Tearing Up the Tech Tree - Science Strategies for Deity Play In the .217 Patch

Martin Alvito

Real men play SMAC
Joined
Sep 23, 2010
Messages
2,332
After significant experimentation, I have found several powerful approaches for attacking the tech tree on Deity in the .217 patch. Not all of those approaches are commonly known at the moment. Babylon's ideal game plays much like pre-patch Siam, but that approach is no longer the unique solution. The three approaches I have found share a common goal: unlocking Education and being ready to build Universities or Wats around turn 70. That, aggressive Research Agreement usage and Great Scientist bulbing will get you to a win condition tech (Globalization/Apollo) between turns 130 and 145.

I'm assuming that you will play the diplomacy game well enough to avoid getting killed, and choose to do this rather than invest :c5production: in building an army for self-defense. An approach like Tabarnak's multiplayer strategy (http://forums.civfanatics.com/showthread.php?t=415715) will enable you to defend yourself with units, but it is much slower at getting through Medieval.

The remainder of this post will share these new approaches with you. All approaches assume the aggressive use of Research Agreements starting somewhere between turn 70 and turn 100. Some require RAs around turn 40 to get to Education, and I will point this out when necessary.



(OCC) Patronage Gambit (Arabia, others)

This one will work with anyone, but it's cash-intensive so Arabia is the best choice. The concept is to grab both the National College and early Scholasticism in order to have upwards of 50:c5science: per turn starting in the 60's. This tears through Medieval rapidly, so you don't have to bother with Research Agreements for quite a while. However, it is vital to get to Medieval before your fourth policy selection, and that can be challenging if you go NC first. Fortunately, the .217 patch introduced a consistent way to do this for all civs: Meritocracy.

By taking a Great Scientist (or Great Engineer if the Great Library is still available) from Meritocracy, you can bulb a Medieval tech before turn 40. This permits you to then binge on early Cultural allies and get Scholasticism up as quickly as possible. Reaching Medieval also increases the amount of Culture you receive from allies, so there is some strong synergy there.

The fun thing about this approach is that it is well suited to an OCC. Attached are a screenie and start/end saves for a turn 232 Arabia OCC win that I hammered out in about four hours despite forgetting to turn quick combat on. With quick combat on, you can probably bang out a normal speed game in under three hours.

The gameplan runs as follows:

1) Move and settle a luxury if possible. This costs you a doubled luxury if you're Arabia, but if you have several copies this will work out fine. Failing to do this can give you trouble at step #3. Your SP beeline is Liberty -> Citizenship -> Meritocracy.

2) Go Pottery -> Writing first unless you have a strong Mining start. You won't need a quick Worker unless you went Mining, so Scout -> Monument is the choice. If you are going for Mining, I'd recommend Scout -> Worker, especially if you settled on a Hill. (Going for Mining means that you're going to throw more income at a Cultural early on rather than bother with a Monument.)

3) Once you have Writing, sell Open Borders to everyone that is "Friendly" for 50 gold. You now have your first predicament. If you hit a Culture ruin and built a Monument, you will have a Worker soon and you can borrow from other civs (preferably at 45:c5gold: per 2:c5gold:/turn) to build a Library. If not, you need to make a Cultural friend with 250:c5gold:. Try to anticipate this situation, because on turn 19 you get 5 fewer influence for 250 :c5gold:, and you need ten turns of buffed Culture rather than five if you have to go that route.

4) Build the National College in the capital. In a perfect world, you tech Animal Husbandry -> Trapping -> Philosophy now. However, most of the time your luxuries are Calendar luxuries, and you need Calendar to hook them up. If that's the case, you need to research Calendar -> Philosophy. Deploy your Worker(s) to the luxuries and hook them up, then resell them for as much cash as you can, remembering to take back :c5gold:/turn if you must.

5) You need a Cultural ally to get to Meritocracy, and the sooner the better. You need to throw 500:c5gold: at a Cultural the moment you get it. Borrow if you have to.

6) Once you get Meritocracy, take the Great Scientist. Often times, you will get Meritocracy before Philosophy. If you don't mind save/reloading, you can take the Great Engineer to build the Great Library if it is still available for the GS point. In most games, the Great Library will be gone by the time you finish Philosophy.

7) Bulb/GL Theology (Calendar) or Civil Service (AH -> Trapping), depending on your tech path.

