sfriedberg

Seriously? You know the stuff that Vector VonDoom does with beautifully shaded objects, with dozens and dozens of nested object curves each filled with a flat color? In Affinity, he's got to do that by hand. In CorelDRAW or Xara or other app with a strong blend tool, you can just draw the "key" contours, and have the tool do interpolation of both shape and color. In fact, you can use the blend tool to generate some intermediate contours, then manually tweak them for node position and color for a better appearance and apply the blend tool again between adjacent contours to generate shading that isn't just a uniform progression from one boundary to another. Suppose you've drawn a cell-shaded human face. Now you've been asked to add value shading for highlights and shadows. With a good blend tool, this is pretty straightforward to do. If necessary, divide your flat-shaded cells to define lines of mid-value, draw closed curves for the high and low value areas, assign reasonable colors, and blend between the enclosing mid-value objects and the highlight/shadow objects. And morphing goes way, way beyond turning squares into circles. Strong blend tools aren't limited to curves with the same number of points or even the same number of closed curves. So you can generate a gradient between, for example, an enclosing circle or square and some arbitrarily complicated shape like a griffon, human figure, or snowflake. This lets you do things you cannot do with the basic built-in collection of linear, radial, rectangular and cylindrical gradients.

My guess is that exporting as PNG is rendering the strokes as antialiased curves, while exporting as SVG is simply recording the stroke width in the output file for the display device to render later. So, if you want the thicker lines in the SVG, try increasing the width of strokes in the Designer file. This is a really excellent example of why Designer should have graphic styles that behave like text styles, instead of like presets. Retroactively changing the stroke width on dozen, hundreds, thousands of curves is a painful exercise that is unnecessary in many other vector editing programs. Just change the stroke width property in the graphic style and all the curves with that style applied get updated automatically.

A long, long time ago, I was an enthusiastic user of "channel operations" to build, manually, effects like drop shadows, embossing, and lots of things that today are expected to be included in the list of built-in effects. This is tremendously easier to do when you can freely switch back and forth between treating a layer as grey-scale pixels and treating it as a mask.

Hairline width is a device-specific property. A "hairline" stroke is to be rendered on the device with the minimum feasible width. That is why hairline strokes, in applications that support them, always display at the same width regardless of zoom level. A hairline stroke on my laser printers is not visible, because there's too little toner deposited to fuse properly. Cutters (laser, knife, whatever) will render a hairline stroke as a single cut line. The Affinity suite doesn't support "hairline" as a width. It would be nice if it did, but then you get the question of it's exported (in this situation to SVG). Sometimes an explicit zero stroke width is appropriate. Sometimes it's not. No one specific number will be correct for all ultimate output devices. For SVG, you need to add the "vector-effect: non-scaling-stroke" property, but you'll still have to supply a width as "stroke-width: 1px".

IIRC, CorelDRAW 5 is the last version where Corel published the details of the .cdr format. Features added in the last couple of decades (!) won't be recognized by apps that read CD5 files. On the other hand, if your version of CD will "save as" CD5. it will quite possibly render things down to the best approximation that can be made with the CD5 feature set. CorelDRAW 2024 will "save as" only as far back as version 15.0 (roughly 2013). CorelDRAW 2018 will "save as" only as far back as version 11.0 (roughly 2009).

I am a longtime CorelDRAW user and the CD blend tool is extremely powerful. I'd like to see Affinity Designer have a comparable blend tool, and enhance (and bullet proof) the contour tool to the level of CD.

When using the lasso to tweak your selection, turn on Antialias and/or set Feather to something like 1 px, 2 px or 3 px, instead of zero. Or, turn on Quick Mask and use the paint tools, setting a low Hardness on the tools.

I'll repeat a comment I made in another recent thread: I really wish Affinity had adopted the direction of directing side/foot/end/bib-notes into a stream ("story") and let the streams be assigned to user-specified linked text frames together with some selection of pin/float controls that related the reference marker position to the reference body position (in different text frames). The latter part would seem to be the harder part!

So basically, if you make a selection "off the grid", you want automatic feathering of the affected area based on the proportion of the selection that falls within a given boundary pixel? If so, you can probably approximate this yourself using masks while waiting for Serif to act on a feature request. And you should make that feature request. Are you doing pixel art or other super-blocky style? If not, I wonder if a simple smoothing operation at the boundary would serve as well as this more precise auto-feathering. If you do expect people to see the individual pixels, smoothing would look somewhat different, I am sure.

