Hadriscus

Time to look for a new home ! Open source alternatives are not where I'd want them to be (otherwise I wouldn't be using Affinity!), but they're building up in strength & versatility. Inkscape has some rough edges but is a very, very capable vector program. Krita is unequalled at painting. Gimp isn't ideal right now but they're nearing a "modern" 3.0 release, let's see how that goes. Graphite.rs is young but veeeery promising. I don't really have a Publisher alternative to offer, though. Farewell ! it's been nice.

Just found this thread and would like to add my voice for better underline control at some point.

Thanks appreciate it

Hi, thanks for your reply. This is exactly right : I live in a world that is, not a world that should be.

Has opacity control be added into AP version 2 ? I am a bit out of the loop. Considering upgrading from 1.10

This is an ideologically loaded statement. I believe reality calls for differenciation. Sure, OS-level scaling is a sane baseline, but different programs achieve different things and need different defaults. I use Blender on a daily basis (which has a masterclass of a UI-scaling feature, btw) and I keep the interface smallish because of the nature of the work that I do in there : many different interweaved tasks, oh so many buttons, and a necessity to display a lot of data at once in several different editors. On the other hand, my usage of AP these days is for texture authoring from photos (make tileable, color correct, that sort of thing), a pretty straightforward usecase for which I wish to have a comfortable, wide UI. For illustration, this is the kind of information density I have to deal with when working with Blender (see attachment). I consider myself lucky Blender does scaling independently of the OS... it's actually a multiplier of the OS base scale value, so it's not exactly independant, but anyway. See the AP interface (other attachment) : there's a lot of wiggle room, and I wouldn't be mad if that entire interface was scaled up a little (even if not fractional). But changing that at the OS level means changing everything else, including programs that are fine as-is (in terms of information density). I understand how this all should work, ideally : developers all over the world ()Windows developers included!) making sure their programs work fine with scaling. But reality is a mess of different UI frameworks, discontinued/EOL programs, and blurry upscalings. I won't be adding to this thread anymore

Because "it's not their fault" if the OS scaling or other applications' handling of it are botched. Presumably, every other program on this Earth should be updated instead, I'll let you judge whether or not it is a realistic goal. I still hope that stance changes at some point, because it seems to be truly ideological in spite of their own customer needs.

Is there any communication available on this topic ? should I post somewhere else for someone in the know to see my question ?

I have to know whether or not this is going to improve in future versions. Is there a roadmap I can look up ? I'm used to Blender's more open development style, and I understand not every developer does that, but at some point there needs to be clarity. What are the plans ? Is this area likely to improve ? Thank you, Hadrien

Hi, thanks for the suggestion! Choosing nearest neighbour does just that, but doesn't appear to change the definition of the images. At full zoom, the preview images that appear in place of the "real" ones are made up of blocks 8px wide, when they should be 1px (because the linked images match the document resolution and are not scaled). I understand this is not the support forum, but I can provide the file if you think there is something worth trying. Thanks ! Hadrien

Hi all, forgive the clickbait title. I use a GTX970 which admittedly is far from new, but canvas compositing is not just sluggish, it doesn't update at all my linked resources when I zoom on the image, even after I've let ten minutes pass. On the other hand any compositing software works smoothly with composites that are much more complex. It seems like it's trying to recalculate everything (including live effects such as blur) every time I change the zoom level, every time I move a layer, basically any action causes a full recalculation. Is there even any caching going on? any plans to move away from OpenCL? (which is a discontinued standard) The card has 4Gib of memory by the way, which is well enough to handle a single A0 150DPI image (as evidenced by most of my other pictures which lack live effects and work somewhat more smoothly). Even selecting layers in the layer stack stops working half the time on that specific image. I'm trying to get work done and it seems mad that even waiting for the canvas to update does nothing. I'm not sure what more I can do? I will upgrade my hardware at some point in the near future but right now it is a daily struggle. So I guess my question is, are there plans to improve canvas updates and general reactivity in Publisher? Thank you

There's hardly anything more I can do, so I'll just let this sit here. The wetransfer link expires tomorrow, to anybody willing to investigate this : hit me up if the link is down by the time you read this. Cheers, Hadrien

Well then, let this thread be my report

I believe this is the place to report bugs ? is there another ?

