Paul Mc

Sorry, @JWalt but the Vector Crop Tool is misnamed (IMHO) it isn't a crop tool. It should probably be called a mask tool. The initial mask is the bounding rectangle of the curve. As you have probably found you can move the sides by dragging the handles but if you move them back the original curve is revealed. The only way to apply the crop of the mask is to Rasterise it and then you lose the curve and end up with a pixel layer. Clearly, this would be no use for your laser cutting requirement. I've had a recent job where this was needed. I'm not sure if there is a better way but here's how I did it. Create four guides for the sides of the bounding rectangle, then with the node tool hover over the intersection points of any curve that crosses a guide until you see a small dot at the intersection of the curve and the guide, then left-click to create a new node on the curve. Once all the intersection points were created, go around the rectangle holding control/cmd down while hovering over the curves outside the box until the red X appears and then left clicking to delete each of the unwanted curve segments. It was a tedious job but gave me a result I could work with.

Are the extraneous curves from the deleted layers, outside the crop area or both?

I don't really think it has lost the clouds: With linear light on a black background you can see that the clouds are still there but as I mentioned earlier the Transparent Glass setting is what "thins out" the reflected and refracted colours. The problem might be that in your render you are using a clear or white background. Without seeing your Blender scene it is hard to be sure. One way may be to render the glass panel as a glossy material with no roughness and then use the Opacity seeing in Affinity to dial in the transmitted colours coming through from behind, then overlay the window frame as that will always be fully opaque. Is there any particular reason that you are using Affinity for this as Blender's compositor can do all this type of work very well.

This is a complicated issue as the transparent film option is not very accurate when you select transparent glass. My suggestion would be to use two renders one with the reflection at 100% and the other using transparent glass and blend between them in AD. You might get a better quality result if you have the photograph image as plane aligned to the camera behind the model that way transparency isn't going to be an issue later in AD

I've just found this nugget in a demo file: scal:dpi="72" x="0px" y="0px" width="864px" height="864px" viewBox="0 0 864 864" This would make sense in informing the application what the DPI value is as it is otherwise omitted. Plus, the absolute pixel dimensions using px as the suffix. Update: I think I may have jumped to the wrong conclusion here. It looks like SVG has no standard for DPI. scal: above probably refers to Sure Cuts A Lot - and not scale as I had thought. So it is probably proprietary setting that gets around the missing DPI problem within the Sure Cuts A Lot domain. Having said that has anyone tried the px suffix for the dimensions?

I'd noticed this as well. It is kind of pointless for most real world uses of SVGs and really only applies when you embed this into something like a web page where the container defines the size. I'd previously tried making the changes you suggested but never ended up with exactly the same dimensions on reloading the file back into AD. Of course, it is possible that AD is buggy in this area and this would never work but your changes might work downstream for the cutter. There's too much experimentation and guesswork going on here we really need Affinity to step in and say exactly what's going on and whether this is achievable or not.

The only other parameter that I can think of that might have any effect would be the Set ViewBox setting. You have it set here so that clearly isn't issue. Do you have a simple SVG file which you know works correctly for a precise and predictable size of geometry? If you are able to share that on here, with a description of what it should generate, then I could take a look and see if there is anything obvious that is missing.

Just for completeness I have no problems opening the files in AP on Windows 10. I presume you've checked for free memory and disk space etc? You are on Windows 11 and although I wouldn't expect this to be an issue it is possible that there's a side-effect from somewhere else causing this. I don't suppose you set things up to be a dual boot with Windows 10 when you did the update did you, or is this a new computer?

Hi @barbara c just to add to what @Old Bruce has said, for any resizing where you have fixed pixel dimensions you may need to crop the image to fill the available area. If the aspect ratio of the source image is the same as the destination then you may only need to scale the image to fit. The Instagram dimensions are a bit awkward in the sense that I know of no cameras that produce images with that ratio as standard. For portrait images you could scale the image to fit the height and then crop the sides or pad the sides if the source image is narrower. Alternatively, you could scale the image to fit the width and then crop or pad the top and bottom. If you are likely to be doing a lot of these then my suggestion would be to create a new document preset that is 1080 x 1350 pixels in size and then drop your photo into that. You can then scale the picture to fit either by allowing parts to fall outside and be cropped or to have all the image fall completely inside the document and fill any gaps with a black or white background.

I tried opening the first one with AP and it works for me: I'm not sure if there are any transformations happening between your upload to the site and my download so a better test, as suggested above, might be to put the file or files into a zip file and upload that.

Hi @P_Nog the export uses the pixel resolution setting under the Document or editable after clicking the More... button, when doing the Export: So my 50mm green square above is exported at 50 / 25.4 * 72 pixels on each side. From the SVG file created: width="141.732" height="141.732" A simpler example would be a 1 inch square at 72 DPI which would appear in the SVG with a width and height of "72" Because your Vinyl cutter has a 300dpi scanner I'd try using a value of 300 as the Use DPI: parameter. If you used 72 DPI in your test you would expect the cutter to operate at roughly a quarter of the side length or a 1/16 of the area. An A5 design would be approximately 5 7/8" x 8 1/4" which if rendered at 72DPI when 300DPI is expected would appear at a size of 1.41" x 1.98" - is that near to what you observed or was it smaller than that?

