JET_Affinity

Levi, Just FYI, "macros" is the generic term for a feature which lets you record and play back a sequence of a program's built-in commands.. Adobe has a (rather annoying) habit of confusingly "branding" terms which are common features in other programs. Actions" is just Adobe's label for a macro feature. Hopefully, when Serif gets time to develop a macro feature for Affinity Designer, it will be something worth waiting for; not merely an Adobe-esque record/playback feature, but one integrated with scripting, more akin to CorelDraw's macro feature (which uses VBA for scripting). I'm hoping for a combination of a straightforward macro capability integrated with JavaScript based scripting. This would actually be better than Illustrator's implementation (at least as of CS6), in this way: Adobe Illustrator provides both a (frankly rather weak) macro feature (Actions) and a scripting implementation (AppleScript, VBA, JavaScript). But--and it's a huge "but" in terms of functional power--as of CS6 the macro and scripting do not reliably interact. (It's been a bug long reported since the second year of the scripting implementation.) JavaScript is cross-platform, and the time invested learning it for an application-specific implementation pays dividends in being well along the way toward leveraging the same familiarity for web applications and SVG. JET

Frankly, I'd rather petition Lazy Nezumi Pro to flesh out its cursor following Rulers functions, and have the same tactile interface working in most drawing programs. If they could just add user-specified incremental snaps to the Rulers (including elliptical rulers) we'd at long last have a close approximation in software of what it is like drawing on a quality drafting machine. JET

There was (is?) Swift3D, and similar vector-based rendering in a few 3D modeling programs (Infini-D was one). But most of them rendered shading as polygonal facets, not as path shapes which are reasonable to manipulate after creation. So (just like autotracing results), the very advantage of its being vector-based is largely rendered moot; you effectively just end up with a vector version of "resolution dependence." I know it's just business, but Dimensions is one of those examples of what I mean when I say that Adobe's market dominance has effectively been a setback to vector drawing advancement. Same as discontinuance of FreeHand. ;-) JET

B13eL, Illustrator itself is also really an entirely 2D program. If perchance by "extrude function and distort like in Illutstrator" you are referring to Illustrator's 3D Effect, you may not be aware that 3D Effect is a plug-in which is a severely debilitated subset of Adobe Dimensions, a separate product which was an actual 3D surface modeling program, the unique advantage of which was that it rendered its results as normal 2D PostScript compatible objects (paths with grad fills, path blends, etc.) I have no way of knowing, but I've long suspected that even Adobe Dimensions was based on another acquisition of a startup program, Satellite 3D, which appeared a year or two before Dimensions. (I actually have an original-package copy of it.) Dimensions was discontinued by Adobe long before the limited pieces of it were re-packaged as the 3D Effect plug-in for version 11 (CS). At the time, many users familiar with Dimensions (myself included) would have much preferred for Adobe to simply bundle Dimensions with Illustrator, so that its full functionality would be available. 3D Effect has some additional capability in that, because it is implemented as a so-called "live effect" (which simply means it re-renders on-the-fly when you make changes to the underlying paths), its results can be used a key objects in object blends. But its performance and stability suffer quickly as you push those limits. 3D Effect's limitations compared to Dimensions are huge, though. You can't, for example, even perform different extrusions or revolves on two separate objects in the same 3D model space. And unlike Dimensions, it doesn't even include the third basic 3D construct of pipeline extrusion. Nor can it really do Boolean operations on separate objects (only what can be achieved from using compound paths in the source 2D artwork. So while even a crayon is capable of nice things in the right hands, 3D Effect is not what I'd call a "professional" implementation of a 3D feature. Serif has repeatedly stated that Affinity's focus is toward creating a seriously purpose-built, efficient 2D program with a functionally elegant interface and without the "everything-including-the-kitchen-sink" clutter of most competing programs--ostensibly aimed at professional users. My vote would be for Serif to doggedly stick to that. Serif doesn't (so far as I know) own an old 3D modeling program to cannibalize and insert a few of its pieces into Affinity Designer. Nor would I want it to. When I need 3D rendering, I'll use a separate program. Several drawing programs (Xara, etc.) include similarly-limited 3D extrusion effect features for the typical "Superman" headline and such, but most render as rasters, which can produce more convincing lighting and shading effects, and which nowadays is fast and is typically implemented as a "live" rendering, so its resolution when exported can be adjusted as needed. But frankly, (just like autotracing), that kind of "me, too" feature is something I really hope Serif avoids in Affinity. The last thing the vector commercial illustration market needs is yet another grab-bag of half-baked features. Now, if what you're talking about is a truly thorough 2D extrusion, lathe, and pipeline implementation, yeah, I'd be all for that, but I suspect that would be a way too ambitious project. I'd much rather see Serif improve its axonometric drawing features, which is an area too long overlooked by competing programs. Commercial illustrator's should not have to resort to CAD programs to do axonometric drawing. JET

