People like new things, in a game it's essential to keep offering new things through the life of most games. Without new things players get bored of the game, they don't buy your sequels, and you get bought up by some larger company if you don't just roll over and die. FPS have new weapons that come up later, new enemies to surprise them, and new Terran that can change the game play dramatically. Racing games have new tracks and sometimes new vehicles. RTSs and fighters not so much if they're competitive only, in campaigns you can get story mode which could be the main draw if they do it right and unlock new fighters/units as you go along but they have somewhat of a different idea of who they are made for.
For RPGs it's always been new loot and sometimes new abilities. In older games it was purely numeral increase, you had to level up to fight the next batch of enemies or be really lucky. This is where the title of the post comes from. You were effectively on a game treadmill, working to stay at the same spot where you could easily dispatch enemies. Go to too high of an area and don't have to reload means you hit the wrong spot and you head somewhere else. So the entire point of the game becomes a goal to go forward while keeping all mechanic application the same.
Thankfuly this has changed a bit. More modern games introduce newer mechanics later on in RPGs or have abilities that make you do different strategies so the treadmill effect is a bit staved off. Though it is seemingly harder in RPGs than in games that have more actual skill in them. FPS has different weapon handling, hit box detection, and other fun things such as vastly rewarding headshots or even crippling enemy limbs (not that those things can be added to an RPG they just feel oddly tacked on and less rewarding). Granted most RPGs today are more action orientated so my old thoughts on it being turn based are somewhat false, though they aren't high on the twitch based skills.
I've been thinking on this with various game progression designs. One problem with me is I don't think about how designs scale. Most games I play tend for you to get on the treadmill and find that sweet spot and the GM or game throws at you accordingly to how "difficult" the game should be. For Combat Project I have not thought of this at all until today. I have little idea how if I did a real game instead of a test would it keep interesting. That's more to think on though, I'll have to look at this subject again later when I have more ideas on what I want at it's basic level.
Edit: I have not played the recent Fallout games. I have played the first two a bit. I can't comment much on them too much. Crippling limbs refers to a few games I've seen where shooting an enemy in the legs actually slowed them down or what little I know of Deadspace where there was an entire system based on it.