1. Draw an oval on a layer above the photo. 2. Apply an fx using Gaussian Blur 3. Right-click the oval layer and set it to be a Mask to Below Vig0001-1578.mp4

You are correct. That's gets me going again. Thanks.

Hi @Old Bruce well it looks like the artboard approach is great for making the changes but isn't able to be used for preparing the presentations I currently have set up in Publisher directly. I need more control over which images are selected and where and how they are placed. It seems that by default only the first artboard image is placed and subsequent ones are unavailable. I know I can open the AD file using AP and it will create a page for each but that's too crude. I'm going to look at the export persona to see if I can set that up to be an intermediary, so to speak, to create intermediate files that can be linked into the AP document.

Hi @Wosven Thanks for that info. I didn't realise that was possible. That gives me another route forward. Only four, fortunately! But lots of items use them.

Right! OK. I'm starting to get it. This is a one off project that I won't get again from this client - but I'd like to build a workflow that handles this situation well as I can see this being something that might happen with other clients in the future. I should probably drop all the current designs into a single file and use artboards for each one rather than separate file. They are separate because it's easier when automating construction of the presentation PDF in AP which brings other content into the mix. I'm prepared for the pain of rebuilding the colour assignments as I expected that to me necessary - at least once. What I wanted was to be confident that the new approach would allow for a more efficient pass the second or third time around.

Thanks @Old Bruce If I do this I presume I'd have to import this palette into each of the documents and then visit every object and replace all the standard colours with Global colours from the palette? Then, when the colour changes again I'd modify the palette in the palette document, resave and then reimport the afpalette file to each of the documents again, or will the act of saving automatically update the other documents when they are opened? Sorry, for all the questions but I can't find anything that explains the mechanics of this. Surely I'd only need the template file if I was starting again or adding another document to the set, no?

Hi everyone, I've several documents which share a set of "brand" colours for text and shapes. I now need to change the colours across all the documents. I'm expecting that this may not be the only time I'm asked to do this so I'd like to rejig everything so that it can be done efficiently on the second and subsequent request. The history: for the first document I created a "base" document with a document palette and used (standard?) colours defined in it. No Global or Spot colours (yes, I know - silly me). I then copied this document as the basis for the others and built the contents for each one. Then came the client change requests. So I'm now in the position where I have several text frames and artistic text strings, for example, spread across several documents that are the wrong colour and need to be changed across all of them. From my understanding I think I need a Global Colour in an Application Palette. However, this seems to be impossible to create. I can create Standard Colours in an Application Palette but this isn't bound to the objects that use the colour, instead it appears that the colour is copied from the palette when first assigned and after that the object's colour is independent. Is my understanding correct? I feel that I've unwittingly painted myself into a corner with this one and that there is going to be a truck load of work applying these change requests. If anyone has any suggestions of how to ease the pain then I'd appreciate hearing them. (As an aside, is there any fully comprehensive documentation that explains how the palette system works in AD, apart from the online help?) I'm using the latest release of Affinity Designer on Windows10.

Ha! This sometimes happens. With the best will in the world sometimes things cannot be fixed. I wonder if they are completely different fonts in the sense that one may have been saved during the development phase and then it was worked on some more before the other was saved as a final version? Just guessing. That last point about the TTF to OTF still smaller seems to me like there is something dodgy about the base data. Thanks for trying though.

I love this community! When you have complete strangers taking time out to help others solve problems like this, it is great. I can see that you are on a mission @LibreTraining I'm curious to see what you find out.

Thanks @Lagarto As was pointed out earlier by @LibreTraining I'm guessing the the font file was not constructed correctly or automatically generated using old or buggy software. Whatever the reason I'm able to move on. As curious and interesting as this is I feel it is not worth taking this forward (at last for me) as there is a solution: don't use the OTF version but use either of the TTF versions. I don't possess any font editing software or skills so this is a little outside my knowledge and comfort zone. Thanks for the help and support on this.

It is very strange. Are you getting the two strings aligned? I've no explanation why this may be happening. Maybe there is a side-effect from the screen font scaling or something like that. Because TTF works I'm content that I can move forward with this. Thanks for your help and trying to get to the bottom of it.

... and bingo! It does work correctly if I use PDF/X-1a:2003 (This is the same file as above just exported differently).

Here's the previous test I did when I discovered the problem. I've taken a screen grab of the PDF from Adobe Acrobat Reader DC, changed the colour of the black text to red and then overlaid the original from AD on top. This is using the OTF file. I used the digital high-quality setting (PDF 1.7 (Acrobat 8)). The Arial text has been used as a guide for eyeballing the alignment and you can see that matches up pretty well. The exported PDF and the AFDESIGN files are included below. If you open the AD file and things work correctly then the black text should obscure the red text - apart from a few pixels that might be visible along the edges. kerning OTF test.pdf kerning OTF test.afdesign

Thanks @Lagarto, that's really strange. As you can see, my experience was different. I didn't realise that kerning is off by default in Word so that explains why they match. The original issue still remains, though, and that is that the on-screen image of the text was different from the PDF export - at least on my PC, so maybe there's something strange in my setup. I'm away from it at the moment but will get chance later to try uninstalling the fonts, rebooting, and re-installing them one at a time to see if I can reproduce what you are seeing. Of course, this is all academic now as I have a working font and document but I appreciate your efforts here, thanks.