If Designer had more powerful blend groups (thinking of CorelDRAW style blends which interpolate both shape and color between arbitrary vector shapes, to any desired number of intermediate steps), you could get the effect of warping a gradient by drawing a few overlapped shapes of distinct color and blending between "adjacent" shapes. If you want to tweak the warp, just delete the blend groups (leaving the original shapes), and edit the shapes so their outlines better define the desired warp, then put the blend groups back on. I don't know how Vector VonDoom has the patience to do his vector-based shading without an industrial strength blend tool.

I would also like to see MathML support. I would prefer that Affinity spend its time on a more general tagged text replacement plugin system, and build equations, bibliographic reference processing, and lots of apparently unrelated things on top of that text replacement system. In fact, they should publish the plugin API and let the user community build the specific plugins. Such a system needs to take "tagged marked up" text of some variety, pass it through the plugin appropriate for the tag, and then replace the tagged marked up text with the output of the plugin, for the purposes of flow and typesetting. The tagged marked up text remains in the Affinity document, and possibly will be replaced with different output in the future. That's the view from 50,000 feet. Lots of details to work out, and you probably don't want a raw text replacement, but rather something based on a well-defined document object model (DOM) so you can insert images and multiline equation display blocks and something that supports fully styled text. That means Affinity would have to open the kimono at least slightly in defining the public API. Furthermore, to support complex 2D equation display blocks "natively", you need access to a lot more than character styling controls. You can do it with CSS controls, but in general you need nested 2D regions with inherited styling. That would have to be exposed through the DOM. The alternative is for the plugin to generate an SVG file (or something similar) and insert the image as the replacement "text", which is how I currently get my non-inline equations into Publisher.

I really wish Serif had structured notes a bit differently. I would have preferred a notes stream ("story" if you prefer), which could be assigned to any user generated (set of linked) text frames. If that had been accompanied by some pin/float references between notes and their reference marks, the result would have been more flexible. Actually, I would have preferred multiple, and more general, notes streams, because there are times I would like to use both end notes (text) and bibliographic references, and it seems pretty obvious to use the same basic mechanism for both, end note tags generating one stream, bib ref tags generating another, and both being mapped into appropriate series of linked text frames at the end of the section.

Some of us have seen the evolution (including ingestion of 3rd party technology) of other companies' graphics suites, such as CorelDRAW and PhotoPaint (current relabeled Corel Graphics Suite). I first used CorelDRAW 3. Not 13, not X3, not 23. Three. In comparison to the current CGS it was a kludgy, ugly, inconvenient, limited and annoying piece of software. The CorelDRAW 5 release was a tremendous step forward. Over the years, it has evolved to the point where I don't even bother looking at new releases more than every 4 or 5 years, because it's stable, reasonably comprehensive and I don't care for UI tweaks just to accomodate the latest trend (flat buttons, dark mode, disappearing scrollbars, no thank you). A user of CorelDRAW 3 would be justified in bitching about missing features, counter-intuitive UI design, awkward controls, and bugs. But if they predicted that CD would never be usable, and no professional would ever consider it in the future, they would be quite incorrect. Despite Corel's persistent corporate mismanagement, they manage to maintain some perfectly usable software. They brought in technology from Xara, which added a lot of functionality (and, I believe, considerable numeric stability). There are aspects of CorelDRAW I consider superior to the industry reference (and 800lb gorilla) Illustrator. They also completely abandoned Corel (formerly Ventura, formerly GEM) Publisher, which is what drew me to the Affinity suite in the first place. Some good decisions, some bad decisions. Everybody has their own list of "must have" features. Release 1 of the Affinity Suite didn't have some of the things on my list. That did not make it useless, because not every job requires every tool. But it did mean I had to fall back on other SW to do some of my work. Release 2 has filled in some of those missing pieces, but some of them are still quite rocky and I'd hope for continued improvement. There are still things missing, but I can do more of my work in the Affinity suite, and I expect this progress to continue. Meliora spero, it's not for you to decide that Affinity is dead. That's a decision for the marketplace. And it's not going to be decided by Affinity release 2 any more than the life or death of CGS was decided by CorelDRAW release 3. Furthermore, you are not accomplishing anything by repetitious venting about how you think Affinity is dead. You've made your opinion crystal clear. While my "must have" list is undoubtedly different from your, the Affinity suite is missing some essential features. Unless your principle is "misery loves company" and your object is to make everyone else miserable, consider your point well made and drop it.

The community can't help you sort this out without access to some or all of the files you're working on.

The Affinity Suite does not (yet?) support true vector pattern fills. It can, however, place SVG files as "images" without rasterizing quite nicely. It's my preferred method for bringing in artboard from Designer into Publisher (and others prefer other methods).