when to start getting pyramids for an SE?

[DrPepper836]

Thanks for the tips everyone. I played 30 turns more, and am starting to pull ahead. I have a big technological advantage, and I just captured a city with horses. I'm losing 9 gold per turn at 90% research. Sparta will soon build an odeon, and I'll get a GS from there soon enough. Unfortunately, it's water was poisoned so it shrank some. I explored the rest of the continent, And didn't find much more. My maps of shaka's land are a little out of date, but I'll manage. All of the cities I'm capturing are killing me. How should I continue the war?

 

[Ultimocrat]

So, I usually tech straight to IW, then agriculture, AH, then beeline CoL. When should I get pottery?

The fact that you have such a set teching order suggests part of why you are having difficulty at low levels. Play in Civ4 is almost above all else, situational. Worker techs are usually but not always a top priority, and they are highly situational -- you don't always start with the same terrian and resources in your capital and surroundings. The best play is, to a large extent, a strategy that has synergies with the leader you choose, and with the map you've drawn from the RNG.

E.g. It's probably foolish to go for the 'mids when you're aggressive/financial, start on a river with floodplains, and have an industrious neighbor with stone.

 

[Indiansmoke]

Cottages!!

Yes Pyramids goes well with cottages...

 

[Roxlimn]

Percentage research is not a good way to gauge research prowess. Never use it as a measure of how well you research. The only way to measure research power is with beakers. At Warlord, you ought to be researching stuff at around 5-10 turns, 10 turns max. If you're researching faster than that, you may want to expand more, since you'll be able to afford it and it'll snowball your research power in the future. If you're researching slower than that, you'll already too small, or your economy is tanking.

As a rule of thumb, during the mid-game, research power at 60% slider is a good benchmark to look at.