Even the document level is not fine grained enough. Snap to grid is frustrating because for all UI/UX designers it should be on 100% of the time.
That is, until you go to work on icons or illustrations. Then you turn it off and forget to turn it back on and then you ruin everything else you touch after that until you realize.
There needs to be a way to identify the intent based on the content. Maybe it’s applied at any node within the doc, like the page or frame with icons, but not the screen comp.
Please help stop the proliferation of fractional pixels