As if characterizing anyone who disagrees with you as having their "heads in the sand" isn't? Then by what logic do you think posting this thread would suddenly cause Serif to "wake up and go all out"? Harpy, I could build a very long list of features which Adobe was years, if not decades, behind its competitors in delivering. Illustrator, for example, is over 30 years old and still doesn't provide dimension tools, connector lines, user-defined ruler scales... You're expecting Serif to have matched, feature-by-feature, in 4 years what Photoshop took 29 years (after having initially been acquired by Adobe from the Knoll brothers) to become? How long will it take? When will it happen? Nobody knows. Not even Serif. Your blustering won't make it happen. Unless you're privy to a full accounting of its resources, you have no idea whether Serif is already "going all out" or not. What I do know is that the Affinity applications can already serve for a lot of what I do, and I've never considered graphics software an either-or proposition anyway. There's nothing saying I'm required to use just one particular raster, vector, or page assembly program, and I don't. It's how I've always avoided mission-critical dependency on a single vendor. But that's just me. If you can't make productive use of any raster program until it matches Photoshop tit-for-tat, well, don't. Good things take time. JET

No real argument with that, Arthur. Panels is fine. So is Tabs. I'm kind of old-school in that, in the early days, palettes was the defacto generic term (in the sense that a color palette stores a selection of colors; a stroke palette stores a selection of pen nibs, etc.) I just don't care for needless software-specific "branding" of long established and commonly understood constructs, just to make them sound unique."Studio" is very Adobe-esque in that regard. ("LiveTrace" indeed. It's yet another glorified amateurish autotrace command.) "Studio" implies a whole environment, like a room, so it has to be joined with "Tabs" to at least give the user a clue. And what the heck is a "persona," in a graphics metaphor anyway? "Studio" would be more appropriate for that. But I'm not really sweating the small stuff. ;-) Nor do I have a dog in this fight. (Well, not a big one, anyway. Maybe a miniature Schnauzer.) I was just mainly pointing out that the Window menu is pretty vacant, and it's a commonly used place for listing control palettes..uh...panels. So that's an already available place for those anguishing over the tedium of digging under the over-populated View menu. Frankly, I couldn't be productive without organizing controls over a second monitor anyway, an engrained habit since the days when there actually were significant functional advantages to Mac OS. ;-) JET

