Andreas Koch

> I tested it again and I still see no difference between before and after this PR. Moreover, the status **before** this PR looks OK (I see both scroll bars...

> I suppose that one issue here is that different monitors can have different resolution, so the shell would help pinpoint the monitor and hence that monitors resolution. But that's...

> Do I understand it correctly, that a shell now should not hold a `Font` reference any more but a (cloned) `FontData`, so it can somehow update the `Font` if...

> Unfortunately, it looks still almost the same. On 125% it became better/good, but with the primary monitor on 150% or above, I still see the issue. E.g., this is...

In each of the following screenshot there is an adapted version of Snippet373 started with commit [2fe01bf](https://github.com/eclipse-platform/eclipse.platform.swt/commit/2fe01bfb14ea7ceb50c029267a073a0b06ff3ff9), 4 times each. The title gives the info of the configuration: 1. autoscaleOnRuntime...

Hmm, I cannot reproduce any vertical spacing issue, neither with quarter nor with int200. 100%  150%  175%  200%  Regarding the horizontal indentation: I agree, that int...

> Can you elaborate in more detail what this change is supposed to achieve and how to test it? I tried around a bit with multiple shells on different monitors...

@HeikoKlare The behaviour should be more consistent now, although I'm not sure, if I was able to reproduce everything as you had. Would be great, if you could have another...

> @akoch-yatta I tested again (this time only the known bug I mentioned before) and the bug is still there. This time you can see it if the window that...

As a remark just for further explanation. This PR is calculating the exact zoom, so the image handle perfectly fits the destination width and height. Therefor it will directly scale...