HuniSenpai

I agree with those things you listed. I hope they get added. I was just going off of the things I noticed Designer was missing in the order I noticed them -- that's a pretty good indicator of importance. It'll be somewhat different for everyone, of course. For me, it was shapebuilder, vectorizing, and line weight tool. From what I can tell on the forums, these are some of the most common requests. I don't think Serif will run out of things to add, but I do think they will run out of big things to add. Big things that would justify an upgrade to version 2.0, 3.0, etc. Scripts would be a cool addition, it might be able to let us add missing features to the program until the devs are eventually able to add them properly.

Saying that is like saying "Well, this car can't go in reverse, but there's another car that can. Just hop in the other car any time you need to reverse." The whole point of vecotrizing is to offer a quick way to bring an object into vector without tracing. And, yes, I know it is not perfect quality. While Inkscape is an okay program (first vector program I've used, actually) it is sluggish and a pain to use. Honestly, by the time it takes for Inkscape to open, to vectorize, and to export the vector, I could have already traced many simpler paths by hand. If i'm just making an sbubby, it doesn't need to be perfect. And, to be clear, i'm not saying that aren't minor things that the Affinity suite is missing. There certainly are. But it's the big things that everyone knows about that will convince people to upgrade. As far as I can tell, the big things for Designer are shape builder, vectorizing, and line width tool. They should and probably will save the smaller things for the free updates.

Let's be honest, they're probably saving auto trace, shapebuilder, and line weight tool for version 2. The guys at Serif have proved that they are excellent software developers capable of solving problems and implementing new features relatively quickly. These are the main things keeping Designer from being a near perfect software. If they give us these features now, what else is there to entice us to upgrade to version 2 when it rolls around? The performance is basically perfect, all other tools are there, and it already beats Adobe Illustrator in a few ways. They already have programmed these features in the past with DrawPlus X8, so they certainly have the skills necessary to program them again. I do not blame them for putting off these features. I sometimes grow a little worried about Affinity's business model, and if they are going to stay afloat. With such a low price and free updates, it amazes me that they can pay all of these fantastic developers. Some time along the line, they will have to raise their price and get people to buy their products again. And, if version 2 has these features, I will gladly pay $75-$100 for it, especially if it means keeping Serif afloat. Additionally, I feel confident that they will make my $75-100 well spent. I have no doubt that they will make the auto trace and shapebuilder etc. as good as they can possibly be. Frankly, if I had the money, I would be donating to Serif. Adobe needs competition.

Alright, I was thinking maybe a driver conflict. The FX-6300 came out in 2012, and the GT 1030 came out in 2017. Since the FX series chips don't have integrated graphics, you must have a dedicated graphics card to use the chip. If you purchased the FX-6300 around when it came out, you must have had a different GPU between 2012 and 2017. Have you done a GPU upgrade? If the old drivers for your old GPU were not cleanly removed, there can be driver conflicts. Anyways, since you're not having performance issues, I doubt it's a driver conflict. I'm starting to think this is a bug related to Windows 7. There's a histogram preview thing behind the regular curves adjustment, but not on the one in develop assistant. This might be the source of the bug-- it's one of the few differences I can find between the curves in the develop assistant and the CRL + M curves. If this is the case, I don't have any solution. I'm not a dev.

Yes, you can always use the tones adjustment panel if you want. The "develop persona" can always be accessed at the top left. So, if you need to do a curve adjustment on a layer, you can use that. However, it only affects the layer you have selected. And it's missing the picker tool and LAB adjustments. That panel is basically a watered-down version of the real thing. Do you have any monitor calibration software or anything? Is your graphics card's driver up to date? Have you recently installed a new GPU, meaning that the old drivers might be left over? Is the program laggy when zooming in or panning around? Also, just a random idea, try turning off Windows Aero.

