Peter Werner

That would have the following disadvantages: Bigger file size in both Publisher and the PDF (compared to pass through) Does not prevent rasterization happening twice (once inside Publisher and once by the printer's RIP) Will cause triple rasterization if the image is then resized in Publisher after it has already been rasterized to document resolution Does not prevent quality loss if the effective image resolution is very close to document resolution (say, a 300dpi image placed and resized so it ends up at 298 dpi) Breaks link to original image asset file Has to be re-done manually when file is replaced (eg. when client sends an updated version) Will not update if, for whatever reason, the document resolution is changed later The "Rasterize" command has no setting for Nearest Neighbour sampling, so rasterizing an image that requires Nearest Neighbour resampling would just cause it to go blurry one step earlier in the process If I hand such a document to someone else to work on, they would have to be aware of the situation or risk messing it up when making changes I think being able to choose per image whether the data is to be passed through without upsampling (better quality and smaller PDF file size for images like QR Codes, screenshots and pixel art as well as matching InDesign's behaviour) or upsampled to document resolution on export (which, for especially for photos will lead to slightly better quality results) would be a better solution. In my case, I worked around the issue for now by rasterizing all the photos to document resolution (which probably uses bilinear or bicubic sampling) and then setting the resampling in the PDF output settings to Nearest Neighbour. I was only dealing with a two-page document, but if I was working on say, a tech magazine with a mix of screenshots and photographs, that would be rather error-prone and cumbersome, to the point where frankly, I'd just do the project in InDesign instead and not worry about it.

When exporting to press-ready PDF from Publisher, the software always seems to re-rasterize images that have a resolution lower than the document resolution (or the resolution specified in the export dialog box). There currently seems to be no way to just pass the images to the PDF at the original resolution. There is no per-image control over how they are resampled/upsampled. I have a case here where a client has provided me a QR code as a PNG image, with a resolution that is much lower than my document resolution. Exporting with any other sampling than "Nearest Neighbour" will obviously result in an blurry QR code. Exporting with "Nearest Neighbour" will result in all other images potentially getting upsampled using Nearest Neighbour sampling. As a user, I actually may not want images like the QR code re-sampled at all. Re-sampling that image even in Nearest Neighbour will just generate pointless data and the RIP at the printer will re-sample once more at native print resolution, so depending on the image and how close it is to the document resolution, there could be some generation loss, artifacts or even moiré incurred. Ideally, we would be able to have a PDF that has images at their native resolution so they will get resampled only once during the RIP phase at the printer at its internal maximum physical output resolution. InDesign can output such a PDF and in fact will do so by default – it does not upsample images at all. Another use case where upsampling could be undesirable would be when creating PDFs with web graphics for client review. In that case, the client may actually want to inspect the image at a pixel level by zooming in in their PDF viewer. Just passing the images through to the PDF at original resolution would allow for this (though the user would have to use a PDF reader that does not do image filtering when zooming in for this of course). Just imagine someone places a 299 dpi image in a 300 dpi document and exports it to PDF. The resampling result will not look great. But if the native image is passed though, it may actually look great (depending on the zoom level in the PDF viewer). Another use case are screenshots in magazines. If these are resampled like photos, they can become blurry or exhibit ring artifacts when resampling with a filter like Lanczos. Of course one could just export the document with Nearest Neighbour filtering, but that would cause all photos in the magazine that are even slightly under document resolution to become Nearest Neighbour filtered as well. Don't get me wrong, upsampling to document resolution is actually a good idea in most cases as printers in my experience only ever do Nearest Neighbour filtering and you may gain quality. It might be useful to allow the user to control these settings at an object level. For instance, for the QR code in my example, an option not to upsample could be activated. Upsampling of images could be deactivated globally in the PDF export settings. Or, with upsampling activated, all regular photos in the document, except the QR codes and screenshots with upsampling off at object level, would be upsampled on export for slightly better quality.

That settings seems to be unrelated. In fact, I was able to reproduce the issue without creating a symbol, only grouping the text and the line and then applying a constraint to the text layer in the example I am attaching. The arrowheads disappear as soon as the constraint is applied to the text when the enclosing group is implicitly converted from a Group object to a Constraint Group object. Changing the arrowhead end position setting on the line does not remedy the clipping.

