Sunday, July 22, 2012

Sweet Dice of Mine

Vampire: the Masquerade has a great dice system. We all love 10 sided dice, but in case someone has never played PnP Vampire: the Masquerade, here's how it works:

You roll as many 10 sided dice as points you have in the feat you're checking. If you're firing a shotgun you'd probably roll Dexterity + Firearms, let's assume your character has Dexterity 3 and Firearms 2, then you roll 5 dice. If you're trying to hack into the police database you'd probably roll Wits + Computer. If your character has Wits 1 and Computer 4, your dice pool is 5. Each dice with a value equal or higher than the difficulty of the action is considered a success. If you roll 4 6 0 1 2 you've scored two successes against the default difficulty of 6 (the 6 and the 0, which in fact is a 10). However, each dice with a value of 1 cancels a success (in the example roll, the 1 cancels the 0). After all, if your dice roll ends up with at least one success your action succeeds (3 9 4 2 6). If your dice roll had no success and at least a 1, your action botches (5 4 2 2 1). Otherwise, your action fails. 

I was wondering if Vampire: the Masquerade dice system could work well in a video game. I've been reading a lot about the pros and cons of the system, its probability flaws and whatnot. Then I opened Unity and coded the dice roll. I run some thousands of simulations and entered the data in a google docs spreadsheet. Sounds like a ton of work but it actually took me less than an hour. 

I simulated many rolls with increasing difficulty and also with increasing number of dice. The results were quite good. Most of the time during the game you will succeed (82.2%). Sometimes, you will fail (14.1%), not too often but enough to make you realize it's not too easy. The unhappy 3.1% will be botches. Here's a pie chart to illustrate the point:

Of course, these are just simulations. If your actual gameplay experience is rolling all your botches in a row (strange but possible), we have a problem. So it's just a starting point.

No comments:

Post a Comment