8) Keep hooking up luxuries and buying up Cultural allies as fast as you can. Get a Maritime as well. Sell every luxury you hook up; your allies will give you more than enough luxuries back. Unlock the left side of Patronage to Scholasticism as you get policies.

9) Once the National College finishes, the game opens up for you. If you want to make Settlers and grab the luxuries that are still out there, feel free. If you want to stay as an OCC, you can also do that if you have a lot of luxuries in the starting radius or are playing Arabia. Your :c5science: will be fine either way. The tradeoff is acquiring more luxuries, and therefore cash, against getting Patronage and later policies faster.

10) Tech strategy will also vary by civ. An Arabia player should beeline Currency after Philosophy, then go for Education. Be sure you have 580:c5gold: on hand when Currency finishes to buy the Bazaar, then resell your duplicated luxuries to cover the cost. Most other civs should either go straight for Education or go for Metal Casting, then Education.

Attached is an OCC with Arabia that finished on turn 232 in a single sitting without save/loads. It's a pretty clean game without major errors, although it could have been a few turns faster if I'd anticipated the aggressive late efforts to buy up my city-state allies and planned accordingly. Having more civs actually need Wine also would have been nice; the Bazaar wasn't totally useless, but it wasn't as effective as I'd like.



Legalism Abuse (Siam)

Here, the idea is to exploit Siam's ability to drop in Wats instantaneously. This requires some micromanagement and forethought. You have to time the acquisition of Legalism to match up with when Education and Temples are both completed, so that you receive Wats from Legalism.

You need to produce a total of 160:c5production: in three secondary cities to complete Monuments and Temples by the time Education finishes. You also need to avoid improving Wine or Incense, or you'll have to buy/build a Monastery as well. Finally, you'll need three Research Agreements for Theology, Civil Service and Education, because early :c5science: output is going to be very poor. You'll be late to the Renaissance, but with only four cities that isn't the end of the world.

The :c5production: problem is the big barrier. For that to work, you'll need to get cities online quickly and working high :c5production: yield tiles. A city working the base tile (non-Hill) will need to minimally be size 2 and working four :c5production: with citizens to complete this build series in 32 turns. However, having Wats in a size 2 city is fairly pointless, so you're going to want Maritime allies and you're going to want to work improved tiles such as Sheep or Horses rather than Hills wherever possible. That way, you can fill the Wats and have functional cities afterward.

Settlement and build strategy are designed with these factors in mind. The last city you found must be a city with tiles such as Horses and Sheep, and those tiles must be improved quickly or you'll be out 580:c5gold: for a Temple. Worker turns are a serious factor, so settling luxuries is a good idea whenever possible. Finally, you won't have time to build more than two Workers in the capital, so you're going to have to steal some. Stealing one directly from a city-state and raiding a barb camp for another are pretty much mandatory, unless you get a luxury-heavy start and can get away with buying a Worker.

Here's the gameplan:

1) You have a choice on tech ordering. If you don't mind gambling, tech luxuries and Animal Husbandry and hope that AIs tech Philosophy early. If you don't like risk, pick up either Mining or Calendar, then go for Philosophy.

2) SP order is Liberty -> Collective Rule -> Tradition. Throw 250:c5gold: at a Cultural city-state as soon as possible (should be around turns 10-14) in order to get to Collective Rule. Borrow as soon as enough :c5gold: is available to make this happen. Send the free Settler to the best multi-luxury site you found while scouting. Distant is better than close; fill the closest good site with the last Settler.

3) The best build order is usually Scout-> Worker -> Settler, so that you're not wasting turns building Settlers without Collective Rule available. This will also generally let you grow to size 3. Use your judgment on whether you want to build the Worker or go Scout -> Settler or Scout -> Scout -> Settler. Be aware that you need to steal in a hurry if you skip the Worker.

4) Hook up luxuries and resell for cash. You need three Research Agreements. You're also going to need a Cultural ally at some point to get Tradition and eventually Legalism, and you're going to want Maritimes so that secondary cities grow while building. Those Research Agreements should be in place by about turn 40 if at all possible; prioritize them.

5) After the capital has built two Settlers (in addition to the free one), build a Worker if you still need one (send the capital's Worker to improve the last, nearby city), then start on the Monument and Temple. Secondary cities should build Monument -> Temple upon founding.

