I apologise in advance for the thread necro, but I'd just like to share some observations after having (finally!) beat the scenario on Deity recently.
Disclaimer: I'm not very good at CiV, so there was a lot of saving and reloading on my part, but the general idea should work.
Anyway, to begin: I finished on Turn 79 and followed Waywocket's strategy of ignoring happiness - except at the very start - and never razing. I decided to work on the following principles:
Cash is King: Rush-buying units and buildings is probably the best way to save time in building up your army. I used most of my actual production on defensive units. In view of this, my goal was to earn as much cash as possible. Therefore, I went for the right side of Commerce. Also, I aimed for Economics since I'd be able to get at least one GE in the course of the game (for Big Ben), and the +1 gold to TPs synergises with the huge puppet empire.
Upgrade Complete: Given the discount to upgrading from Professional Army, it usually helps to build up a decent-sized force of relatively cheap, primitive units (e.g. Triremes, Archers) and delay the relevant upgrade tech until you're ready to deploy them. It gets even better when you rush-buy them.
It's Supposed To Keep Them Out: Rebels spawn in the following patterns:
- Rebels will only spawn to your corresponding tech level. If you can't build Riflemen, they won't spawn Riflemen.
- They can spawn as strategic units (e.g. Swordsmen / Horsemen) but never ranged units.
- They come in groups of 2-6. (Possibly 7?)
Since you get the Great Wall once you take Beijing, sufficient ranged units should make short work of any melee rebels, with a small force of Pikemen to deal with the horsemen / knights that may spawn. So I garrisoned the cities with Archers and upgraded them to Crossbowmen once I had enough.
You Require More Vespene Gas: Take the cities with Strategic Resources first. (This usually applies when fighting any war.) A -50% Strategic Resource Penalty makes them hurt more than your -33% for Unhappiness. You generally want to cripple their Longswordsmen, which are usually the most annoying to deal with on the attack due to their high base Combat Strength. (This is especially useful when taking on Japan's Samurai.) The caveat is that I tried
not to pillage if I could afford it, since I didn't want to waste movement points in mid-combat or extra turns getting Workers to hook them back up.
The Enemy Of My Enemy... Is Toast: Another good general principle is that you should try to bribe your neighbours to fight each other so that their depleted armies will be easier to deal with. It helps at the start if China fights Jin (this is usually luck-based), but you usually have to bribe Persia to fight India. Arabia tends to fight Persia and Byzantium on its own initiative. The earliest time to do this is when you have sufficient extra Luxuries and / or Iron to avoid wasting your gold (in my case, after taking the Chinese capital). Try not to sell your horses unless you're sure you're going to get them back immediately (read: just before declaring war).
In view of this, I decided that my strategy would be to get the following techs: Metal Casting (for the specialist) -> Education (for the university) -> Economics (for the gold and Big Ben) -> Navigation (to take Japan) -> Biology (to reach the Industrial Era and get Autocracy if possible). This way, the rebels that spawned would be Swordsmen, Pikemen, Horsemen or Knights, all of which could be handled by sufficiently-trained Crossbowmen and Pikemen. The Archers would garrison the puppet cities in the meantime, giving bonus culture.
A summary of how my playthrough went:
- Started off by picking Professional Army and upgrading all Horsemen to Keshiks except one, which I sent to meet Western Xia. Farmed the riverside plain and built a pasture on the sheep. Built archers. First beeline: Engineering.
- Sent the Horseman down the Silk Road to meet Almaty, Persia, and (after detouring to the hills) India. In the meantime, declared war on Uighur and took it by turn 4. (Precise positioning required.) On puppeting Uighur, took Military Tradition and finished the tree. The extra gold from killing units helps to keep your economy afloat for the first 20-30 turns.
- Sent the rest of the army back down to Western Xia just as Zhangzong and Wu Zetian decide to declare war on each other. Scammed Zhangsong of his money, bought Sugar from Gandhi and a Circus. (Reason for this is that I needed to keep my happiness above zero so that I could get 6 pop in Karakorum). Declared war on Zhangsong and took Western Xia (which had conveniently allied with Jin).
- After taking Western Xia for the two free horsemen (upgraded to Keshiks as necessary), I went to assault Beijing from the hills in the northwest. While I was doing that, China took Xian.
- Took Beijing, then Liaoyang and Kaifeng in that order. This was by about turn 16 to 18 - I had completed Masonry and Construction by that point, so I used the free tech to bulb Metal Casting, and the money to buy a Workshop and a Water Mill. Ran a specialist. By this point, my aim was to keep Happiness more than -9 for the next step.
- Scammed Wu Zetian out of her money, then went straight for Hangzhou, since that would neutralise her swordsmen. Production was diverted towards building horsemen for the upgrade. Took Guangzhou and Xiangyang, whereupon she sued for peace and gave me all her other cities except Chengdu. This should be around round 23-25. At this point, I received a Social Policy and took the Commerce opener.
