I learned a lot about game design in the time between when I started my RPGMaker 2000 game Kinetic Cipher and 2003 when I decided to dismantle it and completely rebuild it from the ground up. I felt like starting the game over from scratch, applying what I’d learned about the engine and games in general, I could make a much more solid game.
And that turned out to be true! Kinetic Cipher (Again) was much better in a lot of respects. But even going back to look at the remake, I see where I made a lot of really dumb mistakes. See, once again I feel like I’ve learned a lot about game design in the intervening years, and I see a lot of things in KC(A) I would do differently.
(Note: I won’t actually be re-re-making the game. The only way I pick up KC again is if I get funding and a publisher. This is really just a thought experiment.)
1) The world map would be smaller.
The world map in the original KC was a standard RPG map. When I remade it I wanted something more like Chrono Trigger: a world map whose only function was to quickly get the player from one interesting place to the next. I shrunk the map sprites down to instill a sense of scale, and it looked pretty slick. But it’s still way too big! There are places on the map where you can stand and not even see any interesting location to enter. I liked hiding things in little nooks and crannies around the map, so I don’t think I’d use a node-style world map. But I’d probably have a rule like “there will never be more than 12 steps between the last required area and the next one”.
2) I’d introduce Melissa earlier.
Even if you’re an oldschool RM dude who goes way back with KC, you don’t know who Melissa is. She was supposed to be one of the player characters, and a pretty important one at that. Except the plot graph didn’t allow for her to join until well after the 50% mark, and neither version of KC got that far. She’s one of the game’s most interesting characters and absolutely nobody in the world knows she exists except me. Along that note…
3) I’d cut out two party members.
KC was supposed to have nine party members, when all was said and done, four of which could be in your party at any given time. The leader, Karen, was glued into her slot, so even if you had all nine you’d only ever really get to pick three at a time. I think nine characters is probably too many to keep a handle on, though. My gut tells me that Robin and Cor would not have been necessary as PCs. Robin was always a hassle to balance for a lot of reasons, and I had planned to really downplay her role as a combatant in the endgame. Probably would have been better to just have an NPC guest slot for her, than try to shoehorn a character into a combat and advancement system that didn’t really have a spot for her.
4) It wouldn’t be ATB.
KC(A) is too hard. Way too hard. I didn’t make it hard, though. The toughest spots in the game are spots where the RPGMaker battle engine just doesn’t support the player’s ability to make good decisions. Enemies all tend to attack at once, which means your whole party can die on a whim unless you’re picking your options really quickly. A system that pauses the action any time the player has control (something like Final Fantasy X or Super Mario RPG) would be ideal for the kinds of battles I was trying to build.
5) I’d dramatically shorten the freewheeling.
I tried to give the players a lot of freedom to explore things in KC(A). About a full year of work was put into the areas on the second continent, and almost all of it was unnecessary. There was a series of fetch quests that took a lot of work to put together and didn’t actually add anything to the game. I had too many locations that only existed because that’s how it was in the original. I didn’t really need a North and South Orin. The goal should have been to ferry the player through the plot (and the dungeons!) until he had explored all of Tekiel — and then opened up some freewheeling. At this point the player would have six or seven characters to play with, a decent idea of how the world fit together, and the means to travel between locations very quickly. That’s the time to introduce fetch questing, not when the player is on a crucial plot-driven errand.
There are lots of other things I”d probably redesign, too, but those are the ones that immediately jump out at me when I flip through my screenshot archive.