Dybkjær

I have this need right now, enough to search the forums for an easier solution. I imported som svg diagrams with explanatory text boxes which end up as Art Text. I convert them to 'Text Frame by clicking into the text, select all, copy, create Text Frame, paste. And now I can resize the text boxes with the text reflowing instead of resizing. The bold etc. text properties are not preserved, though. So I have a workaround, it is just not optimal for this workflow. There are other features I would like more (such as custom arrow heads, and dashed lines with more than two sets), and with diagrams born in AD I create the boxes as Text Frames from the beginning, so I would put priority at low. I am using iPad with AD 1.8.3.2 (I know this thread occurred in another version).

Prompted by inputs from Dwright and DM1 I tried more things: Preview looks fine, print does not, from iPad. Preview and print look fine from mac (OS X 10.13.6). Printer Canon Pixma iP7250 has firmware 1.101. Newest firmware update is older, 1.070 (https://www.canon.dk/support/consumer_products/products/printers/inkjet/pixma_ip_series/pixma_ip7250.html?type=firmware). The app Canon print inkjet/SELPHY has the same problem. The Printer Pro app has the same problem. I tested with each pdf export preset (actual prints, not preview as that is unreliable): export: Wrong flatten: The side with a white rectangle at the bottom is OK, the other side is wrong. Of course, flatten is a bitmap, so now text is blurry. print: Wrong web: Wrong x1a: OK Nice, but X-1a is n old pdf format. I do not believe I can expect a book printer to accept this format. x3: The partially opaque snowflakes and the gaussian blur are OK, the fully opaque flakes are wrong. x4: The gaussian blur and one partially opaque snowflake are OK, the other three are wrong. I have no idea why there is this difference. Pairwise the snowflakes have the same properties. Even if I extend the white rectangle out under all the objects, the result is the same. The PDF/X format seems to state that the colour profile must be CMYK. I cannot see how this is expressed in the preset. I know that my printing company requires PDF with an RGB colour profile, so I am not sure they would like to get CMYK.

Hi DM1, yes, the preview looks fine on my iPad, too. Which just means that the preview and the printer rendering are produced differently. Note: I still wonder why Apple has not included a proper print dialog on the iPad, with all the detailed choices present on OS X and Windows.

I printed the above directly from the iPad (12.9" 2918, iOS 13.3). Thanks a lot for testing the print. That led me to test it on my wife's Windows computer, and on a 4-year old macbook (OS 10.13.6). Both places it prints fine. Now I "only" need to figure out if it can be printed directly from the iPad. It is inconvenient having to send it to another computer. Also I will at times send a PDF to other people and I have previously experienced problems there (pre-Affinity Designer). Determining problem mostly is with the iPad or also partially with he PDF, would be useful. Best, Hans PS sorry, forgot the printer model: Canon Pixma 7250

It is a well-known flaw of PDF that it does not print what you see if transparency is involved. As I am using transparency in my drawings with AD, my question is if there are workarounds? I would like this workflow: 1) Make a drawing in AD with multiple art boards. 2) Export that as PDF (for print, web, ...). 3) Print it, or send it to a printing company. Notes: Please, do not say "Publisher", my computer is an iPad. My book printer only accepts PDF (in RGB only - but that is not the issue here). What happens is illustrated in the two images below. The original TransparencyTest.afdesign as well as the generated PDF (for print) are also enclosed. The dark image is the printed one. The other is a screen shot of the same PDF file and looks as intended. Clearly, PDF is not wysiwyg. We look at this, back to front: The right side has an opaque, white rectangle in the very background. To make sure the problem is not a black default background. We see that is not the case, the right and left sides of the image look the same. The blue background is a gaussian blurred rectangle. Clearly PDF print handles the print wrongly. The 4 smaller snowflakes are embedded PNG which are transparent outside the actual snowflakes. The layer of the two light grey ones are partially opaque. The dark grey ones are fully opaque. In front, white vector drawings and a bit of text. They look as they should. Apparently, transparency through partial opaqueness works fine. The faulty darkness towards the boarder of the semi-opaque ones seems to be the wrong blur-darkness shining through. Transparency from Gaussian blur fails. Is there a viable workaround? Transparency of PNG fails. Is there a viable workaround? Before you suggest that the printer or the driver are at fault, please try to print it on your own printer. Ι have tried to print from different computers and to different printers, all fails. Best regards, Hans TransparencyTest.pdf TransparencyTest.afdesign

I just tested that this is still an issue in AP. I found this because I have the same issue in Affinity Designer. I have placed a cross-reference issue for this here: Regards, Hans

When I place an image in Affinity Designer on iPad, and select a photo from iPad Photos, it gives me the original version, not the displayed, current version. I note that this is raised as an error in Affinity Photos in April 2018. A quick test shows that is still an issue. I place the new topic here, because it is also an issue in AD. The original question suggests the option of being able to choose between the original and the current version. That is fine, even if it is more useful in Photo than in Designer. To help understand why I would like the current version: I might have made edits I do not want to do again, or cannot do, in Designer. While diagramming origami, a common technique for 3D steps is to take a quick photo, crop/skew/etc. in iPad Photos (which is very good for doing that quickly), place the image in AD, use it for drawing/shadowing support, and then throw it out again. There are workarounds, obviously, but they tend to be slower. Like cropping (place inside) in AD (which still loses any other edits), or saving the image to iCloud and place it from there (which requires extra steps, and creates yet another file I need to remember to throw away). PS edit: Found the efficient flow for the second point: Just share as "Copy" in Photos, and then past in AD. Simple.

