Blend Ranges for the image (not the fill layer) and set them to something like what's shown in the screenshot below. I've set everything to be visible on my painting layer, up to a luminosity of 98%. Above that luminosity level, the painting layer goes from completely visible (at 98% luminosity) to completely transparent (at 100% luminosity). Sliding the node at 98% back and forth will let you see what parts of the painting are being affected, since the Fill Layer will show from below in those areas; (4) Delete, or Hide, the Fill Layer. Or, change the Fill color to pure white.
Then, you can print onto your paper and the areas that were originally "almost white" should now be transparent (or pure white, if you changed the Fill color), and should transition from opaque to transparent/white smoothly rather than abruptly.

Thank you for the shout-out, jmwellborn. Your valentine is certainly appreciated!
- smadell

Today, let’s have fun with Gradient Maps…
Gradient Maps are a great way to color grade photos, since they map dark, light, and midtone values according to a pre-defined gradient. This can easily define a “look” for photos that might otherwise be hard to accomplish.
Although I am not a user of Photoshop, our “arch rival” has a wonderful group of gradient maps called “Photographic Toning” that are specifically designed to color grade photos. Although you have to drill down through a few panels and dialog boxes to find them (and you sort of have to know they’re there in the first place) they are a truly nice addition. Also, they are easily obtained by anyone.
But, as they say, there’s the rub. Affinity Photo cannot import Photoshop gradient maps. Affinity Photo can store pre-defined gradients in the Swatches studio, but for some unknown reason the stored gradients are not available from the Gradient Map studio panel. Gradient Maps can be stored as presets, and can be chosen from the Adjustments panel, but I don’t believe they can be easily exported from one computer and brought into a different one.
So… after a long couple of days of transcription, I have created a set of Macros that apply each of the 38 Photographic Toning gradient maps. Since macros are easy to store, and to share, I am making them available for anyone who wants them.
There are actually 2 sets of Macros included - each as an easy-to-import Category from the Library panel. The first group of Macros is called “Photographic Toning Gradients” and these are named according to the gradient map they apply. Each macro creates a Gradient Map layer, applies the appropriate color values, and names the layer according to the gradient map it applied. The second set of macros is called “Photographic Toning Gradients - reduced.” This category also applies the gradient maps in a similar fashion, but then reduces the opacity of the layer to 30%. This reduces the effect of the Gradient Map, and produces a much subtler effect.
Here is an image that shows the full-strength versions of all 38 gradient maps.

And here is a photo to which I’ve applied a full-strength gradient map, and also a reduced-strength gradient map.

These macro categories are included in the ZIP file attached to this post, along with a Letter sized JPG that includes samples of the gradient maps. Also included in the ZIP file is another macro category that includes one more macro. It’s called “Obama Hope Poster” and, as you might have guessed, it turns any photo into an Obama Hope-style poster. Just because…
Please enjoy them!
Photographic Toning Gradient Maps.zip

Let me give this a shot, although I'll be the first to admit that "color management" is perhaps the most complicated part of computer graphics that I've had to deal with. Getting everything to match up is a tough one, but you can come close if you do the following.
First of all, understand  that the color you see on the screen may not be the color that is saved to your file. Your monitor introduces a "bias" which must be corrected. Much of the time, your computer screen will be (i) too blue; and (ii) too bright. If you use your monitor defaults, your "perfect" colors will be too yellow and too dark, since you will have tried to compensate for that bias introduced by your monitor. This is why calibrating your monitor is important.
Your monitor should aim at a White Point of about 5500K, which is much more "yellow" than its most likely original setting. You should also keep the brightness of your monitor at about 80-120 cd. This is MUCH less bright than the default value your computer shipped with.
So, the first step is to calibrate your monitor. Calibrating your monitor, in simplest terms, means that a file that contains a certain RGB triplet (like "255,0,0" for pure red) is actually displayed as pure red on your screen. You cannot do this by the "eyeball" method. Get yourself a colorimeter, like a Spyder from DataColor, or any of a dozen different other devices. These things, basically, hang in front of your screen and measure the color being shown by software that knows just what color it should be. Calibration software will ultimately create an ICC Profile for your monitor, based on the measurements it makes.
Set your "Display Profile" to use that ICC Profile. On a Mac, you can set this in the "Displays" part of the "System Preferences" application. On Windows, I believe it's found in the "Color Management" app (although someone who is more Windows-fluent can correct me any time).
In Affinity Photo, Designer, or Publisher, you should set the Color Space to be used by your documents. Do NOT set this to your Monitor Profile. This is a common mistake, and it will not get you the results you are after. Your files should almost always be done in RGB (not in CMYK, even though you will eventually print them) and you should use a color space like "sRGB" or "Adobe RGB (1998)". I've included a screen shot from the Preferences part of Affinity Designer below.

