[ I'm not doing a good job of making time to blog about the design. It takes a lot of time to blog, mostly in the editing. But right now I just want to give you some insight into where the game is going and why. Can I blog about that in some faster way? Here I'm going to try to just jot some quick notes about lots of topics, and stop semi-arbitrarily at about 1000 words, because otherwise I'd just ramble forever. Let's see if this style works at all... ]
Skills: too easy, too flat, too simple
I’ve gotten feedback from a few people who’ve said the game is too easy, too flat, too simple. There’s a lot of problems contributing to that (and they’re a topic for later), but I think the problem is made to look more prominent than it really is because of the shortcuts I’m using for testing:
- My skill curves are artificially low, so testers have an easy time leveling up to test higher-end content. But that means you can max out a combat skill in just a day or two if you work at it, which can feel way too fast.
- The skill curves are capped at 50, instead of 100 like they will eventually be. The bottom half of the curve is the fastest part, so that makes the curve seem even faster, since you never see the more stately pace of the higher levels.
- All the skills are basically accessible instantly, with very low prerequisites. I can hardly expect people to test Battle Chemistry if they have to raise Alchemy to 50 first, so I’m skipping those prereqs.
- There’s no limit to how many skills you can raise at once — that’s by design, but feels awkward right now.
I can remove #1. Maybe it’s time to use more realistic XP tables. However, when I tried using a slower XP table for some of the crafting skills, lots of people reported that it felt like a Death March slog. I’m not sure if that’s because the XP tables were way too punitive, or because the rest of the game is still so fast that it seemed out of place, or because those particular skills were just boring and stupid, so making them slower was extremely noticeable. Or all of the above. Probably all of the above. But for the moment, I’m not trying to remove #1. Maybe in a few months.
#3 is still very necessary, because otherwise I can’t get people to test whatever new thing I’ve just recently added. However, I can do more to explain to people how the skills will eventually fit, I suppose. Like how battle chemistry vendor mentions that eventually you’ll need an Alchemy score of 50 to learn Battle Chemistry.
#4 is one of the trickier goals of my game design. It’s not an easy design to make work, but it’s what I’m aiming for in this particular game. But right now it seems insane, because it feels like you can trivially max out every skill within a few hours of acquiring the skill. I need to change things enough to break that perception.
First in the arsenal of tricks is adding money sinks. I’ve been making the cost of recipes and ability training very high, so you can’t afford to just buy everything as soon as you see it. I may have overshot the mark, actually, and made the costs too high in some places, but there’s several other monetary factors involved, so it’s hard for me to tell right now. (See below.)
High training costs aren’t a panacea, though. The biggest problem with them is that earning power doesn’t scale linearly, so a high-level person can earn enough money to quickly level up all sorts of skills to the midway point. Is that bad? Well, it’s not the end of the world, but it’s not ideal either, because some people will decide that the “right way to play” is to race to high level in one or two skills so you can farm money and buy up all the other skills at once. That sucks, because it means you’ll have no new skills for a long time, and then suddenly you have too many new skills to even know what you’re buying.
So I’ll probably want some other caps on training instead of (or in addition to) high costs. Maybe I’ll use a time limiter, so you can only train X times per week, or Y times per day, or Z times per NPC per week, or… something like that. That’s tricky to get right, though, because of different play speeds. If I set the cap at a point where the average player is affected, the more aggressive players are going to hit the cap quickly and be frustrated. But if I tune the caps for the aggressive players, the average player will never even see them. Which… you know, maybe that’s okay. Maybe I only need to have this sort of cap for the people who are playing a ton.
But it seems cheap to have a hard-coded limit in a game that’s about having an infinite canvas to explore, doesn’t it? So maybe there’s yet another option I need to find.
I’ve added a simple new stat called Notoriety, which goes up when you kill “famous” monsters. You only gain notoriety the first time you kill each creature, so the point isn’t to reward you for farming; it’s to make it more exciting when you stumble on a new named monster. It’s a pretty common MMO idiom.
But what do you get for your Notoriety? I decided that it should give you some benefits when dealing with NPCs. For instance, training costs could be lowered, shopkeeper’s caps could be raised, etc. Notoriety can also help you earn Favor with NPCs faster, which makes a lot of sense: if you’re famous and you hang out with some random shopkeeper, or give them fancy gifts, that’s going to have more of an impact on them than if you weren’t famous.
So I coded up these variables. I haven’t tested them for crap yet, but they’re implemented now! So my next thought is always, “Hmm, I have new variables! What else can I plug them into?”
My first thought was equipment. A “Hat of Charisma,” for instance. But I decided against that, because it’s really fucking tedious to always have to switch between your in-town gear and your combat gear whenever you want to hit town. Later on I’ll be adding some sort of equipment preset system, and then maybe it’s kinda fun to have a Shopping Suit that you jump into with the press of a button. But right now, it’s tedious.
My next thought was using it in Alchemy or a similar craft skill. That can work, but it’s a little tricky to get right. Suppose you can craft Potions of Handsomeness which reduce the cost of training. If these potions are cheap and effective, then you’d be stupid not to always drink them before every time you train. That’s tedious, not fun! On the other hand, if they’re expensive and ineffective, nobody will use them. A lot of Alchemy potions end up unused like that, because I set the stats too conservatively, and they just don’t pass muster. So I have to aim for that hypothetical sweet spot, where it’s kinda hard to make the potion, but it’s pretty effective if you do.
I’ll make the potion have a reasonable effect — say a 10% reduction in training cost — but require a rare component to brew the potion, so its not practical to always chug them. Is that going to be fun? Hard to say. In the worst case, people who feel compelled to “play perfectly” may really hate that design, because they’ll save up all their money until they can buy a ton of training with one potion. In the extreme case, that’s really un-fun. But with a 10% cost difference, hopefully that’s not worth the hassle. It’s worth trying, anyway.
(My guess is that it’ll end up rarely used, like most other Alchemy potions. But I can never really tell. And it’s not like those potions are going to be worthless forever — I can keep tuning the stats and costs until they’re more useful.)
But beyond potions, what else can I do with this new ability to manipulate NPC prices and favor? Well, a bunch of NPC-centric skills were part of the original design — stuff like Hypnotism, Charm, Jokes, and so on. But the more I thought about those skills, the more dubious I was, because of how the game is trying to emphasize your friendship with these NPCs. If you’re best friends with Joeh the weapon smith, should you really be able to get away with using guile and charms (and potions!) to take advantage of him? What kind of a dickish thing to do is that to a “friend”?
But if I can make that a fun choice, then maybe it’s worth it. For instance, say you have a Hypnotize skill. You can use it on NPCs who like you, and it makes them train you for free. But they’re 50% likely to remember what you did, and feel violated — and stop being your friend. Is that a fun conceit? Eh, maybe. But is it balanceable game mechanic? Probably not.
If you lost all your Favor whenever you failed to hypnotize somebody, it’d be a terrible skill, because raising Favor back up would cost far more than you ever got out of it. But if you only lose a little Favor, it’d make sense to always Hypnotize people and try to get away with it. Bottom line is that it’s not an interesting choice, because it’s either the right thing to do (mathematically) or the wrong thing. I don’t like skills like that. They give you the illusion of choice, but then punish you for making the wrong choice. Bleh.
But speaking of NPCs, there’s another angle I want to go with them, adding a couple of simple elements to each NPC’s simulation to make this sort of skill more plausible. … and crap, my word count is way past 1000. Next time!