Mike,
The -nowine switch is documented in the LTspice's help manual, see Modes of Operation > Command Line Switches:
"Prevent use of WINE(Linux) workarounds."
I don't know if this happens to all, but when using LTspice normally, that is, without any arguments passed on to command line, trying to resize, for example, the dialog window for adding a spice directive, will resize the perimeter without the contents of it (I hope I found the right words). If there are more than 4 or 5 lines of text, they will be hidden (but you can still scroll through).
Using -nowine will allow resizing the contents, too, so now you can have an entire novel and be able to see all the text.
The downside of this, of using -nowine, is that when you try to move a component (e.g. with F8), it will leave minor dots behind (at grid's spacing), traces of them like crumbles, which cumulate, but they are gone the instant you drop the symbol with the mouse-click, or when the viewpoint of the window changes (for example when you zoomed in and, when moving the symbol, the current view of the schematic shifts to adapt to your direction).
This isn't as bad as it seems, since moving components doesn't mean playing around with them, and there's an actually good indirect effect: suppose you want to copy a symbol and move it, symetrically, downwards or anywhere; the trail will let you know if you strayed from the path.
The custom launcher is used simply to tell WINE that I want to use the INI from the LTspice's directory (custom installation, models, etc) and to add the -nowine switch. The clean installation is in a different location and uses the %AppData% from Windows.
This is about as best an explanation I can come up with right now.
Good luck understanding,
Vlad :)
