RogerBW's Blog

The universal system, the one RPG that could handle any setting you could throw at it, was a popular idea in the 1980s. Why then, and where has it gone since?

To understand the aspirations, I think it's helpful to remember the common style of games then; my experience is from the UK, but I get similar impressions from elsewhere. Typically one GM would run one game, with players coming and going, until he (usually he) got bored with it, or moved away; then he might start another one, perhaps in a different system. Campaigns were expected to be long, but very often not to have a defined goal or ending: you'd go down tougher and weirder dungeons, or make increasing amounts of money with your tramp starship, until the game stopped for some external reason.

And while only Dungeons & Dragons and its imitators have stuck with it as a core requirement, rules knowledge was important. A player was generally expected to know how to generate a character and at least how to play their part in a fight without needing any assistance from the GM or other players.

To my mind those two ideas inform the desire for universal games: you don't want to have to learn a whole new system as complex as D&D when it comes time to change campaign. Wouldn't it be great to have one system that could handle everything? And if campaigns are long (and relatively non-lethal) you won't be generating characters very often, so it's fine if that takes a while.

Another point that was often brought up was character portability: wouldn't it be great if you could take your character from one game to another, not just from one dungeon-bash to someone else's (which was often contentious in itself), but from one setting to another? People (including me) enthused about this a great deal, but it didn't seem to happen very often, even once universal games became available. An imported character has no backstory in the setting, no social network, no motivations beyond the purely internal. And if it's "heroes from across the universes go out and fight evil", well, how do you make sure that each PC can contribute something?

This is where I think GURPS suffers rather: it's very clear that points cannot provide anything like game balance outside a single setting, and often not even there. What is game balance when one character can be a commando-trained security chief and another is an expert spaceship pilot? They won't be doing their things in the same scene. At best, one can argue that every deviation from the mean standard human is a way of calling for spotlight time: "I can do that, I'm an expert marksman". Even the negative ones: "I can't do that, it's against my code of honour" in game terms is still a way of making your character stand out.

But GURPS (and every other point-buy game too, Champions certainly, even something like Savage Worlds where the "points" are whole advantages) has to pretend that an extra level of strength is worth the same to the axe-wielding barbarian as to the power-armoured drone jockey.

Meanwhile in forty years the overall complexity of systems that players are prepared to accept has dropped. You may still need to "learn a whole new system" to join that new campaign, but it'll take you half an hour at most. Character generation is quick and easy, and if it leans into cliché at times, there's no harm in that.

Meanwhile any universal system that keeps producing expansions suffers from combinatorial explosion: when book B comes out, it has to work with the core rules alone, or with book A. When C comes out, it has to be tested with core, core+A, core+B, and core+A+B. And so on. Saying "this book depends on this other book" only helps for a little while (and hurts sales).

But I think there's a hybrid that hasn't been widely analysed: the modifiable core. I'm thinking here of systems like Gumshoe, Powered by the Apocalypse and its derivative Blades in the Dark, where what the players buy isn't a core book plus a genre supplement but a single all-in-one book for that setting. The rules for setting A won't be exactly the same as for setting B, and that's just fine; they're still broadly familiar to players who've played the other game, so there's less learning effort, but at the same time anything can be tweaked if needed to fit the setting. Sanity rules don't need to be in the hypothetical Gumshoe Core, they're in Trail of Cthulhu, and if a different Gumshoe game took a different approach to mental health it could have an entirely different set of rules without anyone worrying about how they interacted. And for this reason you don't need to worry about combinatorial explosion either: even if "this book" is effectively core + A, it doesn't have to be compatible with core+B, core+C, etc.

I think this is what designers who forty years ago would have tried to build another universal system are now doing instead.