Ah yes, indeed. If you really want to create a mask from it, that is something you can do. As a tip, to select the contents of a layer like that, you don't have to go into the layer menu. Just hold crl and click on the thumbnail of the layer. That will make a selection; you can then add a mask. Then you can go back to the blend modes and press reset, since you don't need them anymore.

I agree, Affinity Photo should improve the "select sampled color tool." It's really not too useful -- it's just a simple magic-wand-like tool with a tolerance slider. There isn't even a black-and-white mask preview, nor any clear way to just mask by value, hue, and/or saturation. It really needs work. However, you don't need to do it the way he described in that video. He also made another video explaining why this isn't always necessary, here: Affinity Photo absolutely has the feature he uses in this video, except Affinity Photos implementation is wayyyy better because it uses a graph instead of those darn sliders. The graph offers far more control. At the start of this clip I recorded, i'm just taking note of the lightness values I want to target. You don't have to do this -- I just like to be precise: Also, just ignore me going into the assistant at the end. I had my setting set to child adjustment layers to selected layer by default. Additionally, I opted to use LAB instead of RGB because it doesn't mess up color saturation when trying to do lightness adjustment -- it often looks way better. You can just stick with RGB if you want. If you mess up (like how I accidentally selected a color that was way too bright, and close to white causing curves adjustment to kind of glitch out) you can delete a node in curves adjustment by right clicking it. And if you want to save some time: crl + u = HSL crl + m = curves crl + L = levels

Glad to help! I'm convinced it is in some way related to the bug I found, it's just not immediately obvious. You know how the donut oddly merged with the parent after I pressed merge? Maybe, in your case, it's merging with something transparent, thus making it disappear. Maybe the mask?

Take a look at the last video I just sent. I think I figured it out.

I think I may have found your bug. The elaborate layer structure you are working on has a whole bunch of layers (looks like a person with head, legs, etc.) masked to a layer. When you apply a childed adjustment to one of those masked layers and press "merge," it does a weird thing where the entire thing merges together, and you get, in my example, the donut and background merged together. Take a look at this clip: I am pretty sure this is a bug. We should probably report this to the devs.

Yeah, that's what's really confusing me. It should be working for you, so long as you didn't induce the bug. What version are you on?

Sir, trust me, I know what you are talking about. My point is that this is working for me, but there is a bug that can cause the merge to not work when the layer is childed. The bug is that merge doesn't work when you child the layer yourself. Watch the video I just posted.

Here's a video of the bug i'm referring to (it's currently processing):

Don't worry, I understand what you are talking about entirely. I'm make a lot of infographics for Amazon listings, and I have to affect just one layer at a time with an adjustment quite a lot. I was just referring to a bug I've noticed, because childing it yourself vs having the program child automatically can actually cause the merge button to stop working. Really odd to be honest. I digress, though. I'm really confused by the behavior you're getting -- as far as I can tell, the merge button should be working, as long as you had the program automatically child the layer for you. Based on all you have said thus far, it sounds like you have the program set up to do this automatic childing, so you're good to go. You aren't childing it yourself, right? Because, based on the bug i've just found, doing it on your own breaks the merge button.

So I found something weird about the program... In the develop assistant, you can either choose to have new adjustments do a new layer (meaning that you affect the global image -- all layers below the adjustment layer) or you can tell it to child the adjustment layer (so that the adjustment, like HSL, only affects the layer you have selected upon pressing CRL + U). That's no what's weird, though. That's just some background information. Let's say you have the first option in the assistant selected: make a new adjustment layer, don't child it. If you child it yourself and then try to press merge, it does nothing! If you choose the other option in the assistant and have the program child the layer on its own, then the merge works correctly (at least for me). Still, I'm not getting the same behavior as you, @Ladlon. By any chance, do you have multiple adjustments being done to a single layer before you press merge, or maybe there is a mask or something? Can you send your afphoto file?