Just want to supplement: In dark rooms the contrast does not magically become great with the dark UI. It is still bad, meaning it is difficult to read the dark icons and to discern the dark symbols I define with the symbols studio. The one useful thing of the dark UI in dark rooms is the low contrast between the room an the screen. But what you are really after here, is the brightness of the screen. Apple has of course thought of that and built it in. Any iPad has a brightness regulator, easily pulled down from the upper, right corner. Lower the brightness, and the screen is pleasant to look at in dark rooms, and in good UI-programs (aka light themed) the contrast is still good with the icons, i.e. they are easy to read. So for readability (aka good UI) use light theme, and regulate brightness to match the room light. Otherwise I fully agree with you, including the idea to easily switch between light/dark themes. I might select between lighter themes, and some people with sharp eyes (mostly young, I guess) will love the darker themes.

Thanks (and sorry for the late reply). One remaining problem is that the check/uncheck status of "Check spelling while typing" is no remembered when re-opening the document. Another problem is that it does not follow the keyboard setting, so it seems I cannot get Danish or German spell checking, e.g. Only English.

I am mainly illustrating origami diagrams, a relatively geometric discipline. A common operation is to fold the paper, or a flap, into half. This translates into dividing an edge (a straight line between two nodes) by adding a new node between the end point nodes of the edge. Could we get a feature in AD (iPad) supporting this? Could be: Select a segment (an edge). Press Add point (new operation, places the new point in the middle). This is how InkScape does it. Oe better: Select n nodes. Press Distribute (new operation). This way might be more in line with the iPad AD way, and much more general. Please note I am not asking how to construct the midpoint of an edge and place a node there. I know how to do this, several ways. I am asking for an operation that directly supports something I do often, and which might be of general interest. There is some partial support when the edges are vertical or horizontal (when holding one finger down while dragging the node with the pen), but not for anything else. My apologies if I have overlooked an existing operation on iPad AD.

Same here. It is very straining to have dark studios. Assets and styles are almost invisible on the dark background. The studios (and tools tray) should be white - I seldom, if ever, have white styles or assets, and the icons would be much easier to distinguish.

Good idea. However, I cannot get that to work. I do this: Create text Hi, make it italic underlined Hi. Add symbol from selection. Pull that symbol in twice to have now three objects Hi Hi Hi. They are linked, if I make them bold and edit the text, they all react: Hi! Hi! Hi! Now uncheck Sync symbols. Edit one works fine: Hi! Hihi Hi! However, check Sync symbols again and edit style again to remove underline, only the unaltered instances react: Hi! Hihi Hi! Am I doing something wrong, or does it not work on iPad? My iPad? Anywhere? An aside spawned again by this exercise: The UI is dark. The text symbols in the Symbols tray are black. They are literally invisible - an annoying UX error. The same goes for Assets. Most of my assets are dark and hence near impossible to see. To be honest, all the dark grey icons on the really dark background are difficult to distinguish. A light themed UI would be a huge UX improvement.

Yes, works like a charm. Except that when you edit the text, the text also changes accordingly in all the symbol instances. So close, but no cigar. I guess Serif has to design carefully before they add this functionality, but I hope they do. Maybe they can move a module from Publisher to Designer (or release Publisher for iPad, that would be great).

A much simpler workaround: Group the offending object. Grouping makes a regular box, and these have no problems. Grouping the object with itself works, too. Still makes a permanent, regular box. Even if it is a bit large as it has to bound the irregular, rotated original bounding box.

Hi Sean, I have a new observation that might help resolving this irksome problem.. With reference to the demo1,afdesign enclosed in my previous post of July 22: Some of the transformed line segments lose their boundary in one direction. Click on the problem object, note the orientation of the blue bounding box with the resize handles. In one direction the object is deselected if you click outside the box, in the other direction it is selected, however far away you click: if you are inside the "lane" of the bounding box in that direction, it will be selected if you click (unless another problem object lane also overlaps the click point, in which case that may or may not have higher priority). A guess is that it is related to the "reset bounding box" which still misses a permanent reset (see link to discussion below). So a workaround is this: Select the line segment. Copy. Pen tool. Select blue end of line. Make a node or two outside the line. Close the curve by clicking the red end. Draw a solid box covering the closed curve. Select the solid box and the closed curve. Intersect. Node tool. Delete the nodes outside the original line. Paste style. Now the line satisfies two great properties: a) selection works as expected (the error has gone)., b) The bounding box is permanently regular (at least until I start rotating it again). The (b) is a byproduct of the above procedure, but cf. the discussion linked below, something some users (and including me) really wish for, as the non-regular bounding box is confusing and cumbersome. Even if there are solid technical reasons and use cases where the rotated bounding box is what you want (and I don't doubt that), as a user I have not noticed them yet. But that is a digression. The main point here is that getting rid of the non-regular bounding box solved the problem. It may be because there is an error in the non-regular bounding boxes, or because the procedure incidentally resets some other wrong property, or something else; I hope you can resolve that. As a software developer myself, I really click with Bens so right comment at the end of the other discussion: users have little chance to guess what is the cause of errors and what is trivial ot not (indeed that is often the case even if the user is a developer of the product...) Best regards Hans