so how about i set this as my goal for my prototype :

1. build sandbox, player, a few different “mech-suits”
2. implement mechanic to directly control the bots and each performs some “unique” ability.
3. implement ability for player to assume control of one of the bots (physically rolls up to one of them and jumps inside) and then be able to “switche control” to the other bots in his “Squad” (basically just a keystroke, once connected to one of his bots, he could control any in his “Group”).

I think if done right, the whole package could be fun. But, if there is any variety and complexity to the pets, the game will be a massive job to get working and balanced, to the point of being in any way fun to play. When I talked about not doing tower defense style games, this is a reason why: any game like this will hinge on the balance between a massive number of elements.

If you focus on the pets, perhaps don’t do any training, and don’t have many (any?) variety at first. And, perhaps don’t have any control over them … perhaps you activate them and they follow you with well-defined roles. As player, you are focused on controlling your character, but the robots move around you in a swarm whose movements are (largely) deterministic and so you can control them indirectly with your own (imagine moving forward, and turning sharply left, and having a “tail” of them whip around to effectively strike the enemy).

so a player can occupy a particular suit and use it to protect him and augment his abilities, or he can command the suit to do what he wants it to do (what he’s shown it how to do?) and let it go off and do it, whatever it is (fight stuff, find stuff, fedex stuff, etc).

i was hoping the keyboard manifestation of teaching the suit (pressing a key makes the suit move a part of itself) would present that “starting from scratch” experience in a pleasing, feel-oriented way, but yeah, it could just end up being a PITA. heh.

i had also considered the ship-upgrading thing (sort of thought of it from a “car wars” pov) – this was actually where i started with this. i thought maybe starting with the completed suit would be less complicated.

so maybe not focus on the mechanic of teaching the suit to move and instead focus more on the game play mechanics of “how to care for and raise your pet mech suit”? that is, after all, my desired end result.

The essential problem is that, in the end you know that these things are just a means to an end. The game you describe is structured such that the parts are things you need to do something else; so anything you make them do to get those parts will be hard to make it feel “integrated” into the game. Imagine if whenever you got a weapon in WoW that you had do a bunch of junk to be able to use it.

I’m not sure what to suggest to make it better. And I’m not sure what to suggest as a prototype for this. I think that you need to step back from this and think about what would be the fun.

To me, the fun would be in _using_ the mech parts, and perhaps assembling them in different ways to get different abilities (pretty standard “upgrading your ship parts” kind of thing). I’d dump the whole teaching thing, and focus on some fun combos of parts. More than that, I’d focus on a very few small parts and think about how they could combine, and beyond that, I’d focus on a few ideas that feel very different, and allow you to do very different things. What would make this game different, or what would make it feel fun?

– your normal character has no defense, but can perhaps move lightly and quickly
– add some legs and you can kick, stomp and perhaps jump and move faster. but you are now stuck to the ground mostly. But,you can take damage.
– add arms and you can grapple and punch? Swing? Climb?

I’m not sure. You want to avoid complexity from lots of combinations …
hmmm.