Fantastic @Wosven. Thank you so much for investigating this ! I was at a loss. So I rasterized the whole group for convenience and it worked. I won't be doing any more adjustments on this so I feel safe just baking it into the layer. What is it about the curves adjustment (or the way the layers were arranged) that makes the bleed blank, do you have any idea ? I don't really see a connection. Is it a bug, or something that's expected (given the initial layer arrangement) ? Hadrien

Guys here is the file. No clipping is going on as far as I know. https://we.tl/t-TUsFjhG57k

Thanks for the suggestions. All these would show in the viewport too though, wouldn't they ? I don't see how something would affect only PDF exports. Do you care in taking a look at the file by any chance ? On the attached picture, you can see page number four, left half of the problematic spread (misleadingly the layer group is named P3). It contains the main text on the page, as well as three groups for the sea turtles' endangerement level graphic below. The page is constructed the same as all others, and does not contain a master page at all. As you can see the bleed is full of interesting pixels. The PDF export however, looks like this : All other pages look like this :

Publisher 1.10.1.1142 on Windows 10 PDF export for print outputs a fully white bleed area on pages four and five instead of the colour that's supposed to be there, whereas all other spreads include proper bleed. The file is on wetransfer, please come forward for the link, thank you Is this something I've caused unintentionally ? is it an option somewhere I checked by accident ? Btw some linked documents (PDFs specifically) will still not render on export, I assume this is a different issue ? Thank you Hadrien

Hi, 1.9.1 crashed on saving, now 1.9.2 crashes on resizing inline tables. Resizing them while they're floating works fine, so that's what I do, and then only do I cut&paste them inline. Additionally, on reopening my document two out of my four tables (that each occupied a page) have disappeared. One was invisible (supposedly because it was slightly too big for the page it was on, even though I had not toggled it invisible), and I was able to recover it by cut&paste, resize and "make inline" again. The other one does not exist anymore and all the content that came after it is simply gone. I inspected all the subsequent pages, they're totally empty ! That's several pages of work, gone ! Before noon, when I saved and closed Apu, everything looked fine. How can a file look fine on saving and then look different on reopening, with loss of information ? These problems have been going on for several months now and what I thought was stable quality software is turning out to be completely unreliable, has been making me miss deadlines too many times. I can't spend my time reporting bugs, uploading log files, I just can't. This is unbelievable really. So right now my mindset is to go look for an alternative that's reliable. I don't need anything fancy, no bells and whistles, just that it works for what it's supposed to do. Obviously I don't have time to migrate now, so I'll have to put up with these issues, finds workarounds, walk this landmine until this particular job is done. Cheers, Hadrien Get yourselves some quality control ! something is wrong with your development practice.

@Jon P it appears to work ! Thanks ! Now, APu crashes when I try to move a table. Will report when I get the time.

Yay thank you !! I'll be trying out as soon as I ship my current work, so in a few days.

Hi @Jon P , I saw 1.9.2 was released some days ago. I couldn't tell from the release notes whether this particular bug had been fixed. Would you be able to tell me ? Thanks, Hadrien

Thanks @Old Bruce ! it's an improvement, now my 6 chapters go 1, 2, then start over at 1 until 4. Seems like I have to wrap my head around this numbering thing still, ... what could be the reason it restarts after a while ? should I upload my file instead of beating around the bush ? edit : there's something super fishy going on, and I wonder if that's not related to my file corruption issue (logged after 1.9.1 release) : one chapter appears twice in my TOC, on pages 10 and 14. It's actually on page 14 so I have no idea what's going on. I'll investigate some more...

Hi all, I realize this is a very basic thing, so basic in fact that none of the tutorials I've watched seem to even mention it. I've been searching for hours (on forums, etc) for how to do this : the guide I'm making has chapters -so far so good. But they're all numbered with the number 1. My expectation was that this number would increment each time a paragraph is assigned the text style in question. I would greatly appreciate any pointers, I'm so tired of looking around. Pretty sure it's right before my eyes but I can't find it. Cheers, Hadrien

Much appreciated !