6) Research needed techs, then block unneeded techs by researching them to 1/3 completion. It's usually best at this point to block Masonry, Bronze Working, The Wheel and Sailing unless you have Marble or Pearls/Whales. You need to research Calendar, Writing and Trapping, which implies Pottery and Animal Husbandry. You will almost certainly need Mining. Archery can be blocked or researched. If you end up with turns to spare (due to ruins pops), invest them in Sailing and block Optics. If you were insanely lucky with ruins, take Optics and block Compass.

7) Use the table from the Number Crunching Thread in the strategy guides section to know when you need to add Cultural allies (if at all). You want to take Legalism just as Education and the Temples all finish.

8) Staff the free Wats with Scientists. You now should have a respectable :c5science: output per turn. After this, you need to get up to the Renaissance. Astronomy is going to be the best way to get there. Hopefully, you had enough cash from luxury resales to sign a fourth Research Agreement well after the others so that you could research Compass, then get Astronomy from the RA. If not, you'll have to research it the hard way.

9) If you binged on Cultural allies, you'll probably have to take another policy before the Renaissance unlocks. You can do one of two things. Either you can grab Patronage and go for Scholasticism, or you can take Landed Elite and then go for Democracy or Secularism once the Renaissance unlocks. The former makes sense if you had a cash-rich situation where you have a lot of allies. The latter makes more sense if you're cash-poor. It's worth noting that Democracy is a lot better for Siam than any other civ due to the early Wats, and if you can get it in a hurry you are going to have a lot of Great Scientists for the late game.

10) Now that you have decent :c5science: production, it's also time to clear all those cheap techs you were neglecting. I'd suggest getting up to Metal Casting and Machinery as soon as possible for Workshops and the Ironworks in the capital. Credit goes to vexing for this idea. While you do this, it's time to load up on Research Agreements. Thirty turns of Wat output backed up by a half-dozen Research Agreements should carry you deep into the Renaissance. You should tech horizontally, then let the Research Agreements do the heavy lifting. I'd advise blocking at Gunpowder to try to get maximum value from the RAs, which should push you to Scientific Theory or very close to it.

11) You'll want to deal with two problems. The first is the :c5happy: situation, and the other is acquiring the National College. Circuses are great any place that you can build them. You'll need Libraries everywhere to build the NC. Finally, Workshops can greatly aid the process of getting other buildings up.

12) My advice is not to settle further. Staying at four cities will let you pick up either Democracy or Scholasticism and still get to Scientific Revolution. Adding additional cities just won't outperform that. Be careful in managing your cash with this approach. You tend to bury the other AIs in techs, and this means that Research Agreements get expensive, but your luxury count is low because you stay small. Be sure to sign RAs before triggering new eras whenever possible.



Babylonian Kudzu (Babylon)

The idea here is to exploit Babylon's abilities as much as possible. The free Great Scientist settled as an Academy lets you get away with aggressively REXing while still progressing through the early tech tree. Babylon's doubled Great Scientist production means that ideally you want to settle a city for every Great Scientist you intend to produce, because you will end up with several more Great Scientists that way. Also, doing that lets you remove the specialists from developed cities and focus on turning them into production powerhouses for the endgame.

You'll want to cut growth off in most cities after they reach size 3 if you're going for a Science win, so that you can grow your parts production cities as large as possible. With the way buildings and modifiers work these days, you'll probably want three parts cities which will get all of the production buildings you can stuff into them. If you're going for a Diplomatic win, this matters less because you're going to rush the UN in the capital with a Great Engineer, but you still want to make the capital as large as possible for maximum GE value.

Strictly speaking, Babylon only needs to take six SPs to win by Space. Liberty, Collective Rule, and the left side of Rationalism are all that is necessary. Landed Elite would be great, but it's too far into that tree for the benefits it provides. Democracy is overkill; you make Great Scientists fast enough already, and that bonus isn't worth three SPs. If you manage to get additional SPs, Freedom, Order and the SPs after Citizenship (Meritocracy and Representation) are the ones you want. Scientific Revolution should almost always be last, but be sure to take it in time to unlock necessary techs.

This game looks a lot like pre-patch Science games. Here's the gameplan:

1) Beeline Writing at the start, unless you're settling a luxury that can be Mined. In that case, take Mining first.

2) Assuming a strong :c5food: start (Cows/Flood Plains Wheat, or a :c5population: ruin), the best build order seems to be Scout -> Scout -> Settler. That should get you to size 3, which will give you respectable Settler output without improving terrain. Stealing Workers and settling luxuries if possible are essential to this build order. If you can, find a nearby civ with a barb encampment nearby and camp it with your Warrior while the Scouts explore, and steal a Worker from a semi-nearby city-state with a Scout.