Yes, that's exactly what I showed in my video. The adjustment layer merges down when it's not a child. When you do child it (see second half of video) the merge button no longer works. I think they should add some code to the program to say "if merge is pressed on a child layer, then merge with the parent layer." Instead, it does nothing. And absolutely, it is important to make the adjustment a child when you need to mask it to a certain layer. I do this all the time, and I agree that the merge should tell the program to merge the adjustment to the parent. What confused me is you said that the merge -- when you have a child adjustment selected -- actually does something, i.e. it deletes the content of the parent layer? I don't get that same behavior, hence why I made the video. The later half of the video showcases the behavior I get with child layer merge button.

When you press merge, the adjustment layer looks at the layer below itself and merges with that. When the adjustment is a child, there is no layer that it can look down to. So, when I press "merge" on a child adjustment layer, it does nothing. I'm surprised that, for you, it actually does something. Either way, I feel like it makes sense for "merge" to not work right when the adjustment layer is a child, since it's supposed to merge downward to the layer below itself. Here's a video showing how the behavior is for me: I think they should fix this though. I mean, yeah, technically the merge is supposed to merge down. But somewhere in the code they should write that, if the adjustment is a child and you press merge, it should apply the adjustment to the parent for you.

I mean that would be a cool feature too, I suppose. I'm just asking if it could automatically rasterize 'em when you try to merge multiple vector layers into one. That's how Photoshop does it. I know I use a crazy old version, don't judge me lol. But notice how I can just go and merge the vector layers (the squares) together, and it makes it raster: In Affinity Photo, if I even one vector layer selected it won't let me go through with the merge. I can press crl + shift + e as much as I want and it'll do nothing. I really wish it would just rasterize and merge them. I know what i'm getting myself into: yes, I lose the ability to resize losslessly or change color when things are rasterized. But I also think that the kind of people concerned about layer merging are fully aware of that fact. But if you added a feature to "batch-rasterize" layers, I suppose that would work. The only annoying this is that you'd have two steps instead of one. You would 1) have to batch rasterize and then 2) merge with crl + shift + e.

Yup, I noticed. Regardless of whether it is intended behavior or a bug, I think it should be reverted back to how it used to work. I appreciate predictability in a program. Also, if you have to question whether or not something is a bug, then it's probably not a good feature to add, to be honest. Also, do you think they can let me crl + shift + e merge layers, even if I don't rasterize them first? Can't Affinity Photo rasterize them for me? Look at the video: before I could merge, I had to right click on each layer and rasterize. Sometimes, I have 20 layers I need to merge, and I have to right click and rasterize each one by hand! This is another annoying thing. This has been in Affnity Photo for as long as I can remember.

You know, I never realized there was a merge button lol! But yeah, clearly that's not what i'm referring to since I didn't even know that was a thing. No, i'm referring to crl + shift + e. It doesn't work the same in this latest affinity photo version. Here's an example. Frankly, I think this new "feature," if it is a feature, is more annoying than anything. If I want to change the arrangement, I can just drag the layers around first and then merge. This feature is unnecessary and really confused me for some time after the change was made. If I was a new user, I don't think I would have ever even figured out that the order mattered. P.S. I absolutely, completely and utterly hate how you have to rasterize each layer -- by hand -- before you can merge layers. I really wish you would automatically rasterize them when I merge them together. Sometimes I need to merge a few dozen layers: right clicking and rasterizing each one by hand wastes a lot of time.

Glad to help! One of the odd things is that it merges top to bottom if you select bottom first and top second, but it -- for whatever reason -- merges bottom to top if you do it in opposite order. Bottom to top means that the lower layers take precedent over the upper layers when you merge. So, let's say you have a nice blue sky for a background, and a cartoon drawing for a foreground. If you select the foreground first and then the background and press crl + shift + E, the foreground disappears and you just get the background! The cartoon disappears in this merger. Inversely, if you select the background first and foreground second and do crl + shift + E, you get both the cartoon and the blue sky merged together, just how you'd expect. Really odd and kind of annoying to be honest. I'm frankly not sure what the point of merging bottom to top is-- that just effectively deletes layers above it. Makes me think it's a bug.