- I used my new-found Silk and Iron and bribed Darius into declaring war on Gandhi in the hope of thinning their forces down somewhat. In the meantime, I sold all my luxuries to them for whatever they could spare, and got the workers to connect the cities up to the capital for trade routes. Spent the money on Keshiks.
- During the 10-turn peace treaty, sent the first group of Keshiks down (three to take Dai Viet, three to take Dali), while my second strike force took Lhasa for Trade Unions (particularly useful since maintenance for the Chinese roads is no joke). During this time, rebels started spawning, so I sent the Dai Viet group of Keshiks back to deal with them. Fortunately I'd completed Engineering in the meantime, so moving around within my territory to fight rebels was easier. Subsequently spent a turn on Sailing (so that I could get Triremes), then beelined Education. By this time, I reached a positive balance in cash flow.
- Peace treaty expired around Turn 35; took Chengdu in 2 turns. Conveniently timed so that Education completed on the same turn and bulbed Banking. Annexed Guangzhou and Hangzhou, bought a University in Guangzhou (all that jungle!) and started building barracks. Workers went to TP the jungle tiles. Next beeline: Economics (I went with Physics before Machinery to get a couple more Archers out.) From this point on, all subsequent GGs spawned were sacrificed for Golden Ages. The next Social Policy went to Mercantilism. From this point, you can do what you want with subsequent social policies; the Liberty Opener is always a good one if you want a couple more policies before the scenario ends, or you could go for the Tradition Opener and Oligarchy for added security. For me, I took the Liberty opener and then finished off Commerce, which on hindsight was kind of useless.
- The original Keshik strikeforce went along the southern jungle and attacked India. In the meantime, the second army went up and took Almaty for the additional Horsemen, which were upgraded. Then I declared war on Persia at around Round 40. Around this point, I had Machinery, so I upgraded my home army of Archers to Crossbowmen and started churning out Pikemen from Karakorum instead. Additional gold went towards buying Armories in my production towns, while Hangzhou and Guangzhou occupied themselves with producing Triremes.
- The Persians are dangerous, as their Golden Age movement bonus combined with double-healing Pikemen means that they can catch your Keshiks off-guard very easily. Fortunately, the Great Wall meant that I could use puppeted Almaty as the staging ground to attack Persia as long as the enemy had to pass through my territory. Having said that, though, I spent about 10 turns sniping a neverending human wave of Pikemen streaming through Samarkand, since those hills posed a mobility problem.
- At around turn 50+, Army 1 flattened India at about the right time to bulb Economics with the free tech. I sent it through the southern pass towards the other Persian cities. The GE spawned at about the same time, so I used it to rush Big Ben (still took another 8 turns or so). In the meantime, Army 2 finally took Samarkand after killing nearly every man who could hold a pointy stick in Persia. Subsequently went for Urgench (since Nishapur is hard to take safely from the northeast, I left it for Army 1) then moved northwest to Russia. Next beeline: Navigation. In the meantime, I was rush-buying Triremes at a discount, and annexed cities as required to purchase Crossbowmen for Homeland Security.
- Army 2 took Vladimir, then Kiev. It's probably better to take Kiev before Novogorod so that you can deprive them of Knights, which will seriously cause trouble for your Keshiks.
- Crushed Persia at around Round 60, bulbed Navigation and met Arabia. Subsequently cheated them out of their money and used it to upgrade my standing navy of about 16+ Triremes to Frigates, all upgraded to Bombardment II. Sent them along with a lone Pikeman and bombarded Dazaifu into submission. (If Korea had allied with Japan, I'd have taken Korea first.)
- Took Abbasids and Georgia in one turn each before declaring war on Arabia, since I had to cripple their Camel Archers and I needed the roads. Aleppo was next due to its strategic resources. In the meantime, Russia fell and I sent Army 2 to Hungary. Hungary is tricky to move through due to the hilly forests, which eat up a whopping 3 movement, but I took it eventually and cut Byzantium off from their own source of Horse and Iron. In the meantime, the navy went to Kyoto first (Iron) - I managed to bring 8 ships to bear on the city per turn, so the rest went to bombard the ranged units and anything they could reach.
- The remaining Arabian cities fell easily except for Cairo, which is a major chokepoint. So I sent Army 1 across the Red Sea, Moses-style, and took it. Then I annexed the Crusader States (which Arabia had taken over previously) to get a suitable site to send out some ships. As for Japan, Kamakura's fall was inevitable once I got enough frigates.
- Finally, Army 1 attacked Constantinople, using frigates as spotters, while Army 2 occupied itself with sniping Thessalonica's units from behind (there's not much room to move around from the back, and the hills and rivers limit your range drastically).
tl;dr: Screw happiness, steamroll city-states, scam everyone, buy / upgrade plenty of Keshiks and Frigates, cheap-shot rebels with the Great Wall and crossbows / pikemen, and take over the world.