Surrounding your artwork-in-progress with darkness (or garish color) has been recognized as bad practice since long before graphics computers. It's why the prepress color houses went to great lengths to provide neutral gray environments illuminated by color-balanced lighting for client proof-approval of color-critical work. Human vision seeks contrast and even exaggerates it where it is low. Dark software interface backgrounds cause your artwork to look more "brilliant" and higher-contrast than it really is, which is the last thing you want, especially if designing for reflective print. It's why it's also bad practice to work in low light rooms. The problem is bad enough given that monitors glow. That problem is just exacerbated by working with the lights off (or dim) or by surrounding your artwork with a black interface. And it is not easier on the eyes. It contributes to eye strain; the same reason the non-backlit grayscale reflective Nooks and Kindles were more comfortable for reading a novel. All the self-proclaimed "highly visually sensitive and color-discerning" graphic artist who works in the moody dark (while, ironically, sweating blood over all the hair-splitting nuances of every color management setting) has to do to see the comparatively blunderbuss principle in action is: 1. In your low-light working environment, launch your favorite graphics program. Fill the page with a rectangle and color it with the darkest "rich black" you can muster. Looks really black, doesn't it? 2. Now turn on the lights, shut down the computer and ask yourself: How dark is the monitor? Maybe 80% grey? How, then, can that same monitor be showing you a darker black while turned on and glowing than it actually is when it's turned off and not glowing at all? It can't and it's not. Your eyes are playing tricks on you even in proper lighting and you're just making the problem worse with dark interface surroundings. What's one of the hallmarks of stunning professional-quality graphics and dead giveaways of amateurish work? The range of detail in the shadow areas. View RGB (which is what your eyes are seeing on your monitor, even when the document is in CMYK mode and you've sweat a gallon over color management settings) in dim light or with blackened surroundings in the interface, and your eyes will see more detail in the shadow areas of your artwork than will be there in the final. Then you'll look at the printed (worse, published) results and think the printer mucked up your masterpiece. The same principle is at the root of why beginners so often think a red graphic on a black background makes a great logo. No, this "blacked out" interface nonsense is just a trendy crowd fad, like dressing up in black leather pirate costumes while riding completely black motorcycles and wondering why drivers don't see you. I'll rejoice to see both fads finally fade away into the curiosity annals of laughably bad ideas. The only place for them is Halloween parties and Ripley's Believe-It-Or-Not museum. JET

Just relocate the Studio flyout list to the Window menu (as in other programs). Rename it Panels or Palettes. JET

Please, no. The auto-join behavior of Illustrator's Pen Tool (which cannot be turned off) is one of the classic examples of its horrible interface design. The Pen Tool is not a selection tool. It has no business altering unselected paths. That interface element violates the fundamental meaning of selection. The behavior can seem like a convenience in certain scenarios. It just as often becomes a stumbling block. It's just as common to need to start or end a separate path at the same location as a pre-existing unselected path. So to facilitate that normal and proper behavior, Adobe had to provide a workaround. You have to: 1. Mousedown somewhere you don't want to start the new path. 2. Press and hold the Spacebar. 3. Drag to where you do want to start the new path, hoping Illustrator's unreliable snap-to-point behavior works. 4. Mouseup. 5. Click elsewhere to continue the new path you just started. That is so convoluted it is just laughable. All that just to momentarily disable the infernal auto-join behavior which shouldn't be affecting unselected paths in the first place. No, the proper way to extend a pre-existing unselected path without breaking stride while drawing with the Pen Tool is to simply press a keyboard modifier (Ctrl in the case of Affinity) to momentarily invoke a selection tool, click the endpoint to which you want to join so it is selected, release the modifier (automatically resuming the Pen) and then click the selected endpoint with the Pen. I can see I'm gonna need a T-shirt: "Illustrator IS NOT the program to emulate." ;-) JET

?? A spunky, relatively tiny competitor methodically and carefully building fresh-from-the-ground-up elegant, powerful, and affordable alternatives to a set of decades-entrenched applications "shows no enthusiasm" just because they're not done yet?! JET

MEB, Any chance that may be in the plans? I use Illustrator's Transform Each command quite frequently. It's another easy opportunity for competitive advantage. It's yet another modal dialog. It provides no proportional lock. It provides H and V scaling, but no explicit width and height. It seems to me an elegant no-nonsense way to provide all that functionality in any program would be to simply provide "transform separately" and "random" checkboxes or buttons in the Transform palette. With the Separate option on, each object in the current selection would be transformed relative to its own coordinates, and the Random checkbox would become available. With the Separate option off, the entire selection would be transformed as a whole and the Random checkbox unavailable. A Scale Strokes on/off would also need to be included. (Thanks for your participation in these forums, by the way.) JET

Among other problems. Mission-critical dependency upon third-party plug-ins is not something I'm interested in. Been there; done that. It's a model that I consider passé. Far more valuable to me is a full-blown user scripting implementation (which, as I recall, has been mentioned in the development plans), preferably JavaScript based. (FreeHand's Graphic Find & Replace has yet to be equaled in any mainstream drawing program, by the way.) JET

1-bit rasters are not just for particular repro methods, either. Colorizing a 1-bit diffusion dither is a great low-overhead way to achieve "grainy" textures in any vector illustration style, regardless of printing method. JET

