Haven't checked this thread in a bit because of reasons. Will try to answer those questions about my most recent idea. I'll probably end up repeating things a few times because my sleep cycle is currently in an awful state and I barely proofread this.
Firstly, are we limited to conventional combiner mecha parts? As in, could someone control a tail, if they wanted to? How about the sword, or helmet, or even a parasol?
That'd probably be up the the person running the game. Ideally it should be plausible for a mostly human shaped robot to fold up into whatever you're making, but that depends where the game will be on the super robot vs real robot scale. Given the somewhat silly premise, it'd probably be more interesting to lean towards the former. Heck, your initial robot might not necessarily have to be human shaped.
Second, could pieces be attached in unconventional ways, and if so, to what degree? Can a torso fit onto another torso? Can an arm have another arm attached to it? Can the head not be attached, and be held by an arm or another piece?
Good question. Limbs could probably just join at the... well, joints! A double arm bot would probably move by flexing around. Before you even ask: Yes. That would be simultaneously really dumb and amazing to witness. That's probably one of the simplest cases though. As for a torso to a torso... That'd probably end up either making some sort of (potentially mobile) fortress mecha or a flying "borg cube" sort of machine, depending on the robots involved. I'm not yet sure what implications a robot without a head would have to be honest.
It may be best to just have the head be a sub-part of some sort. Every robot might have a "command module" of some sort. Maybe it'll shaped like a head and placed somewhere on the robot. Maybe even somewhere inside the robot. Maybe not shaped like a head at all? Who knows? There's also the issue of self piloted (AI) robots of course... There's a lot of things to consider, needless to say. I do think having a clear head on the body should provide a notable bonus to most actions. I can't quite get across why I think that should be the case at the moment, unfortunately. Multiple heads should also be allowed, but probably be mostly redundant.
Third, how would the abilities of mechas and their parts be defined? Are abilities determined by bonus/malus to roll? Maybe different sized dice? Would mechas have stats that are compared for effects like damage, action order, etc? Do the pilots of the parts cumulatively decide on an action or set of actions, or can everyone just do their own thing?
When it comes to abilities/stats/dice systems/etc, it all depends how complex the gm and players want the game to be really. Anywhere from full minimalist to some gargantuan rule sheet with a billion sub tables like some people occasionally make here. I will say that being able to target individual parts of a robot should be possible, purely from a strategic standpoint. Meaningful decisions are key.
Pilots being in agreement should probably help in some respect, but they also shouldn't be forced to comply with the masses. That would basically just force half the players in a machine to just quote a post and say +1 rather than actually engage with the game. Not ideal, for sure. One option would for each player in a robot to have two actions. One would be their overall goal for the machine, and the second would be how they want to achieve that goal with their limb. Alternatively they could have some secondary goal to work towards.
Lastly, what would the end goal be? Defeat Alien invaders? Other mix n' match mechas? Would they be other PCs, or NPCs? Would they just be trying to do mundane tasks like go grocery shopping, walk the dog zord, get that promotion to assistant manager?
A monster of the week format honestly wouldn't be the worst way to go. However, a part of me is deeply amused by the prospect of our heroes (realizing they overslept) deciding to use an impromptu giant robot as a sort of carpool to make sure they get to work on time.
It seems extremely difficult to keep player actions meaningful. I think you need both a functional game when players are not combined and conflicting reasons to combine and stay separate.
Indeed! It's important for all the players to have things to do whether they're fused or not. One of the things that I (and probably a lot of other people) dislike about combining robots is that the new robot is basically only in control of the hero. Everyone else sort of gets sidelined in a lot of cases other than a cut in to yell stuff. I'd obviously want to avoid that.
I don't have anything concrete as of yet unfortunately, as I've been busy. The difference between passive and active effects does seem to be an important one. Ideally, all but the worst and most impractical combinations should be able to defeat your average robot. It'd be really dumb to try and fight with "literally just a pair of legs bot", but with enough jet thrusters it might not be impossible.
Here's how I'd do it:
Each player is a Robot Ranger, with a custom, unique robot. Each robot is its own distinct character, with its own sensible abilities. These can include things like flight, a weapons system, a force field, or less boring things. In addition, all robots have combiner abilities, which are broken into two parts.
I'll try and break this down a bit. Your method as a whole is certainly an interesting take that I quite like, but I'm gonna respond to each part separately because of reasons. This basic premise seems simple enough.
Primary abilities are used to act, but only one primary ability can be used per robot per turn, and there is probably a cooldown to prevent the same player or players from dominating control of the robot. This means that players have to take turns controlling a combined robot. Everyone except the player using a primary ability gets to use a secondary ability, which changes how the primary ability works.
I'd have liked some examples to differentiate between the two types, but given that I didn't exactly give everyone much to start with it'd sort of be hypocritical to ask someone to just make up example machines on the fly like that. All the same, I'd like to get a better idea about the difference between the two types.
I don't necessarily agree about the restriction on one primary ability per turn. However, you do want to be able to balance what exactly a machine can be capable of in one turn. I can understand that. Cooldowns do certainly seem to a good idea in theory to hand control between each player.
I've actually literally had a moment of mild inspiration right now so I'll put it here. The Dwarven Race RTD series had different modes of driving you could use during your turn. Aggressive driving added speed, but was riskier and might damage the car. Cautious driving was basically the opposite. Driving focus made you really good at driving but forfeited all actions. (Probably the least relevant example for this idea.) Naturally, standard driving had no bonuses or penalties.
It may be possible to port these "styles" over somehow or create new ones. Not sure how at the moment, but I'll think about it. Basically, give players different tactics to how they operate their limb, and potentially the robot as a whole. Ex: If you're a leg, walking aggressively might increase melee power due to speed, but make it harder to dodge.
The only way I think you could run it is to hide the mechanical side of gameplay from the players so that cheesing the system is impossible and if two abilities are incompatible you can just make something up under the hood, but that seems like a very risky way to run a game. Some sort of combining robot game is definitely possible, but maybe not one where multiple players form a single robot. It's a very interesting game design challenge, at least. I think I might keep trying to figure out how to make it work.
I do think character sheet submissions should be private messaged if this game happens. Players shouldn't know right off the bat what machine turns into what (other than their own of course). The game should probably start with a "There's no time to explain!" moment to shove the inevitably hideous robot right into the fray. I'm not quite sure about what to do after that though, as players would probably just ask each other what their robots do and engineer a functional (but mechanically boring) robot. I suppose that would depend on the total player count and the overall deadliness of the game. It might actually be a good idea to keep each round separate, and have new players each time. In that case, it would be important to keep the game relatively simple so that submitting and processing character sheets doesn't bog down the system. Alternatively, cycle the list of players (possibly somewhat randomly) for each mission. People can keep their old machine and character, but they'll just have to wait their turn.
Anyway, thanks for working on my little idea. Will try to check back again soon.