Here is how to reproduce: Add a line Add a stroke and extremely large arrowheads to it (eg. the Bar type at maximum size) Add an Art Text layer Mess with Constraints and create a symbol (Constraint Group) from both Now the arrowheads are not taken into account in the bounding box computation and appear clipped. Additionally, maximum size for the Bar arrowheads is capped at 500%, even when higher values are typed into the text input field. In certain instances, this can be too small. Same on both Mac and Windows.

I can confirm the issues with the Lighting filter being out of place after changing the document DPI. I have a lighting filter on a group of multiple Art Text objects with layer effects (this lighting effect is always missing from PDF exports by the way, but I'm not sure if that's maybe a separate bug). If I change the document DPI in Publisher via the "Document Setup" dialog box, the positions of the Lighting effect, which are apparently resolution dependent, don't seem to be updated. This also messes up the undo stack – it's not enough to undo the DPI change via history to get the Lighting Effect back to its original position. The document in question was probably originally created in 1.7.2. Would be useful, by the way, to be able to directly change the DPI in the export dialog box for raster image formats to be able to render vector elements at higher or lower resolutions than the document was originally set up with. It just allows to set absolute pixel dimensions, but if I have a document that's set up in points or millimeters, that's less useful than a DPI/PPI control. Sending stuff to be professionally printed onto CDs/DVDs, in photo labs, certain sign companies, engraving firms, mug manufacturing companies etc., sometimes necessitates sending designs in the form of rasterized data rather than in vector formats like PDF like a "proper" printing company would accept. Also guys, if at all possible, please move the compressed file size calculation in that export dialog box out of the UI thread –for very large files (such as exporting a 60 cm x 90 cm poster at 1200 ppi to be printed by a photo lab), the hang caused by that can make changing all the export settings a major pain because every keystroke or click causes a 15 second hang. System: Affinity Publisher 1.7.3, MacOS 10.11.6, late 2008 Aluminum MacBook upgraded to 8 GB RAM, NVIDIA GeForce 9400M 256 MB, external Monitor connected via DisplayPort with custom ICC profile Eizo Color Navigator 6/Spyder3.

I just ran into this as well and it is definitely an important feature. In my specific case, I'm working on prototypes for DVD boxes and the inlays are to be printed by a photo lab as regular 20cm x 30cm photo prints and then trimmed manually to the exact final size. What's even worse, the workaround I would normally use (going through a temporary PDF) is not an opting this time since there is a bug that prevents the live effects on the type from exporting to PDF correctly in this document. Having the option to include bleed on placed files is very useful whenever something is printed/prototyped on an office or photo printer to be trimmed manually – for me this often happens with prototypes, CD/DVD covers, postcards, small signs/labels etc.. InDesign's "Add Crop Marks.jsx" example script makes this really easy to set up.

I can't seem to find the link right now, but I recall Adobe Photoshop product manager John Nack posting about the possibility of a Linux version of Adobe Creative Suite (at the time) on his blog many years ago. He stated that the reason why Adobe didn't do a Linux port wasn't the size of the potential market for Linux versions at all. Their market research showed, however, that offering Linux versions would not get them a significant amount of revenue from new customers, it would instead only shift a large part of their existing customer base to a different operating system. Serif today is in a very different position – shifting a large part of Adobe's existing customer base to Linux sounds exactly the kind of thing they might be interested in.

Thanks, that's pretty cool, I must have totally missed that feature! Though this of course doesn't fix the problems when it comes to images that have been opened, pasted etc. as well as user created raster layers that have been transformed. But it's still a great improvement for many of the page layout-related use cases.

