I understand your distinction, TGS, and it's also the True Zero Player Mode that interests me -- the full "clockwork universe."
From the sounds of it, that's very nearly a necessary component of the kind of game it sounds like Josh is trying to make. So I actually think ZPM is not just possible with out-of-the-box LT, but likely.But
-- and there's always a big but
-- I have a couple of chin-scratchers regarding ZPM.
1. A clockwork universe is neat, but not as much fun if you can't see the gears turning. To really enjoy ZPM, we need a way to be able to explore the universe as a kind of omniscient, disembodied consciousness -- one that's able to travel anywhere, see what's happening, and maybe even query individual NPCs to learn what they're trying to do and why. ZPM needs to do more than just be LT running without the player; it needs to let the human Observer appreciate that universe at a high (and maybe high-speed) level. Also having access to a low (inside the NPC's brain) level would be nice as well, but my feeling is that that's less important than watching the whole game universe change.
In other words, ZPM isn't enough -- there must be a way to display the effects of ZPM to the human "player."
Alternately, maybe ZPM should be defined as fully hands-off: you can simply say, "automate [or remove?] my character." That activates Zero Player Mode, in which the game goes into a very simple display mode, simulating all parts of the game world at the highest level without displaying anything about it. Then when you're ready to play again, you re-enable control of your character, at which point everything local to your character goes back into fully non-simulated mode.
That might be easier to implement. Not as much fun IMO as an interactive ZPM, but better than nothing.
2. How does research work in ZPM?
Once again I come back to what I'm increasingly concluding is a fundamental incompatibility between a "play-forever" simulation and a "play-to-win" game, and that is the tech tree. I tend to really enjoy researching technology in games to be won, such as Civilization and Master of Orion. But those are games to be won. In a game that's meant to be a simulation running for a very long time (in in-game terms), how do you allow characters to continue to improve their available technologies through research without research eventually becoming pointless?
Not only do I see that as an interesting question for LT in single-player mode, it also affects Zero Player Mode -- assuming the goal remains that executive NPCs ought to be able to do as many of the things that a human player character can do as possible. If exec NPCs can do research in the regular game, I assume they'll also/still be able to do it in ZPM... so how long can you run LT in ZPM until research advances become so trivial as to be meaningless?
(Note that designing research so that some breakthroughs randomly yield numeric "advances" that are actually worse than what a character currently has doesn't help here. No rational actor would switch to a tech that's worse than what they currently have, so all this would do would be to slow down the insertion of new technologies. It does nothing about the problem of eventually running out of numerically meaningful tech advances, it just delays the inevitable reaching of that state.)
The alternative to making research advances a numeric change is to define them as new combinations of functional effects. ("Blaster" plus "anti-gravity" equals "gravitic shredder," for example.) But even that must eventually reach an end as all combinations are generated.
If it were up to me, I think I might say that research progress is actually switched off completely in ZPM (as a full simulation mode).
But maybe someone else can square this circle. It's possible that I'm just not being imaginative enough, that I'm just wrong to see a fundamental incompatibility between simulations and games where technological progress is concerned, that there's a way to implement progress that remains meaningful no matter how long a game universe runs.
I know there've been threads about this; I've contributed to them. But to my knowledge they've never generated anything like a consensus agreement on how research progress can be implemented in a way that delivers a consistent amount of fun regardless of how long someone plays LT, or how long it runs on its own in Zero Player Mode.
(Note: I don't see this, and am not talking about it, as a "problem" to be "solved." The incompatibility of research progress with world-simulation isn't a defect in either playing a game or running a simulation -- it's just an effect that arises when a game designer decides to make a game that includes both progress and ongoing simulationist features. The question is not how to "fix" that -- because nothing is broken -- but rather which of those incompatible elements to emphasize under particular conditions or player preferences.)
Summing up: I like the idea of ZPM, and I think it's not only workable, it's probably already working. The only things I wonder about are being able to see the results of ZPM, and how technology advancement through research should be handled in ZPM.