Wedge009

The request can certainly be phrased better. Subjective, like #7479. Presumably #7106 would be a sub-set of this.

> Insofar I know, debug-proofing the mainline candidates was not done, likely due to Dalas not knowing about it being a thing. I'm not sure how 'official' a 'thing' this...

This sounds like a low-level Direct3D-related problem: https://learn.microsoft.com/en-us/windows/win32/api/d3d11/nf-d3d11-id3d11device-createtexture2d https://www.thewindowsclub.com/the-gpu-device-instance-has-been-suspended-on-windows-pc suggests it's a hardware and/or hardware driver issue. Doesn't sound like something directly caused by Wesnoth nor something that Wesnoth can...

This seems like an extension or another aspect of similar issues such as #1167, #4912, #5575, and #6663.

~What kernel/s are you using? This doesn't bode well for #2993 -> #2939.~ Sorry, realised this is an issue with llvm, not amdgpu-dkms.

Thanks for reporting. I think this is related to the selection of font sizes relative to the resolution available - probably only occurring at lower resolutions since I didn't observe...

The images you've posted suggests there's some scaling increase done in 1.16.2 and 1.17.3+dev is rendered at 100%. I have the inverse problem/behaviour: I just tested 1.17.3+dev (f17b8b76e98) and it...

Yes, I remembered seeing that, didn't realise it was merged already. @mesilliac Any idea if @knyghtmare's observations might be related to your changes since it looks like what I observed...

I recognise I'm a minority, but I'm actually quite comfortable with 1x scaling at 2560x1440 and above. But I probably don't have the very high DPI screens that many of...

If a decision is made one way or another, how do we resolve this without causing yet more changes and translation work?