This is also an issue in Photo where you can non-destructively scale a raster layer. Since there is no "Reset Layer Transform" command other than "Rasterize", which kills your resolution, you may end up painting/cloning etc. on a layer that has vastly different resolution than the rest of your document. Being able to see the scale of a layer may be crucial information in that case. Or you may have opened a raster image, transformed it slightly for a quick test, and now when you create raster layers, they won't match up with the resolution of your base image so you want to reset that base layer to its original size to match up 1:1 with the document pixel grid without losing all your undo stack. I recently had the case where I was quickly throwing together a comp for a client and then had to do more retouching work to an image in the comp stage in context of the final design than I had anticipated. However, the image was transformed non-uniformly to fit a design and now I was required to move everything into a clean 1:1 retouch of that image, but there was no easy way to find the scale factors and/or reset them so I could transform everything back so I could do further work on the image in 1:1 resolution. The only solution I found was to select the base image with the transform tool where it displays the DPI and then work out the inverse scale factor, but that only works for non-uniform scale, not for rotations and the like. In other cases, where I can't just re-import the base image, I'm also worried that this will lead to rounding errors since DPI is only displayed rounded to whole numbers, so I might end up with an image that's half a pixel or so off from 1:1, which is definitely not ideal since then every single pixel will be resampled if I export to a file at my document resolution. And when it comes to Publisher or Designer: In some cases, being able to use exact percentages like 50% or sqrt(2) times the original size is also important, for instance if a design was created for A3 and has to be resized to A4. With the current implementation, I often find myself re-importing the original image in such cases to get it to a guaranteed 100% and then multiply the values accordingly. That's obviously not a good workflow. The only option would be to look up the exact target dimensions in mm for A4 or know them (which I admittedly do, but there are other more exotic formats). Same thing again when I then need to check that the size is correct, where otherwise I could just look at the scale factor in a panel and know it's accurate at first glance. I'd therefore vote for the Transform panel to show the exact layer transform of image and raster layers and also to allow the user to edit, copy and reset them. Moreover, being able to Copy&Paste transform and inverse transform would come in handy in many cases.

I'm using an external (Apple) keyboard that does not have an Fn key. Pressing the Fn key on the built-in laptop keyboard does not change the behaviour I described, neither in conjunction with the F-keys on the external USB keyboard, nor using the F-keys on the built-in laptop keyboard. Further investigation shows that furthermore, the shortcut input control in the text style editor dialog box doesn't accept F14 and F15 since these seem to automatically be tied to adjusting screen brightness on the laptop screen. Holding the Fn keys has no effect on this behaviour. I think if all of this was a case of system shortcuts taking precedence, they would just carry out their system function when pressed instead of nothing happening, as they do in case of F14 and F15. By the way, I've come across another oddity: Assigning F16 shows a warning that the shortcut is already mapped to "Zoom Out". Turns out this shortcut actually works for zooming out the document, but in the keyboard shortcuts editor in preferences, only Cmd+- is listed for Zoom Out. Maybe this has to do with the shortcut editor only allowing being able to handle one (probably the first) shortcut per command.

Just did a quick test, and It would seem you are correct. You may want to post this in the Affinity Photo bug reports section.

Your comment makes me suspect I'm actually dealing with a bug. I thought I had mixed something up, but checking an older proof PDF I sent out to the client, actually I think the numbering re-started with each story at first and then something caused it to suddenly continue across stories without me linking the text frames. Removing the "Restart Numbering Now" attribute from the first paragraph of the second story causes numbering to continue from the previous story. The text styles were all created from scratch. Maybe it's just a really obvious setting I am missing. I'm afraid I can't share the document in question publicly since it was created for a client, but I would be able to send a copy to the development team if necessary. Still, even if it's a bug, it would actually be really useful to have both options.

The "Restart Numbering" setting currently allows resetting the counter per-document, per paragraph sequence, or manually only. Sometimes it would be useful to have a numbering sequence specific to a text frame or sequence of linked text frames, i.e. per story. This small handy setting would eliminate the need to manually reset numbering at the start of each story and therefore also one source of potential errors since it's quite easy to forget.

I think it is worth mentioning that now with that Publisher has shipped with Studio Link enabled, you can just go to the Photo persona and use the Channels panel to check your separations. For some reason, page/spread borders disappear, but it's still a viable workaround. You need Affinity Photo installed in order for this to work.

Works fine for me, I only had "Allow JPEG Compression" deactivated on the PDF/X-3 preset, otherwise my settings should be identical to yours. The screenshot shows the PDF re-imported into Affinity and as you can see it consists of a raster image with a clipping path. I made the file by simply creating the S as an ArtText object and dragged the placed image inside in the layers panel. If you are seeing different results, I'd say it is definitely a bug and not a design limitation of the Affinity backend or the PDF library.

