[SPOILERS]Luddite: A Man, A Plan, An Also-Ran

[luddite]

Gotta get my spoiler thread up so Mist can give us our starts.

[luddite]

I actually do have a secret plan for this game. Just waiting to see the start before I type it up.

[pindicator]

Are you calling GES an Also-Ran?

[Gold Ergo Sum]

Are you calling GES an Also-Ran?

Sounds about right.

[luddite]

Hmm, a lot of discussion in the map thread and still no screenshots of our starting position. My guess is that Mist put in some evil twist on the map.

[Mist]

Not evil, just interesting

This is a very good and well behaved map

Start :

Caveat : Some stuff might still change

Pick related map info :
-Noble
-Cylindrical wrap
-Somewhat cramped

[luddite]

I'm still on a high from seeing the win in PB6 (which so far is the only victory I've had in a multiplayer game here), and since this game hasn't started yet, I guess I'll post some more random math thoughts I had about civ.

When I was doing research for this thread, one thing that really jumped out at me is that all the demos in civ are locally exponential. That means that if your food and hammers have been increasing 5% per turn for the past 10 turns, they'll probably continue increasing at that rate for the next 10 turns unless something major happens to drastically change the game (which usually means a war, or running out of expansion room. You'll know it when you see it).

I went a little crazy in that thread trying to find a way to calculate what rate food, hammers, and beakers would increase at. If you could figure out that growth rate, you could come up with exact calculations to make the sort of long-term vs. short-term tradeoffs that are often so difficult to do in civ. I did come up with some interesting (to me anyway) results, but ultimately nothing that I would trust more than regular game intuition.

However, when I was thinking about how I wanted to play this game, I had a much simpler idea. You don't need to calculate that growth rate. As long as it's locally constant, you can just measure how fast you were growing for the last 10 turns, and that will tell you how fast you can expect to grow for the future. You don't need to bother measuring your micro skill or the map settings or anything else, because it's all adjusted for in that growth rate anyway.

If this all seems totally pointless and disconnected from any real game well... you're probably right, but bear me out.

Suppose you want to decide whether to use or Great Scientist for a bulb, or settle him? Once you know your tech growth rate, you can discount all research into the future at that rate. Add up all the turns of research it'll provide, with each future turn being worth less and less. It becomes an infinite convergent series, with the sum of initial value/growth rate. If a settled great scientist in a city with no buildings provides 6 beakers +1 hammer, and we say that 1 hammer is worth 2 beakers, and that our research growth rate is 4%, then the value of a settled great scientist is 8/.04 = 200 beakers right now. Nice and simple.

Admittedly, a real game isn't quite that simple- the value of a settled specialist will go up over time, as the city adds more infrastructure. But unless those buildings are going up very soon, they'll be discounted so much by this formula that it hardly changes anything. And I think that's reasonable- it doesn't make any sense to settle a scientist and assume "well in 50 turns I'll have oxford here so it'll become good then". Academies work the same way- they'll grow a bit over time, but most of their value is what they can do for you right now and in the very near future.

So how does that compare against a bulb? Well, a bulb gives you all of those beakers instantly, of course. But on the other hand, it's usually not a tech that you would have researched immediately if you had a choice. For example, bulbing theology with a prophet can give you a lot of beakers very early on, but nobody beelines theology with normal research because the payoff is so delayed. So for a fair comparison in this method, I think you have to discount the beakers of a bulb in the same way. Estimate how many turns in the future you would have researched that tech normally, and discount the value appropriately.

Doing it this way, both bulbs and specialists depreciate at the same rate, and you end up with a handy rule of thumb: if the tech that it bulbs is something you would have researched within the next 20 turns anyway, bulb it. Otherwise, settle. (the math of this is that, assuming an 800 beaker bulb, 800/1.04^20 is about 200. Actually it's more like 23 turns, but 20 is close enough) As you get more infrastructure in your cities, the value of the specialists increases a lot, up to a maximum of 33 for settling a representative scientist in a city with an academy, library, university, observatory, forge, and oxford. At a 4% growth rate, that's worth 33/.04 = 825 beakers immediately. The value of bulbs also goes up, but not as much. Note that knowing the tech growth rate doesn't actually change the decision here directly, since it affects specialists and bulbs equally- however it DOES determine how many turns in the future you can expect to research a particular tech. Overall, faster research should encourage more bulbing.

I'll probably still trust experience and instincts more than this type of math (in particular, this doesn't account for the value of winning liberalism or key wonders) but I think this is getting to the point where it can actually give some useful results. We'll see how useful it is, in this game.

[luddite]

Oh... and as I'm writing all that, Mist posts the map.
Wait, do we really not start with any units at all? Or is that just to show the start.

[Mist]

That's just to show the start. All proper units will be placed on the "Starting Tile".

[ironstar]

Interesting thoughts about the settle / bulb dillemma. So in the case of a GS, whose bulb represents significantly more beakers than any other GP(erson), the rule of thumb will be skewed towards choosing a bulb, right? Or am I not fully understanding your thinking here (bear with me)

EDIT: of course he also represents more raw beakers than any other GP when settled. Maybe I should start to take a minute to think before I post.