I'm really a DemiGod player myself, I still regularily lose on Demi if I get a bad start or if I mess up
In regards to the Atilla game I chose them not for the Horse Archers but for the Battering Rams. I grew my capital really fast and got an early library to speed up Mathematics (where Rams are now) so I could rush them. I figured in such a crowded map they would be really useful. My plan was to take two capitals as soon as possible. I went Liberty left side, settled five Cities on my own and then switched over to unit production. Assyria had the biggest military of all, so my decision to attack him first might have been bad in hindsight. He even got Terracotta army just a few turns before I DoWed. Regardless, after twenty turns of zerging I managed to take his capital with Rams and Horse Archers. I razed two of his cities and left him with one. Finished Liberty, moved my army north to Korea for an ambush. Took their capital in less than ten turns since his units were stuck somewhere else. Really disappointing though since he had no wonders, his capital was very low population and had no useful buildings whatsoever. That's it for my write-up.
As for the opening turns - I think you're completely right, they're by far the most important turns in any game so I will be going a little more in-depth:
Surviving the Early Game
With the mod doing city state quests is so much easier. They are also more important, especially mercantile, cultural and militaristic ones. This should always be one of your main focuses. In this section I mainly want to discuss how to deal with "angry neighbours". Whether or not the AI will declare war on you is dependent on a few factors: a) Their pathing, meaning your cities and the terrain you picked b) Their army score and your army score c) World diplomacy. A few tips: Settling "worse" cities for a good strategic position is almost always worth it. Even if you are the tech and military leader you will still get DoWed at some point in the game. Better be prepared. Settling on coast is always a risk, as is settling on flatland. Archer units used to be the prime defensive and offensive unit, but they're not anymore. Melee units, in this case warriors and spearmen boost your military score by a much higher value. Even if they were worse at defending, the idea is to get the AI to never even declare war on you because it will definitely be a war of attrition that slows you down heavily. So you want to get your military score up early which means building units early on instead of buildings like the granary, water mill, library and shrine. This has the added advantage of being able to tribute city states, especially when you have a quest for it. You get gold or potentially a worker out of it. Speaking of workers - try not to steal from a warmongering AI, they'll never make peace again. Stick with the peacemongers or science civs and slow them down. Use all of your units to do city state quests, tribute and buy more units. Having a decent standing army is important. Settling cities later than in the unmodded game is possible, so if there is a key wonder you want desperately just go for it. Most of the time it will work out. From my experience: Even at DemiGod wonders like Pyramids, Hanging Gardens, Petra and Temple of Artemis sometimes go on T60, Oracle sometimes goes on T80 or later. Great Wall is also a decent wonder to invest in.
Tradition or Liberty?
In most games this is the biggest decision you will be making. There are a lot of factors going into this, but in my opinion it is quite different from the unmodded game. First off, four city Tradition is no longer a viable tactic for peaceful victories. You are very likely to lose a science or culture victory to someone with a bigger empire. Four city Tradition is however viable as a domination strategy - you focus on nothing but science until you get to a key military unit like Artillery, Bombers, Atomic Bombs, XCom et cetera. So what does this mean for peaceful victories? Even when you are going Tradition it is heavily advised to settle at the very least five cities, preferably six. In the unmodded game it is rather easy: If you have space for four cities you go Tradition, if you have space for six or more you go Liberty. In AckenMod you need space for five or six cities that are at least decent to have a chance of winning a peaceful victory condition with Tradition. Tradition empires have faster border growth, need way more workable tiles. The penalty for wide empires has been reduced heavily, which means Liberty can both settle cities closer and settle cities with less good tiles available. What this all amounts to is that
Liberty and Tradition essentially need the same amount of space. So now going Liberty or going Tradition is finally a real decision. Liberty however still relies on having a lot of luxuries early game since the happiness policy is so much worse than Monarchy is. Cues for going Tradition would be: Mountains, abundant rivers, lots of grassland tiles, few tundra or desert tiles. Cues for going Liberty: Lots of great production tiles, pastures or camps, some tundra, desert, mountains (not next to your cities, mountains taking up workable tiles). If you are in a situation where there is only space for three or four cities in total the only way to win, be it peacefully or domination, is to either go Honor or go Tradition and do a timed push. Without getting at least one capital your chances are looking pretty bad.
