Subscribe
Dystopia’s Weapon Balance Scheme
January 25, 2007 1:53 pm
In the Slashdot thread for the aforementioned Gamasutra statistics articles, somebody pointed out Dystopia, a Half-Life 2 mod that used beta test metrics to balance weapon DPS. There’s an audio interview on the game’s website; I listened to a bit of it and I’m still intensely curious as to how they corrected for player skill bias in accuracy.
This is the first public word I’ve seen on FPS weapon balancing with metrics.
There’s very little out there on FPS balancing in general. Here’s a conversational article at Unreal Wiki, where the writers argue that math doesn’t really have a place in it. Maybe that’s why no one talks about methodology. When you say “I made some shit up and iterated until it felt right,” people tend to think that you don’t know what you’re talking about.
Psychochild wrote:
I believe that design and balance are still more of a black art rather than a real science. “I approximated then refined the values based on feedback” is the more polite way of saying it.
But, I think sometimes people fall in love with mathematical purity in “balance” and it really doesn’t work out that way. I actually wrote a bit about this in my article in the book Massively Multiplayer Game Development 2.
A simple example is to give someone a +2 to damage but a -2 to hit for a skill. Balanced, right? What if the damage range is 1-4 but the to-hit range is 1-20 (hello, D&D!). You are giving a 50-200% increase in damage for a mere -10% chance to hit. Even if both damage and to-hit were based on a 1-20 random result, in most games doing more damage is more important than hitting; doing enough damage to kill someone outright is much more powerful than the ability to hit someone every time in many combat systems.
Generally, “balance” is achieved when you give the player a variety of actions, and all actions are viable in at least some situation in the game. An imbalance is created when a few (or, at worst, just one) options are clearly superior to all others. The real fly in the ointment is that often this is not based on any objective reality, rather on the opinions and whims of players. The “flavor of the month” uber-builds are a prime example of this. Sometimes they’re not powerful, but since there exists enough information to show how to build a relatively powerful character, they come to dominate and be seen as “the most powerful.” Until someone else (usually an Explorer archetype) comes up with an alternative and shows that character to be more powerful.
My thoughts.
Posted on 02-Feb-07 at 8:53 am | Permalink
Anonymous wrote:
>>I’m still intensely curious as to how they corrected for player skill bias in accuracy.
I’m a tester for the Dystopia team, and I can tell you that, well, they didn’t really. More often than not, the data was used to justify the lead programmers own biases and preferences than to help supply a fair and fun gameplay experience.
The prime example here would be the rocket launcher of the game, a massive gun that can kill most players in one shot and if laser guided. It was ridiculously easy to aim (to the point where you always had >100% accuracy (from hitting more than one target, generally).
This, of course, wasn’t particularly fun for the victim. The only thing that kept the weapon somewhat in check was the massive reload timer of 15 seconds, which kept down the damage/second and prevent good players from going on killing sprees (instead, anyone, regardless of skill, could get a kill every 15 seconds).
But anyway, this meant that the weapon really didn’t look that bad in the stats. But I assure you, in the actual game it was hell. It was a tool that allowed any player (of any skill) to kill another player (of any skill), period. But the stats seemed right. Possibly a bit weak even.
But the general playerbase loathed the weapon. The only ones that didn’t were the ones that used it exclusively
Anyway, the lesson I’ve learned from it (I doubt the programmer in question has, considering that the test team, after a large amount of discussion, and at times outright pleading finally managed to get it nerfed from it’s 1-hit-kill status. But since then, it has gradually gotten buffed over and over in the release candidates for version 1 (the “complete” version of the mod) (incidentally, the test team isn’t used anymore, instead using semi-public betas)) is that statistics mean absolutely nothing by themselves and that you have to be very careful to not use them to reinforce your own biases and “mis-decisions”.
Statistics are very easy to miss-apply, and Dystopia should be held as a shining example of how NOT to use them.
Apologies for the rant, but this has been a somewhat sore issue for me the last couple of months. I hope was worth your time regardless.
Posted on 17-Feb-07 at 11:35 am | Permalink
Propecia. wrote:
Side effects of propecia finasteride mg….
Propecia. Propecia side effects bad. Propecia generic. Propecia sexual side effects. Propecia baldness. Propecia side effects. Propecia side efects. Buy drug propecia….
Posted on 22-Sep-07 at 3:25 am | Permalink
Meridia weight loss. wrote:
Prescription weight loss medications - meridia….
Meridia weight loss. Prescription weight loss medications - meridia….
Posted on 25-Oct-07 at 10:45 pm | Permalink
buy hydrocodone without prescription wrote:
hydrocodone prescriptions…
firkin recondense…
Posted on 16-Dec-07 at 1:17 am | Permalink
Order phentermine 37.5mg without rx. wrote:
Order phentermine phentermine online….
Order phentermine. Order phentermine phentermine online….
Posted on 02-Jan-08 at 9:43 pm | Permalink
Wellbutrin. wrote:
Wellbutrin effects depression rxlist….
Teenage cutting wellbutrin prozak. Wellbutrin side effects. Wellbutrin and cheap and no prescription. Wellbutrin xl. Wellbutrin and weight loss. Wellbutrin. Seroquil wellbutrin….
Posted on 02-Feb-08 at 1:10 am | Permalink
Allergic reaction to zithromax. wrote:
Does alcohol affect zithromax….
Zithromax usage informatin. Zithromax treatment length. Zithromax gohnoreahdosage….
Posted on 22-Mar-08 at 5:50 am | Permalink
Buy viagra. wrote:
Buy viagra online….
Buy viagra online. Buy viagra in canada. Buy viagra….
Posted on 22-Mar-08 at 8:16 am | Permalink
Lexapro side affects. wrote:
Lexapro….
Lexapro….
Posted on 22-Mar-08 at 9:53 am | Permalink