doubts on exports to svg

[lsilva.m]

Hi everyone,

I am facing some issues exporting .pdfs in Word documents, where images (.svg exported from AD) look pixilated in the pdf document. I check the svg by opening in Inkscape and PowerPoint, and the svg looks ok (not pixilated and with layers). So, I am guessing that there is a problem during the export of the pdf by Word. I know this is not the place to solve this issue, but it is just to give you the context.

So, after some troubleshooting I found out that like when background layer is locked the svg exported from affinity is pixilated, which was easy to solve. However, this lead me to think that maybe there are rules or better practices to prepare documents that would reduce the problems (rasterized element within the svg). Since I am in a hurry, I just rasterize elements where I used transparencies and blend modes, and (so, I used the new "pixel" layer, and turn off the group), but in here I found out that naturally different dpi gave different results.

What is the recommended dpi to use to rasterize when you want to print (A4)? I tried 300 dpi and it was bad, so I used 3000 dpi to rasterize (because it was when I notice that the original image was not pixilated), then changed back to 300 dpi before export. Is there a better way to work that avoids this possible bad practice?

Is there a rule for the grouping of the elements in the file to guarantee a good export? For example, is it ok if a .tiff file is within a group that also contains art text? or should this elements be ungrouped instead?

When you have a picture that you need to apply a black&white adjustment, how do you do to export the svg properly? what settings to used?

I used artboards to make the figures, and applied during the export the preset svg (for export), and a custom svg export with the option not to rasterize anything. 

Hope it is not confusing.

Thank you  in advance o/

[stokerg]

Hi @lsilva.m,

3 hours ago, lsilva.m said:

However, this lead me to think that maybe there are rules or better practices to prepare documents that would reduce the problems (rasterized element within the svg)

Best practice is to avoid anything not natively supported by SVG, so things like Adjustment Layers and Layers Fx are the main ones that spring to mind as these will all be rasterised, which if you are trying to only use a vector file in your project, could cause some problems.  

3 hours ago, lsilva.m said:

What is the recommended dpi to use to rasterize when you want to print (A4)? I tried 300 dpi and it was bad, so I used 3000 dpi to rasterize (because it was when I notice that the original image was not pixilated), then changed back to 300 dpi before export. Is there a better way to work that avoids this possible bad practice?

 

What was the DPI of your document before you rasterised?  If you could provide an example file before rasterising the page, that would be helpful.

3 hours ago, lsilva.m said:

Is there a rule for the grouping of the elements in the file to guarantee a good export? For example, is it ok if a .tiff file is within a group that also contains art text? or should this elements be ungrouped instead?

 

That should be okay from my understanding but if you are exporting this to SVG, it will contain raster elements as you've got a TIF file in there.  

3 hours ago, lsilva.m said:

When you have a picture that you need to apply a black&white adjustment, how do you do to export the svg properly? what settings to used?

 

As SVG doesn't support Adjustment layers, you would need to export with Rasterise enable.  If you were to export this to SVG with Rasterise set to nothing, you'd see the adjustment would be missing from the exported file.  

 

[lsilva.m]

Hi @stokerg,  

Thank you for your answer

On 7/15/2022 at 2:30 PM, stokerg said:

What was the DPI of your document before you rasterised?

I had different files (microscopy photos and smartphone photos), that ranged from about 72 to ~2800 dpi. At first I change the dpi of the affinity file to match the exact dpi of the photos, but then, because there were too many, I just rounded the dpi of the highest image. I can't share the photos because the publication is not out, but I did a grid of microscopy images (something along these lines https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3001161)

Alright, I guess that in the future it would be better to avoid adjustment layers when making svg objects. I end up using them because with the blending modes it was really easy to make the desired effect. For example, I use a background color (e.g. green) and then with a gray layer on top with screen to make Petri dishes.