Tradition Early Game
Tradition has received an indirect buff because Acken made culture progression a little bit faster, which helps Tradition a lot. Getting the Aqueducts up earlier in your expos is key. So is allying cultural city states. For Tradition I always skip the Monument and go for a solid Scout / Scout / Scout or Shrine build order. The focus on your scouting should be
where you settle your four core cities. The AI plays very aggressively, so you want to settle your first three cities rather far away from the capital so that you still have space for one or two good cities in between. This is really important because
your border cities now get free walls. Abuse this. If you are starting next to Atilla, Caesar, Shaka, Harald, Oda, Genghis or anyone else with a warmongering focus it can be a good idea to get the policy for free walls if you see an army next to your borders, even if it delays Monarchy and Landed Elite. Caravans and cargo ships aren't nearly as important. It is not worth to settle coast if there are only few good water tiles. I often delay my internal trade routes until I get at the very least three food from caravans. Sometimes longer. Going for a four city national college should still be your priority, but as I mentioned earlier it is advised to settle one or two more cities after that is done. The national college is also far weaker than in the base game so delaying it is not nearly as problematic. Your build order is much more free. After going for the national college it is definitely viable to get workshops before going towards universities. Both options are equally viable I feel. Since trade routes are weaker your cities will be smaller, meaning universities have much less of an impact.
Liberty Early Game
Liberty works fantastically on its own, but it also works well in conjunction with Piety. This holds especially true if you have a civ with a shrine or temple unique building. As mentioned earlier you can
settle cities much closer. This is incredibly important because early roads are very costly, so you want to get them done asap while still being profitable. It will also increase your worker speed by a huge amount. Workers for early tiles improvements - pastures, camps, hills - should be the focus of your Liberty empire. Personally I will only go Liberty if I will self-found at least six cities. Liberty lacks the early walls that Tradition provides, but it has nice production bonuses that you will be using to pump out some units inbetween key buildings. Be very careful not to piss off your neighbors, Liberty cities are typically less developed and therefore harder to defend. When playing Liberty I almost always have a worker as my second build in every single one of my expo. I try to get eight or nine workers by the time I settled my last city in order to abuse the worker speed bonus as much as possible. Old strategies like Liberty composite bow rushes still work to an extent, but are much harder to pull off. On the other hand a civ that has a swordsman unique unit will benefit greatly from Liberty since a mixed army of swords and composites can easily nab you one or two early capitals. A prime civ for this would be rome or greece. The national college is not nearly as important and can be delayed until T100 although that is not advised. Your capital will probably be rather small on T100 so with Liberty I'd say workshops before universities is almost always the way to go.
General Science Advice
If you have six or seven cities constantly growing your chance of catching up to the AI is much greather than with four or five. Being eclipsed in research is not a problem. I usually catch up way after public schools, if science civs are involved it may take til after research labs. Never forget that, if you are going for a peaceful victory condition, you can slow down your opponent by starting a war. Even if you are not trying to win, your opposition will likely build some units therefore slowing him/her down. One thing people tend to pay little attention to is spies. They are hugely important in this mod, a triple promoted spy can technically get you a technology every few turns. From carlsguides section on spying:
Spy Steal Rate is directly related to the amount of Scientific Output a City has, related to the costs of Tech that Civ has available to research. Thus, big Cities of Civs with a high Science output will be the best targets. This is why the turns to steal a tech vary so wildly. If you see 40+ Turns to Steal, you are likely targeting a small City in a wide Civ whose Science is evenly distributed around its Cities. Small Cities of Wide Empires make the worst targets, even if that Civ is technologicaly superior. Move on to another City with a higher Population and greater Scientific Potential, else move to another Civ entirely. As a user on reddit.com/r/Civ pointed out to me, we can use this both offensively and defensively. When you have only cheap techs to research that can be completed quickly, your Cities will be very high potential targets because you're able to research those techs in just a few turns.
I could do a write-up on Honor or Piety but truth be told is I pick Liberty or Tradition in close to 90% of my games. Both are viable, but I think Liberty or Tradition are almost always superior.
, I can look into it by integrating what another AI modder has done.
