Game Dev Story Story

I had planned on spending a plane flight yesterday reading Kevin Kelly’s What Technology Wants and Dan Abnett’s Triumff. Instead, shortly after takeoff, I decided to give Game Dev Story a try. In the end, there’s not a ton of game to it, but it’s a blast to play through an alternate history of recent game development and create genres that never were (such as my experiments in the “Romance Puzzler” genre.)

One of the things I love most in the game is naming the finished projects from my studio. There’s a limit of 14 characters, but I still try to have some fun. Here are the first 30 games I made:

Fallout: New Vegas, Fable III, and Rock Band 3 are all in my near future….but I bet I’ll spend a few minutes every day helping my studio keep going.


Consider the simple, six-sided die.

Better yet, consider two of them.

To those who speak gamer, a six-sided die is a d6. Two of them is 2d6, natch. When many games ask a player to roll 2d6, they are asking the player to roll both dice and sum the results. The standard 2d6 gives a nice bell curve of probable outcomes, a curve peaking around a result of 7 that allows game designers to present players into situations with probable–but not certain!–outcomes depending on the target number the result needs to meet or beat.

I spent a lot of time thinking about rolling two six-sided dice while working on HeroClix. The core mechanic of the game was a classical 2d6 roll (modified by an Attack value versus a target Defense number scaled to match), but feats, special powers, event dials and more each had their own unique, new rules text and occasionally gave me a chance to play with alternative ways to use the dice.

When I did get a chance to play around in the design space, I was always pleasantly surprised at how many ways you could coax different results from a single roll of two six-sided dice. Some thoughts on a few of them, as I think about how I might use two d6 in a design I’m working on right now:

Pick a number between 1 and 6 and roll two dice; if the number comes up on either die, you succeed.

A 31% chance of success. If you pick two numbers that jumps to 56%, and three numbers 75%.

When you roll 1d6, every possible result has a flat and equal 17% chance of occurring. Even if the second die is rolled at the same time as the first, it will feel like a backup plan and give many players a psychological cushion that makes 31% feel like acceptable odds.

Roll two dice, take the highest roll as a result.

Another roll with psychological benefits to the player. Who doesn’t love throwing out a bad outcome and keeping a good one?

Looking at the table of probable outcomes for the roll shows how it puts the player’s thumb on the scale and tips the results toward their favor. More importantly, it shows how provides the same range of results as 1d6 roll, but on a curved distribution rather than the flat distribution of the 1d6 roll.

Roll two dice, and subtract the lower die from the higher die to get a result.

This gives as narrow a spread of results as rolling a single die, but what makes it interesting is that zero is one of them. So it could be interesting to use as a means for generating a modifier (“You get extra points of damage equal to the result”) when you want “no modifier” to be a possible outcome.

Roll two dice; a result of doubles is a success.

Of the 36 possible outcomes of this roll, only six are successes. 6/36 is equal to 1/6.

I hear your question across the Internet: So why not just roll 1d6 and say only a 6 is a success, Mr. Game Designer?

Mathematically, no reason whatsoever. Again looking at the roll psychologically, I think there’s an argument to be made for rolling doubles as a more ‘exciting target’ than rolling a 6 on a single die.

Roll two dice, and multiply them to get a result.

The minimum result of this roll is 1 (from rolling two 1s) and the maximum result is 36 (two 6s).

Interestingly, though, within this spread of 36 there are only 18 possible different results. Despite its prevalence in the normal 2d6 roll, 7 will never be a result here. Nor will any number between 26 and 29. If it can’t be factored completely using only the numbers between 1 and 6, it will never come up.

But check out the results that do come up! A maximum result in the 30s! 18 different results when 2d6 can generate only 11!

The chart of possible outcomes only appears to be a sawtooth when you’re charting the number of times each possible outcome occurs, as numbers with multiple results as factors like 6 (1,2,3,6) and 12 (2,3,4,6) spike above the rest. Yet if you click through that link and scroll down a bit to look at the curve generated when you look at possible outcomes through a greater than/less than lens, you get a reasonably smooth curve with more gradations than 2d6 thanks to its larger number of possible outcomes (though there are larger intervals at those same multiple-factor outcome points.)

Often games that use d6-centric mechanics have a tiny number of possible target numbers; when rolling 2d6, if you stray far from 7 you have large effects on the possible outcome. If the roll is an ability or skill check (or an ability plus skill check), it will likely be difficult to improve your abilities or skills, lest you slide down the probability curve.

Rolling 2d6 and multiplying to get a result versus a stat or (stat + skill) as a target number could be interesting in that it would allow for a much wider range of character advancement without breaking the curve.