Thou shalt not throw excessive amounts of dice.
As far as I am concerned two dice is about right for the standard maximum to roll. Three already is a bit too much. Although I always enjoyed GURPS less would have been more.
And then there is my next corollary:
Thou shalt leave room for development.
Which basically means that I do not like very narrow attribute ranges. I want to able to develop characters and one of the best ideas to be taken from D&D 3rd edition IMHO is to be able to raise attributes as characters get more experienced. This makes the game less static and gives even weak characters something to strive for. But this doesn't work too well with narrow attribute ranges because advancement again becomes a pretty static things and I also like the idea that every new level (in a level based game) should offer excitement and new experiences.
Cyborg Commando (also a post apocalyptic game with a very different kind of apocalypse) that - as far as I know - was the first game to multiply two dice in order to find a die score. Cyborg Commando used 2d10 (probably due to the pretty neat distribution spread), for Gaia Gamma I am going to use 2d6. For all major tests, that it.
The notation will be 2dx which means that you roll two six sided dice and multiply the results to get the final score. E.g. rolling a 2 and a 5 yields a final of 10.
The final spread ranges from 1 to 36 (although quite a few numbers can't be rolled at all). The latter is ok for me as this leads to the situation that not every attribute increase (or skill level increase for the matter) will lead to an actual improvement. Even better - the higher a score gets, the fewer numbers can be rolled. While I do not want to discuss the merits of the actual spread (is it realistic or not? who cares in a world of mutants?) I like it's general makeup: Wildly varying results can be occasionally achieved but the main distribution still centers around 10, which is nice for the classical attribute range of 3-18 (and ensuing attribute tests). As a consequence skills also will be centered around an initial 3d6 throw (at least for primary skills - more on that in a blog post to come regarding areas of expertise).
The chart below shows the actual distribution of rolling a specific score or less with 2dx:
One interesting problem arises from the actual number spread on the chart below: It becomes very hard to define meaningful difficulty modifiers if you go for a "roll under or equal to" system - which I intend to do as I believe that doing something more difficult after already being required puts too much strain onto the player.
But I have what I believe to be a good idea about that - more on that in tomorrows post!