You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
On Linux, it is possible to set DPI scale less than 100%:
Doing this, and passing the scale to Ultralight, breaks rendering. It seems when using ultralight::View with initial_device_scale < 1, foreground elements are not rendered for border regions of the framebuffer that correspond to the scaling.
This happens in the Sample8. Note the text missing, but background being rendered:
It also happens in my custom application. I was able to confirm that pixels of ultralight::Surface come already with this defect, so this is not related to blitting onto the window surface. The bug appears to be entirely contained within the internal logic, triggered by the initial_device_scale input, not connected to the operating system functions.
My application exhibits interesting behavior, where the "extra" region gets randomly hidden and then revealed when I move an element through it:
On Linux, it is possible to set DPI scale less than 100%:
Doing this, and passing the scale to Ultralight, breaks rendering. It seems when using ultralight::View with initial_device_scale < 1, foreground elements are not rendered for border regions of the framebuffer that correspond to the scaling.
This happens in the Sample8. Note the text missing, but background being rendered:
It also happens in my custom application. I was able to confirm that pixels of ultralight::Surface come already with this defect, so this is not related to blitting onto the window surface. The bug appears to be entirely contained within the internal logic, triggered by the initial_device_scale input, not connected to the operating system functions.
My application exhibits interesting behavior, where the "extra" region gets randomly hidden and then revealed when I move an element through it:
DPI.scaling.reveal.webm