sup/tg/ - Articles

Got a concept? Awesome. Written some fluff? Perfect! Decided that you want a stat called Asskicking and a skill for Anal Circumference? That's wonderful.

Now throw all that shit out. It's just going to get in your way. When you want to make a game, you should forget EVERYTHING else until you have finished the very most basic part: Resolution.

How do you resolve shit in your game? For example, in 3rd edition D&D, you roll a D20 to try and get above a certain DC. In the Storyteller system you roll a pile of D10's, and each one above a certain Target Number(henceforth "TN") adds to the level of success.

It sounds basic, and intuitive, to start with this, but it really is not. A lot of people fuck themselves over by saving the actual mechanics of how shit works, on the most basic level, until they've already bogged themselves down with some completely unplanned vision of the REST of the system.

At this point you might say: "Sure, okay. What does it really matter? I'll just use 3d6 or something. GURPS was neat." Firstly, GURPS sucked. Secondly, it matters surprisingly much.

Let's take, for instance, D20 vs the Unisystem. As mentioned before, D20(same system as 3rd edition D&D), rolls a D20(surprise!) you add your modifiers, and if it's above some pre-decided number, you succeed! In the Unisystem, the TN is always 9, you add the relevant stat and skill to a D10, and the more you get above that 9, the more you succeed(mostly critical in combat).

In the Unisystem, the random factor is small compared to the skills and stats(1-5 on both, generally), meaning that your stats matter considerably more than random chance even early on(unless you're totally untrained), and that it takes a relatively low amount of skill to generally succeed unless it's an opposed action. The upshot of this is that planning is more important than chance, because you can generally rely on what you try to go as you plan it, then it's just a matter of understanding the consequences of your actions.

Conversely, in D20, taking it completely without any sort of DM arbitration, the random aspect outweighs your stats for a LONG time unless you're absurdly crunked out at something, effectively meaning that ANY undertaking is rather risky and that planning becomes less vital(because half the time you'll fuck up whatever you're trying to do, anyway.). The lack of a static TN(the DC being rather fluid and up to the DM) also risks causing "stat inflation." That is to say, your odds of failure remain constant no matter how good you are, because there's no upper limit for how difficult anything can be, you do the same things, same risk of failure(unless you decide to skip the story and go pick on peasants), just slightly flashier.

Using multiple dice for your resolution mechanism also means that maximal or minimal results are less likely than average results. This results in something akin to rolling small dice, in that your stats matter more because there's less wild variation in the results you get whan actually rolling.

Now, once you've decided on what dice to use, you need to decide on a couple more things. #1: Do your stats and skills apply a static boost or do they make you roll more dice? #2: Do you want people to succeed often?

"Dice pools" are used by stuff like Storyteller and, again, helps make the crazy results less likely(but they do add more rolling and a bit more work to it). As for success levels... It depends on what people roll for. Is it assumed that they basically snap their fingers and simple stuff just works and you only roll when they're trying to stab that dragon in the eye? Or do you roll for EVERYTHING and the difficult stuff is only when someone's trying to oppose them(the damn dragon doesn't want to get stabbed!)?

And you're still not done. Just rolling to see whether you, say, can pick a lock, is generally simple. But what happens when you get into a fight? This is usually where a lot of games break down completely. Opposed actions. Almost every fight has several opposed actions per round. This is where you want stuff to be as simple as possible. For example, replacing an opponent's Dodge roll with a static penalty to hitting the guy? Less rolls. Static damage reduction from armour rather than rolls to see how much it reduces? Less rolls! You can even go so far as to have static damage, but that reduces one of the more exciting(or nailbiting) elements from combat.

Anyway, this has turned long and incoherent. Here's the TL;DR version. Figure out how the system works and how you want it to work before you get started on anything else. And shit, don't be afraid to adjust an existing system to do what you want it to. There's no reason to reinvent the wheel.