Not 100% sure about early versions of InDesign, but I at least don't remember running into any issues in that respect. But you may well be correct since around InDesign 1.5 and shortly thereafter 2.0.X when I started using the software, printers here always still asked for native files and not PDF so back then it wouldn't have been something I would have to worry about. That being said, if I create a shape in any of the Affinity apps, import an image and drag it inside to clip it, it exports to PDF as an image clipped inside a path (tested with the PDF/X-3 setting). The edges are sharp and not pixelated no matter how much I zoom in (tested in Apple Preview and by placing the PDF into an InDesign document, can't try Acrobat Pro right now because the updater corrupted my installation once again). Re-Importing the PDF into Affinity retains that path as a clipping mask for the raster image. Ergo the Affinity backend as well as the PDF library it is using (apparently PDFlib + PDI) are perfectly capable of outputting bitmaps with clipping paths, they are just not doing so in the case of the effects in the original poster's case for whatever reason. One possible explanation is that the partially transparent raster image on the left that is overlapping the lines somehow triggers rasterization of the objects below if any effects are applied, which to me looks like a bug.

I cannot possibly imagine that the Serif team would choose a PDF library that cannot clip a pre-rendered raster image into a path, that's pretty much as basic as it gets.

My apologies, I must have used different wording when searching for existing threads. Thanks for the replies!

I have a document with a few two-column text frames. On the left pages of two double spreads, the column resizing double arrow mouse cursor appears in the wrong spot, either two the left or to the right of the column gutter. On the right pages of the spreads, the hotspot is centered on the right column gutter guide (rather than the center of the gutter as one would expect). The feature can still be used since the offset remains constant if the user clicks and drags. One just has to find the right spot inside the text where the phantom handle appears. In that area, clicking to place the text caret is not possible since the column resize behaviour "steals" the mouse events. I have not tested if this is related to that particular document or not, but just in case, I have made a copy of the file where I encountered the bug that I can provide staff with if the issue cannot be reproduced otherwise.

These effects should still export as raster images clipped inside of the corresponding paths and not raster images only, so there is clearly something wrong here beyond just the resolution of the effects. Whether this is by design or a bug I cannot say.

I can assign F1 through F16 for paragraph styles, but pressing the corresponding key with a text cursor in a text frame has no effect. Assigning other shortcuts, say, Ctrl+Y, works as expected. Tested with German keyboard layout on English language setting of Publisher, USB keyboard connected to MacBook. OS is 10.11.6 (El Capitan). Curiously, pressing any F-keys hides the transform controls and frame/column outlines temporarily until the mouse pointer is moved. If this has to do with certain F-keys being taken over by OS shortcuts like launching Dashboard etc., I would expect Publisher to check for that and display a corresponding warning when assigning the shortcut.

Affinity currently lacks a lot of options required for full professional CJK typesetting, including things like Furigana and Warichu, dedicated composition/line break options etc., it's not just vertical text. My understanding is that they are going to get the basic features for roman typesetting done first to make Publisher and the rest of the Affinity range as usable as possible as a layout software for the roman languages and only then add features for full CJK and RTL language support a later point in time, which is completely understandable. At least IME for CJK languages seems to work fine now in Affinity 1.7 after being completely broken in earlier releases. That's already a good indication that this topic isn't ignored by the development team. I hope that one day we will have full CJK and RTL script support in Publisher. The last thing we could hope for is a messy and fragmented situation like with InDesign where you have to install multiple versions of InDesign in parallel, all of them developed by third party companies where you can only hope they don't mess each other's files up.

