I noticed in this HOF emperor level game (standard size map, pangaee, http://hof.civfanatics.net/civ4/files/included/413_1217_Shillen_Rome_Julius_Ceasar_Emperor_Standa rd_1_52_AD1440.Civ4SavedGame, Caesar has gotten a domination win in 1440 AD using purely massive farm (all specialist, pyramid+representation) at the end, and his research is pretty respectable (3 or 4 turns per tech). So I am wondering if this would be the way to go for post-classical era domination victory. Obviously some cottages would have to be built in the beginning to offset the expansion cost, but it would be purely farming and windmill after code of law and civil service. One way to obtain those two expensive techs would be via Oracle, but if that is not possible, then should one beeline for those two techs or take a more balanced approach. What are best ways to increase the health+happiness limit fast? What conditions do you look for to know that time is ripe to go conquer your neighbor(s)? I can deduce there would be rampant chopping early on to get those praetorians out fast, so maybe this approach wouldn't work as well on resource poor maps (i.e. highland, great plain). Is it a good idea to turtle until your UU arrives? For example, samurai/cho-ko-nu/conquistadore conquest? What are your thoughts on this?

I have been debating the importance of specialists to myself lately. Lets say, for example, that we are going the cottage/town route to handle our research needs. A fully developed grassland is worth 2f/0s/7c. That is the ideal, you farm only enough to work every tile and maximize the number of cottages being worked. If your happiness limit is 14 you would ideally be able to earn, 14*7 = 98commerce from all grassland.

Now lets look at the specialist method. Typical grassland farmed, 3f/0s/0c. For every 2 farmed grasslands you could support 1 specialist. Running representation that specialist gives 6 beakers, very comparable to the 7commerce a town provides. For a happy limited town of 14 farming all grassland you could have 9 farmers and 5 specialists. 9*3=27 food and 5*6 = 30 beakers from all grassland.

98 commerce compared to 30 beakers, seems like it is a landslide victory for the cottages for an all grassland scenario. So when would we be better off using specialists instead of cottages? Lets take a city with 9 floodplains. Lets farm all 9 floodplains for 9*4 - 3(floodplain health negative) = 33 food. Now lets crank our culture up a bit so we can pass our normal 14 happiness modifier and get up to 17 population. We now have 8*6 = 40 beakers. 40 beakers still falls way short of 98 commerce unless you are forced to run at 40% science.

Now why would this guy run specialists instead of cottages? There are a few reasons i can think of, #1 being that population plays a significant role in your end game score so boosting pop up will boost your score. #2 it is easier to transition to a production economy from a food economy and vice versa. Cottages take a very long time to mature into towns which leads to #3, cottages are vulnerable to pillaging while farms are easily replaced.

Considering, I think the main reason was to boost his score up as high as possible. Cottages and towns are simply far superior.

Edit: If you are aiming for a very early game massive conquest then the specialist approach might be ideal. Early towns only earn 4 commerce while the specialist would still earn 6 beakers. You could quickly assimilate captured cities because you wouldn't need to wait for towns to mature. Also a research city could double as a very high production city. A good cottage city requires lots of tiles to cottage while a good specialist / production city just needs very high food output from a few tiles leaving the other tiles to be shield production. Hrm I am going to experiment with this tonight.