~PurpleXVI

I figured I might want an avatar that, you know, was vaguely fitting.

game design, ramble, rant, advice - View All by Author

Comments

1 CAPSLOCKGUY

07:02am UTC - 10/19/2008 [X]

>Firstly, GURPS sucked
In b4 shitstorm.

2 Lord Licorice

10:23am UTC - 10/19/2008 [X]

Damn good advice. Seeing as building game mechanics are just as interesting to me as the actual fluff of a game, I'll be extra certain to work out the dice system in full before moving on to larger steps. The dice system can pretty much set the mood of the game, particularly when you take into account (as you mentioned) the method players achieve success, e.g. random chance.

3 Cainen

10:33am UTC - 10/19/2008 [X]

A suggestion for some readers is to have semi-static damage that changes with the attack roll's margin of success, rather than requiring an extra roll. Account for the opponent's opposed roll/static defense first, of course.

Tri-Stat did this without a dice pool by tiering your damage. For the best result you could roll, you'd get a critical(200%). A couple of steps lower had your attack at 100% of the attack's original damage, and it stepped down from 75% to 50% and bottomed out at 25%. This approach had its problems, namely that without a variant rule it didn't actually increase your chances of getting at the higher damage tiers. It just meant the blows that would've missed glance instead.

The other system that did this was Shadowrun. Whether its resolution mechanic is better or not isn't for me to decide, but its implementation is much, much more solid. Basically, whenever you scored a successful attack, you'd do your weapon's damage code in either Physical or Stun boxes. For every two net successes with your attack roll, the damage code would scale to the next step. Because the game handled getting better at something by adding to the amount of dice thrown or reducing the number you have to beat to get a success, your skill would increase your net successes quite naturally. It was a very graceful way of handling criticals, too, though without knowledge the system itself is pretty rough to handle.

4 PurpleXVI

10:50am UTC - 10/19/2008 [X]

Another important thing is to make sure your resolution system for combat resembles your resolution system for non-combat stuff as much as possible. I know most systems do it that way these days, but REMEMBER IT, because it's kind of fucking bawls to have to juggle TWO systems.

Though, if your normal system would get kind of complex if used for combat situations, you COULD have a simplified system to make the combe fun and fast. Then again, in that case, why not just use that for normal interaction as well?

Personally I liked how Alternity did it. You always had a chance of scoring Fail, Mediocre, Good or FUCK YEAH as a result. With Fail being above your stat on a D20, Mediocre being 0-50%, Good being 50-75% and FUCK YEAH being 75-20%. (1-5, 6-7, 8-10 if you had 10, 11+ being FAILTASTICO) For some reason it confused people, but as long as you remembered to note down your intervals for each skill, it was workable.

Another thing to consider is that systems are more easily surviveable as complex and requiring much rolling in real life. I mean, Alternity, somehow it gets really bogged down online, but it works WONDERFULLY in person. Especially because people can more easily remind each other of rules, see how others do it, etc. Some stuff is difficult as fuck to put into words, but easier to show.

A system mostly designed for gaming online(say, over IRC) should generally be lighter and easier than one actually designed for the gaming table.

5 HiddenKrypt

03:12pm UTC - 10/20/2008 [X]

I'm in love with he FATE system and it's damage mechanics. It to has damage based on the margin of success against the targets defense roll. Player and target roll at the same time, look at the numbers, and then it's over. Next person's turn.

6 Issyl

06:02pm UTC - 10/20/2008 [X]

I don't know, i think that you can come up with most of your fluff before picking a mechanic, because obviously your system's feel can depend on the fluff. However, rolling mechanic is obviously the first number-related part of your game you should pick.

Also nice article Purple.

7 Ettin

01:30am UTC - 10/21/2008 [X]

THIS ENTIRE ARTICLE IS BAD AND PURPLE SHOULD FEEL BAD

8 PurpleXVI

01:37am UTC - 10/21/2008 [X]

I'm pretty sure there's a 99% chance that you're a cocksucker.