But unless initiative is simultaneous, one side gets to do all their stuff before the other side gets to do its stuff. That's the "freeze frame" issue right there.Initiative is rolled per side and when during a particular side's turn an individual gets to act depends on what they are doing.
I've always played AD&D (and B/X) as go through each of those phases for the winning side, then for the losing side. But here you seem to be saying here that you go through the phases collectively, with the winning side going first in each phase and then the losing side going next, but all archers going before all casters going before all melee-ers, etc.But looking at the initiative rules they are totally freeze frame. I shoot arrows you shoot arrows. I move and attack you move and attack. Wash, rinse, repeat.
From memory Rolemaster 2nd ed works this way. I don't think I've come across AD&D run this way, though.
I've used this sort of "continuous initiative" in Rolemaster. But I don't think it's the default approach in AD&D, is it? I don't remember any discussion of it in the DMG, nor in Moldvay Basic.If my action is to move and the magic-user's action is to cast a lightning bolt we need to figure out whether I've just stepped into his blast or not. And so on. (this last example is one of many reasons we went to individual initiatives in 1e rather than init.-by-side - if my action is to move and my initiative is 'x' then I'm assumed to get where I'm going on my init. and it's pretty easy to figure out where I am at init. x-1, x-2 and so on in case anything happens to me en route)
I think Lanefan is talking about continuous initiative. But I agree with you that without continuous initiative, or a system for off-turn actions (a la 4e), then talking about constant and simultaneous motion is just colour, but has no effect on resolution.There may have been assumptions about the action but there is no actual support for those assumptions.