Running a web server for an API (or a web browser or Flash Player for dialog boxes) is exactly the same type of bloat that we all criticize Adobe for. They, too, are running at least one local node.js web server in the background even if none of their applications are active. You may have noticed it because it occasionally hangs, crashes or blocks system shutdown. Scripting isn't that performant in the first place, and running single command through a full-blown set of script interpreter, API in that scripting language wrapping and parsing the HTTP communication, HTTP server, system TCP/IP stack and what not is such a colossal waste of resources, not to mention it opens up the entire system to security vulnerabilities. If you really have a compelling case why you absolutely must automate Affinity applications over an HTTP API, you could easily implement the relevant parts of such an API yourself, either via a native C++ plugin, or even easier, with a Python script and Flask (a combination which by the way could probably replace InDesign Server with a desktop version of Publisher for quite a few use cases). This is by no means about what is trending, but what makes the moste sense in terms of providing an efficient, robust, powerful, extensible and future-proof scripting workflow. Besides, Python has been a staple of application scripting in the VFX industry (among others) for decades, and for a reason. This is by no means just a "trend" and there is no indication whatsoever that it will go away any time soon. As an established solution, there are a lot of companies invested in it so continued development is pretty much guaranteed for the foreseeable future. Lua is indeed great for smaller quick and easy scripts and easy to integrate in a C++ program, but not that suitable for more complex solutions (hello, Adobe Lightroom Classic mess) and its object-oriented language features are lacking compared to Python. C is not even a scripting language and does not scale too well, which is why it has been replaced by C++ as the language of choice for larger native software projects decades ago. It is also not true that the complexity of implementing one solution in different languages is always the same – it is true for very, very simple code, but as solutions become more complex, some languages, and their ecosystems, are significantly more powerful than others. As someone who has worked with both Python and VBScript I can tell you that there are some things that would be a nightmare to write in VBScript and a five minute exercise in Python. We don't need a hundred haphazardly integrated scripting solutions so everyone can avoid spending 30 minutes to learn a few basic new syntax rules, but one great scripting solution that is well thought-out, seamlessly integrated, simple and easy for basic tasks but powerful enough for the most complex requirements that arise, and interfaces well with the outside world (AppleScript/COM, network/HTTP APIs, third party libraries and applications, native code etc.).

I just came across a workflow stumbling block while working on letterheads with Publisher and delivering them as MS Word templates. There were multiple variations of the letterheads for different departments, which were created as different pages in Publisher in a single document based on common Master Pages My workflow was as following: Design everything in Publisher Create slices of header and footer for export to PDF Iterate on the design in Publisher until approved by client Place PDF graphics of header and footer in Word, recreate text that needs to be editable and text styles in Word, make color Palettes via PowerPoint (due to Microsoft logic) Last minute fixes: Make changes in Publisher, re-export the slices and replace the images in Word Deliver MS Word templates and color palette files To my knowledge, this is a pretty standard workflow for doing branding projects where MS Office templates are required. Since Publisher has no Export persona, I created my slices for my letter header and footer by opening the Publisher document inside Photo. Trouble is, Designer and Photo only support one page. When opening a document, they always just display Page 1. This can be circumvented by opening the file via the File menu in Publisher. In that case they will open the page that is currently active in Publisher. For me that meant that I had to go my target page in Publisher, use the menu to open that page in Photo, create my slices, save, close, re-open in Publisher, go to the next page, create my slices and so on. This is not only very cumbersome, but it also makes it really difficult to copy slices, settings or dimensions to keep them consistent across pages. Imagine that I find that I want my footer slice to be a bit larger after all because one version has some extra elements and I want them to be a consistent size – I'd have to close, re-open the document in Photo via Publisher for every single page and make my edits while remembering the dimensions to type in. The same applies to exporting the slices after a design change – when not using continuous export (haven't tested if that works from Publisher), this workflow also requires closing and re-opening the file for each page I want to re-export slices from. Imagine a change on a master page with slices on a 15 page document. That's a waste of time and unnecessarily frustrating. I guess the limit on page navigation was imposed onto Designer and Photo to prevent people from just downloading blank files with multiple pages and using those instead of buying Designer or Photo. My proposed solution would be to offer page navigation in the Export persona in both Designer and Photo. Then just always return to the default page 1 when going back to the main Photo or Vector persona if Publisher is not installed. Another possibility would be to take it further and globally enable page navigation in Photo and Designer if Publisher is installed, similar to the Vector and Photo personas in Publisher. The last option would be to add the Export persona to Publisher, but I'd say that's probably overkill.

Oh, I see – my apologies, I must have somehow missed that post earlier when I did a forum search on the topic. For quick sketches or design mockups, a simple Shift+Tab is still much faster, but I can live with that in exchange for being able to right-anchor tab stops. The only real issue I see is that once an Adobe InCopy/Quark CopyDesk type editorial workflow is implemented, this is something that will have to be done on the design side, rather than being available to the copy editor by default. Also, a unique right alignment tab character is useful for use in a "nested styles" scenario where everything after such a character can be assigned a specific character style automatically (like in the aforementioned end-of-story marker example where the character can then automatically be formatted with a special font or color).