Agriculture and a worker first were quite obvious. The hut popped Hunting on turn 2, which I had tested ahead of time before releasing the scenario; if it had been something very swingy like Mysticism or Bronze Working, I would have removed it.
There is always an open question about the general approach to expansion. You can go for a granary to whip settlers, or slow-build them which is less efficient but doesn't incur whip anger and can put off the granary. Here I decided to take the former approach. Settlers are more valuable when the cities cost no maintenance, of course. And Timbuktu had the right amount of food surplus, only one resource plus a few food plains, in order to be able to regrow size 3 to 6 about every 10 turns.
Bronze Working was my second tech, in order for the first worker to chop out the second. Build order was worker - worker (chopped) - scout - warrior (police) - granary (double chopped) - warrior (escort) - settler (whipped 6 - 3) - warrior (explore) - warrior (killed) - warrior (escort) - settler (whipped 6-3 again) - worker (with overflow and a chop).
I had the opportunity to steal a worker from Carthage. With no tech trading, the diplomatic hit is less important. But I passed, mostly because I was playtesting, and also because getting the worker home through barbarian territory would have been chancy.
City two, at an extremely strong site.
Well, duh. Heh.
Not much of interest happened in the early game, so let's skip ahead to an overview at 1000 BC.
I had settled a city for the marble, in order to make the Oracle - Civil Service jump as usual. Except that I wasn't keeping track of game time, and Egypt grabbed the wonder several turns before I could! I had figured Oracle-CS as a slam dunk with this sort of variant, but that wasn't true at all, not at all like the silliness we did in Hannibal's Muse. Notice the gold count of 4999741 in the picture. That means I've spent only 259 gold on deficit research, or only about 5 extra turns worth. Not at all a crazy cheaty advantage. Anyway, I just kept on researching to CS the normal way.
Djenne with its big food whipped a library and ran scientists as usual, and the Academy plus Bureaucracy arrived right at the same time. And yes, I know Timbuktu has way too many forests still standing. They quickly got chopped into workers and settlers with the Bureaucracy bonus.
I had intended to grow Timbuktu like crazy, using Hed Rule police for happy, since their maintenance cost is not an issue here. But it ran into an unexpected problem: the health cap! I hadn't really thought about health resources when making the map. There were several more available - rice, banana, deer - in fringey locations that took some effort to connect. I hadn't planned the map that way but it came out with some decent strategic options there on its own. Anyway, I still needed a bunch more workers and stalling Timbuktu's growth to produce them was the best answer there.
