I'll spare you the
usual rant that only demoralises people anyway, and say just this about this thread in general: Ideas aren't people. They're not little children all frantically waving their hands and going "Oooh! Pick me! Pick me!" They're not going to care that you are a human being with limited capability and limited time who can't implement them all. It's
OK to just pick one that you think would make a great game and roll with it. Even the most dedicated and productive game designers have ideas that they'll simply never get around to implementing, in fact given that you just started this thread with one idea... they probably have a lot more ideas than you do. And that's OK. All that really matters is the games you make.
I mean, take this thought you have here. It'd make a wonderful game and would be entirely doable in a 7DRL. Let me just see if I can give you a few pointers and see if I (definitely not a game designer by trade!) can spot a few possible pitfalls in the idea.
One, remember that it's the player's emotional reactions that make a game worth playing. Roguelikes are a very abstract medium and can have a hard time arousing much emotion... but, still, there
is the anxiety of exploring, hoping for healing equipment rather than monsters when you're on your last 10 HP. There
is, sometimes, a sort of fear when you know a very out-of-depth enemy is near (and far better if you describe its presence by sound!). There's a sort of exhilaration in slaughter after leveling up or gaining some powerful new ability or equipment. Even frustration, in controlled amounts, can be a powerful tool for the game designer, as it makes accomplishment taste that much sweeter.
What kind of emotional reactions do you want the player to have to this game you're going to be making? One thing you could do is make them feel hurried. Tell them that an enemy is on a spree gaining powerful abilities and you have to get to them fast. Maybe you could try to make them feel determined, determined to take out an enemy that just slipped from their grasp, or even angry at another that just stole a kill from them - but be sure to be extraordinarily fair or you'll make the player angry at the game rather than at the enemies!
Two, speaking of fairness, players are sort of sensitive to what I believe should be considered a technical term in game design theory: Bullshit. Here's one bullshit scenario that might occur if you're careless about design: The player never hears about one opponent becoming all-powerful and killing all of the other ones and being absolutely unkillable once the PC eventually runs into him. Here's another bullshit scenario: No opponent ever actually gets around to killing a lot of others, and the player just walks through the game without being challenged at any point.
You have to be strictly fair to the player and give them both challenge and the tools to defeat it. Deviating from that is bullshit (though in general not giving the player the tools is far more bullshit than not challenging them is). Here's one way you might be able to avoid bullshit, you could make it so that the opponents are guaranteed to have a certain amount of fights where one of them dies within some given time bound, and make a dungeon with a branching structure where the player always knows where to go to find the currently most urgent kills.
Three, remember what players are thinking of in the thick of gameplay. Is it the plot, perhaps an epic of wrathful gods, or a comedy of arcane tricksters, or maybe a tall science fiction tale? Is it far-away bosses and places they don't even know how to get to yet? Is it even the quests they are on right at the moment? Of course not. The player is almost certainly thinking of what direction they're going in, of what equipment they're looking for, of what resources they're trying to preserve, but really that's pretty much it, and so far that seems to be a pretty well-working paradigm of certain sorts of games such as roguelikes and FPSes.
Is your idea something that the player would be thinking about in the thick of gameplay, or is it something that would give context and interesting constraints to their actions? I'm thinking it's most likely to be the latter, meaning that this would be a fairly traditional roguelike except with special bosses you can go around killing and gaining the powers of. It doesn't have to be a
big roguelike, of course - you don't need a food clock since the player knows that tarrying around is going to make the bosses more powerful, you don't even particularly need an inventory or a lot of character advancement since the player has meaningful and interestingly constrained choices about which boss to go for and gain the powers of next. You just need enough tactical complexity to make clearing areas vs. hurrying to get to a boss an interesting choice, and to make killing monsters fun in itself.
(you could also base the entire game around the special opponents, leaving the usual roguelike stuff out of it... but I'm afraid it would end up sort of like Shadow Of The Colossus without the breathtaking art and wonderful colossi that way)
So how about it? It didn't take me long to mash this post out and a lot of it could well be bullshit, I don't mind if you don't make any use of it. But I would love to see a new roguelike developer going at it, whatever idea you happen to base the game you'll be working on on
