When working on large canvases and complex drawings, where accuracy isn't an imperative, it would be nice to have an ability to choose accurate preview of moved content (which is a current default) and fast. Later it roughly what we already have with transform tool - no layer effects, no blending modes, layers hierarchy doesn't have to be regarded etc. When user have ability to do something fast, he doesn't mind if it is "wrongly" presented to him, as long as correction is also so fast to do. With current Move tool it is even impossible to do certain things. Using pan as input device will always produce jitter. Having to deal with large amounts of data and tendency to present them accurately on-canvas, Krita sometimes fails to produce any kind of preview at all.
I'm not convinced that adding another mode to the move tool, to make it work like the transform tool is the right direction... Like I said on irc, it might be better to make the transform tool work on all selected layers. One problem is that it's really hard to make a useful preview image of multiple selected layers.
Me neither. I am always in favor of removing options. Obviously the right solution is to speed up move tool, which is almost impossible to do since users will continue enlarging canvases in the future. The problem is not how (in)accurate preview itself is. I am talking about being unable to see what I am doing, which is far worse than having something wrongly presented. Should I open a new request about allowing transform tools to operate on multi-layer selection? I can't find similar report in tracker.
Yes, that probably should be a separate ticket.
We now have a nice outline that helps with the move tool.