Something weird about the latest version of Affinity Photo is that, when merging, it is important the order in which you select the layers that you want to merge. If you want to merge down, then select the lower layer first and upper layer second. Then do crl + shift + e. If you select the upper layer first and lower layer second, then it will merge backwards, and it'll be all weird. I honestly don't know why they decided to do this, maybe it's a bug idk. Anyways, I recommend just masking the adjustment layer to the layer and then rasterizing that layer and its masked adjustment. I show that here: https://youtu.be/MAboIcG0Eos Please note that I have a different setting than you. When I make a new adjustment, I have my Assistant set up so that it makes a new adjustment that affects the image as a whole. It sounds like you've told it to child the adjustment to the layer you have selected. Here's how you can see that setting: In the end, though, i'm not fully sure I understand your question. I get the feeling you're having a problem with the new way layer merging works, i'm not sure though. Let me know if this information answered your question. And yeah, I agree. They should add an option to quickly do a destructive adjustment to a layer you're editing, like crl + shift + u instead of crl + u in order to do a destructive HSL adjustment. This is a great idea.

I agree, a dedicated replace color tool would be really cool. Still, this HSL menu way of replacing colors offers tons of control and doesn't take much longer unless you want it to. Even when showing my steps, it only took a few minutes: https://youtu.be/2E2eonztYn4 There's two benefits that go along with using the HSL tool to do a color change. By doing it this way, you are masking by hue, not by color. This is an important distinction, because the overall color includes hue, saturation, and lightness (or value, depending on what model you want to use). As you can see on your example, the software that you are using masks by overall color. Because of that, it didn't select the shadow on her sweater, and that part is still red. For color replacement, using hue to guide your selection is important, and that's exactly what HSL allows you to do. Additionally, by cranking up saturation or hue, you retain the relatively variability of saturation and hue (unless you clip the saturation). In other words, even after the color change, some pixels are still more saturated than others. Saturation on HSL naturally goes down as things get darker. Take a look on this image, which retains hue and saturation but has only 1 lightness value: So that's the other important thing. The software you are using is forcing the color change. It forces the same hue and saturation across the entire sweater. This is not how things are supposed to be done because, as we can see above, the shadow regions should have a different saturation than the lighter regions. This is another reason why you want to do it through HSL. Still, this is just why the current method is superior quality. As far as speed, I would love to see a feature like you're describing. Note: I didn't make her sweater that blue, although you could if you wanted (just crank up saturation a lot, and luminosity a bit). I just think that it's impossible to get a sweater the blue. If it is possible then I want one

You mean content aware? Affinity Photo already has that -- it's called inpainting, though. Here's an Affinity Photo tutorial on inpainting brush: https://affinity.serif.com/en-us/tutorials/photo/desktop/video/297061442/ If you want to do a content aware fill, then Select what you want to fill (using marquee tools or lasso tools, for example) Press shift + F5 and select inpainting. Press "ok"

Three of my wishes have to do with color picking: I wish the hexadecimal value was always visible in the color picking panel. I have to change to RGB hex every time I want to enter in an exact number. There's no reason why there can't be a hexadecimal field for every panel, except for perhaps the grayscale one. Also, you should add and option for HSV for artists. I personally don't do a lot of digital painting, but I know that the HSV triangular color picker is much preferred over HSL for those doing digital painting. It allows you to easily change the apparent lightness or darkness of a color without messing with the apparent saturation. Here's a video highlighting my point... notice how, in order to maintain the perceived saturation, you have to curve the HSL color picker in a really strange way, which is not easy to do: 3. This reminds me, you should be able to hold alt while in the color chooser window to color pick from the image. I always have to move my mouse over to the color picker icon and click and drag. You can see just how annoying and slow that is in the video. Worse yet, I have to click on the color picker icon to actually go through with it! While in the color picking menu, I should be able to just hold alt while aiming at my image, and then I just left mouse click.