I’m now into alpha but I can’t help making some design improvements. In particular, I’m wondering about the length of the game. At first I had it set at 30 turns starting in 2011. That was too many turns, so I made each turn cover five years instead of one. That made the game a little short, so I extended it to run until 2050, making for eight turns.
Now, though, I worry that running until 2050 won’t demonstrate the real danger of anthropogenic climate change. By 2050, ACC will be causing problems, but nothing we can’t handle. If we don’t start doing something about ACC soon, though, by the year 2100 the world will be in deep doo-doo.. Perhaps I should extend the game to go that far. I balked at doing so because it extends beyond the lifetime of a likely player – why should they care? After some consideration, though, I’ve decided to proceed with this. The year 2100 may not be within the life span of the current generation, but it is certainly within the lifetimes of their children and grandchildren. And the likely results by 2100 are so catastrophic that I think they’ll make an impression.
So, how to handle such a big change? The simulation itself will need much more careful tuning; with a long time span like that, small variations become huge. Simple equations that yield plausible results over forty years go crazy after ninety. I may not be able to get it tuned that well. As it is, the do-nothing scenario has things improving for another ten years, then going to hell.
Perhaps I shouldn’t jump so far into the future. What if I extend the game only to, say, 2060 or 2070? Those years are still within the life expectancy of my audience, but further down the road. They won’t push the simulation as hard. Yes, that’s a good idea.
But that raises a new problem with the history display. Right now the bars, representing a single year, are pretty slim. If I try to squeeze any more years into those displays, they’ll get too thin to be accessible to the mouse. Besides, the color interior of the bars will be too small relative to the one-pixel black boundaries. No, I have to collapse the bars. The obvious solution is to make each bar cover five years. That might make the bars too fat. I think I can live with that, though.
Should I readjust all the coefficients in the simulation to handle a five-year step size rather than a one-turn step size? Right now the game uses turns five years long but runs each five-year turn in five separate steps. I think I should stay with the five-year step size, because that permits all the numbers to expressed in annual values, which more closely corresponds to what people already use. For example, it’s one thing to say that the total GDP of humanity is $60 trillion; it’s quite another to say $300 trillion over five years. Yes, I’ll keep it at five-year turns with one-year step sizes. But I’ll merge the one-year bars into five-year bars in the history display.
OK, I think I’ve made my decisions. To work.