The treatment of the bounding box skew handles is not weird or undiscoverable. Other drawing programs do it quite similarly (CorelDraw, Canvas, Inkscape, Xara Designer Pro among them). But yes, it's different from Illustrator. No offense, and you're entitled to your opinion about what constitutes "professional" quality software. But I don't equate the term with Illustrator's worst-in-class user interface, and I've been making my living with this stuff since it first appeared in the mid-80s. If this skew thing is a deal-breaker, I can offer you at least one consolation that comes to mind: Consider that Illustrator's interface has long displayed rotated bounding boxes which let you manually scale in the directions of its rotated axes by dragging side handles...but not by numerical input, and only as long as Illustrator "remembers" the object's rotation. And nowhere in the interface does Illustrator provide you access to the angle at which its on-again-off-again rotated bounding boxes are rotated, so that you could, for example, numerically return it to orthogonal orientation, or rotate other objects to the same angle. Affinity's rotated bounds, on the other hand, persist pretty much no matter what you do to the object, and are far more useful. The angle at which the object is rotated from its as-created orientation is displayed right there in the Transform palette whenever the bounding box is visible. And you can perform all of the on-object transformations (including skew) relative to the rotated bounds, not just horizontal and vertical scaling. So no, it's not like Illustrator (thanks, Serif). In this regard, it's more like ACD Canvas, one of the venerable "big four" drawing programs (FreeHand, Illustrator, Canvas, Draw) that date back to the early days. And Canvas blows Illustrator's doors off for CAD-like accuracy and no-nonsense technical drawing in an environment that's as capable as any of the others for general commercial illustration. Which treatment of object rotation do you consider more "professional"? (There are plenty of other consolations, too, by the way.) ;-) JET

Bounding box handles are fine (Affinity's treatment of them is not unlike several other programs), but they don't cut it when you need numerical accuracy to properly align objects. JET

While I agree there is much room for improvement, a word of caution: We must be careful what we ask for, and how. If I were allowed but one thing to suggest to Serif, it would be this: Adobe Illustrator is not the model to emulate for anything pertaining to user interface. One of Illustrator's most fundamental interface faux pas is the pretense that a segment exists independent of the anchor points which define it. Not only is this one of the most chronic stumbling blocks to newcomers, but it is at the root of why Illustrator is so more prone to littering files with accidental "stray points" than any of the other mainstream drawing programs (so prone, in fact, that Adobe had to tellingly add "clean up" commands to find and remove them). Simplest way to see it, in Illustrator: 1. Line Tool: drag to create a single-segment path. Deselect. 2. White Pointer: Click the segment to "select" it. 3. Edit>Cut. The current "selection" is copied to the clipboard, right? 4. Edit>Paste. The pasted content becomes the current selection. The pasted content, of course, is not just a segment, but two joined, complete anchor points, because a segment cannot exist without its defining anchor points. And you now have two invisible "stray points" left behind on the Artboard, because they ostensibly were "not selected" when you invoked the Cut command. The steps above are only to easily demonstrate the issue. In real practice, ramifications of Illustrator's ill-conceived selection interface cascades throughout the program. Getting the selection interface right is foundational. Getting it wrong cannot be "backed out of" later. Yes, there needs to be a way to open an existing segment (which frankly, may be a better term for it). A simple selection in the contextual cursor menu may suffice for that. The Break Curve button should break all selected Nodes, not just one of them (just as the full set of alignment and distribution functions that work on whole objects should also work on any current subselection of Nodes, be they nodes of a single path or any number of paths). If I were allowed two requests of Serif, the second would be: If you need a model to emulate for path creation, selection, and manipulation, emulate FreeHand's as far as you can get away with. I'm not saying FreeHand was perfect, nor that an even better interface is impossible. But I've been doing vector drawing for a living since the mid-80s, and in the intervening decades have maintained licenses to most of the mainstream drawing programs. I have yet to see any vector path drawing and selection interface better than FreeHand's. All path manipulation interfaces foist a measure of smoke-and-mirrors upon the user. To my mind, an interface which doesn't overly isolate the user from the reality of what's going on is preferred, even if it means an occasional extra click or two. For example, what we all nowadays call a "segment" is actually a complete cubic Bezier curve defined by its four coordinate pairs (the "segment" is its two on-curve endpoints, its two off-curve handles, and the plot they describe). Illustrator alludes to this fact by the naming of its Select>Object>Direction Handles command (confusingly dissimilar to its pretense of selecting "just" segments, but is really the same thing. Regardless of all that, Nodes are universally treated as single entities, each having two handles (either retracted or extended), when in fact the two handles are coordinate pairs from the two joined curves. A node is really just a joint between two curves. A joint can be straight (tangent) or bent, locked or bendable. A straight and locked joint would be what we call a curve node. A bendable joint would be what we call a corner node. I've often wondered if a more intuitive Bezier-based path interface couldn't be based on such a metaphor. A joint can also be broken or...disjointed (new metaphor for dragging new joints out of existing ones?) But, of course, I wouldn't presume to suggest Affinity rebuild the foundation of its selection interface in favor of my musings. Whatever future enhancements are built upon the foundation, they need to be conceptually consistent with it. And I can't think of a more inconsistent interface than Illustrator's. JET