Now, the tricky part. If you've calibrated your monitor, then your computer file contains the correct colors. But, when you send that file to your printer you have to be sure that you compensate for any bias that your printer may have (and also compensate for any weirdness introduced by the coloration of the paper you use). This is where Printer Profiles come in.
I have not found any printer profiles for your Epson EcoTank printer online, but I only looked for about 5 minutes. It's possible that you can find them. If not, getting them custom created is the other choice. You would need to print a test sheet that contains some known colors, and compare that sheet with the colors it should contain. Once again, an ICC Profile is created that is specific for your Printer and your Paper, and which corrects any color bias that your printer introduces. When you print, you need to specify that your printer should use THAT particular printer profile when it interprets the colors your computer sends over. Usually, that is done in the dialog box that opens after you choose "Print..."
And, even after you do EVERYTHING right, your printed image will never match your computer screen perfectly. If nothing else, your monitor uses light and your printer uses ink. Light is additive, and ink is subtractive. They will never be exact matches, and your printed image will almost always tend to be a bit darker that your computer image. The goal of color management is to come as close as you can.
Best of luck! It is, admittedly, a struggle…

For completeness sake, I'll attach a macro I've used on occasion to do this. Like the macro above, it creates two Curves adjustment layers, one for dodging and one for burning. Three differences here: (i) I have not attached separate Mask layers, since this is redundant – the Curves adjustment layers already have a built in mask, and I don't see where anything is gained by duplicating this; (ii) I have set the Blend Mode of each of the Curves layers to Luminosity – this keeps them from affecting saturation and hue, and limits them to changing the lightness or darkness of the areas being affected; and (iii) I have changed the Blend Options on each of the layers, so that (at least in theory) the Burn adjustment layer should preferentially affect the darker areas, and the Dodge adjustment layer should preferentially affect the lighter areas – this is the technique one might use for dodging and burning portraits, for instance.
One other thing: when using any form of dodge and burn, I typically use a brush that (i) has a soft edge, so 0% hardness; and (ii) has 100% opacity but only about 1-2% Flow. This way, I know I can eventually build up to a 100% effect, but I only do so very slowly because of the low flow rate.
NB - this is an .afmacro file, so import it through the Macro panel, NOT through the Library panel.
Dodge & Burn with Curves.afmacro

You’ve split your image into a High Frequency and a Low Frequency layer. So far, so good. You’ve applied a black mask to the low frequency layer, which will make it completely transparent. This is as it should be.
The reason that your screen shows complete transparency is that your high frequency layer is turned off. Click the checkbox next to the layer (in the Layers panel) and you should see the High Frequency layer instead of a blank screen.

Sorry, but no one seems to have mentioned the (to me, at least) most obvious answer. Apply a blur using a Live Filter layer, invert the live filter layer (which makes its embedded mask completely black), then paint the blur effect back in where it's wanted. This is completely non-destructive, can be made as strong or as weak as one likes, can be turned on and off at will, can be modified with opacity and/or blend modes, and (best of all) it's so damned easy!

Blurring a Face.mp4

Sorry, but no one seems to have mentioned the (to me, at least) most obvious answer. Apply a blur using a Live Filter layer, invert the live filter layer (which makes its embedded mask completely black), then paint the blur effect back in where it's wanted. This is completely non-destructive, can be made as strong or as weak as one likes, can be turned on and off at will, can be modified with opacity and/or blend modes, and (best of all) it's so damned easy!

Blurring a Face.mp4

Sorry, but no one seems to have mentioned the (to me, at least) most obvious answer. Apply a blur using a Live Filter layer, invert the live filter layer (which makes its embedded mask completely black), then paint the blur effect back in where it's wanted. This is completely non-destructive, can be made as strong or as weak as one likes, can be turned on and off at will, can be modified with opacity and/or blend modes, and (best of all) it's so damned easy!

Blurring a Face.mp4

Somebody make it stop!

Mesh Warp is another possibility. Start in "Source Mode" and map the mesh to the existing light beam. Switching to "Destination Mode" can straighten the line.

Mesh Warp is another possibility. Start in "Source Mode" and map the mesh to the existing light beam. Switching to "Destination Mode" can straighten the line.

Mesh Warp is another possibility. Start in "Source Mode" and map the mesh to the existing light beam. Switching to "Destination Mode" can straighten the line.

I will leave the "meaning" of the story to the reader. I can tell you that, although I have not read a great many of Phillip Dick's works, I have always enjoyed his themes. Chief among them, how do we know that we are real, and are we defined by our own memories (two of the questions in those stories that became "Blade Runner" and "Total Recall")?
To me, tracking is just "plural" kerning – instead of changing the space between two letters, this is the wholesale changing of space between larger tracts of text.
As to the questions about font choice and negative tracking, these were based on my subjective opinion of the "look" of the typeface. I found Perpetua to be rather formal and it seemed to fit the story, which had a philosophical bent. The negative tracking was applied because, without it, the text just appeared too "airy."

Visions of sugarplums danced in my head. Merry Christmas to all (and to all, a good night).

NB - for those who want to know these things, the font is Perpetua 21 points with -10% tracking.

You're absolute right, MikeW. The font is Warnock Pro.

Or, to be completely non-destructive, put the “outlines” layer above the posterized photo layer, then set the Blend Mode of the outlines layer to Multiply. This will get rid of the white entirely without erasing anything.