Capturing IMF variations in observations and simulations - how far along are we?