It's got a good beat. Easy to dance to. I give it an 80. Maybe we should make it a T-shirt. "Please put the fix in one-dot-six." ;-) JET

Okay. Done. JET

It's rather cumbersome and tedious (especially to newcomers) to have to enter two separate transform values just to skew a selection vertically. From a UI perspective it's conceptually tantamount to providing only a height field and requiring the user to enter 90 or -90 in the rotate field in order to set width. Consider the common scenario of distorting the artwork for the top and sides of a cereal box into an isometric orientation. You've drawn the top, front, and side artwork upright. The front is positioned with its upper left corner at the iso origin. Real quick: How do you vertically skew the artwork of the front of the box about its top left corner without moving it, to serve as the right-side surface of the isometric view? Is that intuitive? I'm all for feature elegance and avoidance of clutter, but the interface for numeric skew needs to provide horizontal and vertical, even if it's just a radio button or keyboard modifier. It would also be worth considering putting the Reset Bounding Box icon in the Transform palette in the blank space under the 9-point proxy, rather than having to go back and forth across the screen for reiterated transformations. I find myself having to keep the Transform palette at the top of the screen just for this reason. Generally, I've never been wild about 9-point proxy icons. How, for example, does one skew a selection about the off-object center of transformation? (The COT, by the way, is called the "Rotation Center" in the button's tooltip--implying it's only for rotations--but moving it is recorded as "Transform Focal Point" in the History palette.) JET

...Which is absurdly cumbersome and tedious. From a UI perspective it's conceptually tantamount to providing only a height field and requiring the user to enter 90 or -90 in the rotate field in order to set width. I'm all for feature elegance and avoidance of clutter, but the interface for numeric skew needs to provide horizontal and vertical, even if it's just a radio button or keyboard modifier. It would also be worth considering putting the Reset Bounding Box icon in the Transform palette in the blank space under the 9-point proxy, rather than having to go back and forth across the screen for reiterated transformations. JET

No offense; just a suggestion: It's bad practice to post ideas as a list of a particular individual's wishes. The purpose of a feature request forum is not just to privately submit one's list of feature desires to the developer, but to expose feature suggestions to fellow users for discussion of their merits. (Otherwise, there would be no need for the forum.) Posting one's ideas in a "my desires" thread obstructs discussion or constructive criticism, because there's no way to tell from the thread title what its multiple subjects are. Everyone is not going to read through everyone else's wishlist in search of a feature they are interested in commenting on. If everyone started a thread of his or her collection of feature requests, there would be no way to tell which ideas have already been suggested. Please post each feature request in its own thread. JET

