Author's notes

This week's game felt like a much cleaner process. I knew the broad shape of the game in advance: it would be a battle gauntlet against a series of "identical" enemies, and the gameplay would be about deducing which of your randomly hampered stats needed to be remedied. (Some of these details were left over from Sketch #1 and past visions of knife fighting.) Because of the heavy use of status ailments for the various "flaws", and for the "Assess Foe" skill, I wasn't sure if I'd have time to implement everything I wanted. But I was able to get all the skills and structures working fairly quickly, leaving me more time to fiddle with the math and add a few extra flourishes (like having new elfs walk in from offscreen).

Indeed, fiddling with the math turned out to be the biggest time sink. I used the Defense stat for this game, which meant a lot of extra calculation to get the numbers how I wanted - especially since Defense can change mid-battle. I usually bypass this mess by using fake Def scores in menus and using piercing damage for enemy attacks. When I need damage reduction, I'll use elemental resistance instead, because in RM2K it's a multiplicative modifier, which is much easier to reason about than a subtractive one like Defense. But it's easier to change Defense mid-battle, which was necessary for this game, so I spent a good 10 minutes on the calculator just trying out different Atk and Defense values, for both the player and the enemy, including with form defects and enemy specialties. After finished, I realize I had messed up the numbers anyway: Flawed posture increases the damage you take by 5 points, while it should be 15. Use this dark knowledge to your advantage.

The focus on math, bound to a relatively simply game structure, meant I was able to plan out less of the game in my head. (Besides which I haven't felt much need to.) As a result, there was more room in the process for figuring out what my ideas actually meant. How long should battles last and how many do I want the player to be able to fight? How many of these form defect ideas can I jump on? Do I have time for enemy variants? Which ones can I get working quickly? Which ones need extra skills to make sense? The work of translating ideas into concrete form and making judgments on the fly about what I can or can't afford to put in the game: this is more fun for me than working off a plan, straining to remember exactly how I wanted everything to be and rushing to implement it in time. This was a less stressful, more rewarding hour than the last one.

Sketch #3 had a big combat design experiment at the center of it. My vision was of having a combat system that only needed "one" enemy, and which provided variation from battle to battle by randomizing the player's condition. Winning the battle is a question of efficiently identifying and correcting the issues, which are deliberately obfuscated so you have to "feel them out" by fighting the battle. I think this experiment was successful in that it seems more efficient to accurately identify your flaws and correct them than to just hack it out. In an ideal world, the effect of a "flaw" might be a little subtler so there's more tension between leaving it in place and spending the time/MP to confirm and correct it. But for a game this small, giving flaws a pronounced effect and a low cost makes it easy to recognize and engage with the system. I felt I had to do this when I played, so I consider it a success.

What changes would I make if I had more time? I might like Assess Foe to cost more MP so using it isn't such an obvious choice. I'd like there to be more reason to discern an enemy's strengths by feel and to engage with ambiguity: am I getting hit hard because my defense is low, or the enemy's attack is high? Assess Foe gives you the answer right away, while Correct Posture costs more and runs the risk of being wrong. But in a game of this size, perhaps it's better for the Assess skill to be an easy sell. Half the appeal is the fantasy of "sizing up" the enemy. On that note, I would like there to be more to size up - more variation in enemy characteristics. I'd especially like it if enemies could have weaknesses, too, to create moments where you might want to leave a flaw uncorrected. But I think the four enemy versions I have at present are plenty for a one-hour dev process, especially since I'm more interested in the idea of sizing yourself up.

The scoring system, of course, is heavily subject to luck. In a larger game it wouldn't be fair. But it in this case, I think it only needs to give you a reason to try a few more times and see how far you can get by learning how to play. When you reach the point where it feels like luck is the main barrier to a higher score, that's a good time to quit.

My hope was to design a fairly simple combat system I could take and transplant into future weeks' games with dungeon-crawling and exploration structures. I wanted something simple that could provide a little attrition as you explored, and a little more texture than plain old trading blows. That way I could focus on the other aspects of the game and maybe get a little more done. But I don't think this system is actually well-suited to that: fights take too long, and are too costly. At the very least, I would need to design a new set of enemies if I wanted to put Ashua in a drungeon quawler. But I'm not confident this style of play would be very interesting if the enemies didn't have variants themselves. Or would it be? I suppose I don't have to speculate: I can just make another game and see what happens.