It occurred to me in considering Dick Terrell vs. Daemon X Machina that an appealing, if uselessly vague, way of describing strategy games is "puzzle games with statistics instead of rules". Decisions are weighted and evaluated in parallel rather than binary in series.
Armored Core is in a sense the apotheosis of that idea, since there are so many decision points (equipment) weighted across so many variables (the stat sheet for a mech has a dozen or so parameters in the first game and probably 50ish by AC5), yet they are all codependent (since there's only one mech per mission). If you think about buying/loading gear as the fleet-building stage of Sins of a Solar Empire, with the choice of arm/leg/missile/etc corresponding to choice of frigates/cruisers/carriers/etc, I think they line up fairly well in terms of what the player ought to be calculating and trying to predict. There's more pressure in the action phase of Armored Core because the player is in full control of the mech, which is the advantage (design-wise) of unifying all the statistical parameters. The player needs to design a mech that they are also capable of piloting, and some decisions will be based on accomodating that (for instance, a novice might prefer a slow machine with higher armor because they find the controls clunky and can't take advantage of the ability to dodge). On the flipside, there's more pressure in the decision phase of Sins, because the entire game is played out in real-time. The action itself is almost completely automated, sheer number-crunching.
It makes me wonder what a mech game with real-time mech building would play like. Probably similar to a traditional RTS (StarCraft), since Sins and AC are kind of sitting at two extremes. StarCraft does have full unit control, it's just - there are so many independent elements that the technical bar for maximizing strategic efficiency is incredibly high, and the casual level of play relies on a lot of Sins-style automation. For the real-time element (in this theoretical mech game) to make sense, there'd probably need to be some kind of resource race, and the mech would need to be upgradeable on the fly (since there would be no additional/replacement units to build - otherwise resources would become irrelevant as soon as combat began).
Actually, if you took the standard RTS model (building drones and exploring a map to gather materials), this starts to look a lot like a MOBA. MOBAs after all were built from an experiment with WarCraft 3, asking what would happen if the player controlled just one powerful "hero" unit instead of a bunch of disposable little ones (the original Defense of the Ancients, if I'm not mistaken). MOBAs just tend to have really boring/convoluted combat, since they use an action model designed and for decades optimized for controlling dozens of independent units in parallel. They don't do nearly as good a job as Armored Core at taking the aforementioned advantage of unifying your parameters. The Dynasty Warriors/Musou games are somewhere in the middle here - only a few player units and combat more complex than League of Legends, but much simpler than Armored Core. They don't have much of a prep phase, though, and therefore play more like tactics-action than strategy-action.
Huh. So what really obvious game am I forgetting about that does exactly what I'm describing? What about if we add a metroidvania backtracking structure, roguelite procedurally generated encounters, and soulslike boss battles? Tune in tomorrow to find out.
It makes me wonder what a mech game with real-time mech building would play like. Probably similar to a traditional RTS (StarCraft), since Sins and AC are kind of sitting at two extremes. StarCraft does have full unit control, it's just - there are so many independent elements that the technical bar for maximizing strategic efficiency is incredibly high, and the casual level of play relies on a lot of Sins-style automation. For the real-time element (in this theoretical mech game) to make sense, there'd probably need to be some kind of resource race, and the mech would need to be upgradeable on the fly (since there would be no additional/replacement units to build - otherwise resources would become irrelevant as soon as combat began).
Actually, if you took the standard RTS model (building drones and exploring a map to gather materials), this starts to look a lot like a MOBA. MOBAs after all were built from an experiment with WarCraft 3, asking what would happen if the player controlled just one powerful "hero" unit instead of a bunch of disposable little ones (the original Defense of the Ancients, if I'm not mistaken). MOBAs just tend to have really boring/convoluted combat, since they use an action model designed and for decades optimized for controlling dozens of independent units in parallel. They don't do nearly as good a job as Armored Core at taking the aforementioned advantage of unifying your parameters. The Dynasty Warriors/Musou games are somewhere in the middle here - only a few player units and combat more complex than League of Legends, but much simpler than Armored Core. They don't have much of a prep phase, though, and therefore play more like tactics-action than strategy-action.
Huh. So what really obvious game am I forgetting about that does exactly what I'm describing? What about if we add a metroidvania backtracking structure, roguelite procedurally generated encounters, and soulslike boss battles? Tune in tomorrow to find out.
No comments:
Post a Comment