One of the most useful (and most often overlooked) functions in Illustrator's Pathfinder palette is the awkwardly-named Merge button. When invoked, it does two things to the current selection: Automatically "punches" (Boolean subtracts) intersecting areas of all paths behind each path. Automatically "unions" (Boolean adds) all visually touching or overlapping (contiguous) areas of paths with the same fill. And it does it by actual paths' geometry, not by effectively "auto tracing" the screen display. In a design involving a moderately high number of overlapping path objects, that one command gets a design destined for vinyl cutting equipment "halfway there," saving a lot of tedious and time-consuming individual punch and union operations. The "missing element" typical of many of Illustrator's features and commands is an overlap setting, which would allow the user to specify a distance by which to inset the punched paths (programmatically the same thing as the Offset Path command with a negative value) in order to create registration "traps" to facilitate assembling the individual cut-out vinyl shapes in the real world without ugly sliver voids. As it is, that part still has to be done path-by-path. Most of the relatively small handful of features which actually constitute a legitimate Illustrator advantage, contain similar "almost there but not quite" Achilles' heel omissions that represent golden opportunities for surpassing Illustrator's functionality, rather than just forever playing "me, too." JET

​Understand, having paths effectively partially "mask" others behind them is not necessarily "improper"; and having paths always "kiss" or abut is not always required to be considered "proper." You simply always have to construct your designs with the intended downstream use(s) in mind. ​Generally speaking, embroidery software can either semi-automatically "fill in" the area of a closed path with stitches, or it can stitch a "running stitch" along the path. Generally, the individual paths in the file are individually assigned particular stitch types. For example, a yellow hexagonal closed path may need to have a black border stitched around it. Although in your drawing software, it is routine to apply both a fill and a stroke to a single path, that does not guarantee that the embroidery software which imports your drawing can do that. The embroidery software quite likely needs a separate path for each operation. ​Regarding fills, the embroidery software doesn't automatically limit itself to what are just the visible bounds of a path when viewed in your drawing software or when printed. It's entirely normal for an embroidery stitching operation to be performed on top of a preceding background stitched area. Other times, that's not desired, and you may have to use compound paths so that the stitching of what looks like a "background" leaves voids to be filled in by a subsequent stitching operation of a different color. ​Similar considerations are required when preparing artwork destined to be cut on a vinyl sign cutter. A design which looks and works fine for print commonly has many places where paths partially overlap others behind them. Again, that works fine for print. But sending that same file as-is to a sign shop can end in expensive disaster as the cutting knife simply follows each path in its entirety. There will be wasted vinyl and unnecessary and unwanted cuts across areas which should be cut as one piece. ​Even in print, overlapping is sometimes needed and other times not. Overlapping is analogous to overprinting which is an important element of building traps (chokes and spreads) where shape-following overlaps are necessary to accommodate slight miss-registration on press. That same principle is often needed in cutting sign vinyl; it's usually advantageous to provide a shape-parallel measure of overlap between adjacent dissimilar vinyl colors to make physical assembly possible and avoid gaps as vinyl shrinks over time. ​So again, there is no universal "proper" way to construct your drawings. They have to be constructed appropriately for the intended reproduction method. ​JET

That's great. Thanks. Feel free to take all the time you need. You know...10~15 minutes. ;-) JET

This is good news. At least functional parity of this feature set with DrawPlus is needed. Being able to "send" a live ellipse to any of the three axonometric drawing plane grids (complete with its start and end arc points adjustable) to use as an elliptical axonometric protractor is key for correct construction of off-axis edges. By way of suggestion: One thing that limits practicality of a grids-based interface for axonometric drawing is that the grid increments end up being used as measuring rules, but (unless I'm overlooking it) the user has no means by which to snap the current plane grid to a particular vertex of a shape already drawn on another plane. Instead, the interface just lets the user set an origin for all three grids. It seems to me that a keyboard shortcut which effectively causes the nearest increment of the current grid to shift to where the mouse is clicked would be advantageous. As a long time technical illustrator, I do applaud Serif's willingness to include explicit support for axonometric drawing. I know it must be kind of a "labor of love," since there are so many misconceptions about it. When this feature set is fully fleshed out, it will rival similar functionality (in this regard) in Corel Technical Designer, which is too "vertically priced" beyond the budget of most general commercial illustrators. So this is a great differentiator from other mainstream 2D drawing programs. (That is not a slam against Designer; it's a great program. I suspect much of its cost is no doubt due to other license-based related features, such as importing and flattening 3D CAE system formats. But I remain convinced that most mainstream commercial illustrators could benefit from adding axonometric drawing to their repertoire.) JET