Another plus with double Scout is the likelihood of getting at least one free Bowman, which will annihilate early units on the defensive. Use your judgment on whether/when you need to send it home rather than continue exploring.

3) The first 250:c5gold: you can produce needs to be thrown at a Cultural city-state to get Collective Rule. Borrow as needed.

4) Settle the free GS as an Academy. Cows are far and away the best tile. You always want to work them in the early game, and giving away a point of :c5production: in the capital doesn't hurt all that badly.

You may be wondering why I don't advocate bulbing Theology or CS here. The answer is simple. With Babylon, you're going to settle a ton of cities and you're not going to get a lot of SPs, so trying to get Patronage isn't a good plan. You want to delay SPs as much as possible in order to get Secularism as soon as possible.
Research luxury techs, Philosophy, and Construction.

5) Again, settle distant multi-luxury sites and get to at least four cities. Five or six is better if you can swing it, but don't let :c5unhappy: cause you to end up stuck with a bunch of cities with only one unit of population. If you didn't steal multiple Workers, you need to settle on every luxury you can to save Worker turns and keep growth going.

6) Secondary city build order is Library -> Colosseum -> University unless you have Horses, in which case Library -> Circus -> University is the way to go. One of the particularly nice things about Babylon's Academy is that you have time to pick up techs like HBR in a REX and still get early Education.

7) You can often sneak a Worker (or a Granary if you have multiple Wheat/Deer) into the capital's build order, which otherwise should be Library -> National College until Education unlocks.

8) Early Warrior/Archer rushes can be mauled by rushing Walls of Babylon in the targeted city if necessary. Don't do this unless forced to. Money is precious.

9) Hook up your luxuries, resell them, and get at least two Research Agreements by turn 40. Anything up to five is workable if you can afford it. A third will let you avoid hard teching Theology or, if delayed about ten turns, will let you hard tech Compass and get to the Renaissance. A fourth and fifth mean that it's vital to get Optics before you block techs (so that Astronomy unlocks). In any event, block what needs to be blocked and otherwise get as many helpful techs as you can.

10) Other than that, prioritize Maritimes with cash from your luxury resales. That will get you some food in the satellite cities, which will let you work Mines and finish builds faster. Time allying Culturals so that your next policy lands as the Renaissance unlocks, then pick up multiple Culturals so that you get Secularism as soon as possible. If you are forced to take a policy before the Renaissance, I'd recommend Citizenship. It isn't a great policy, but it gets you closer to Meritocracy (free Spaceship Factory, Solar Plant or tech) and Representation (free Golden Age), which you do want and may get if there are sufficient Culturals available.

11) Once Education lands, cities should be ready to start on Universities. The capital should switch to Settlers for a second wave of settlement once its University finishes. Library -> University -> Colosseum/Circus is usually best, since you won't need a third wave.

12) Fill Universities as they complete, and take the specialists back out of your future parts cities once they finish a Great Scientist and focus on growing them as large as possible before Apollo finishes. With Secularism, you will produce more raw :c5science: than any other approach can at this phase of the game.



Hope this helps! I'll edit the reserved post as I come up with new approaches.
 

Attachments

  • AutoSave_0000 BC-4000.Civ5Save
    437.9 KB · Views: 511
  • Harun al-Rashid_0232 AD-1610.Civ5Save
    988.5 KB · Views: 573
  • Martin_Arab_OCC.jpg
    Martin_Arab_OCC.jpg
    363.3 KB · Views: 3,315
Amazing stuff. Also probably applies to Cultural wins, though Stonehenge's value is so great that the Meritocracy choice becomes a bit tricky. GE for SH means you lose out on the early Medieval tech. It also means you get extra SPs before you unlock Patronage making those more expensive (though 8 cpt offsets this by quite a bit) unless ou are willing to sit on your GE and play a game of chicken with the AIs to see when they will build Stonehenge.

Will be trying out a lot of variations on the Patronage gambit in the next couple weeks for the cultural gauntlet game (despite the fact that at Emperor level AIs are as likely to be broke as flush with cash).
 
The AI seems to produce a lot more cash post-patch. I'm still not entirely clear why.

I spun up the Patronage gambit on Emperor (was contemplating it for TSG9), and it's not quite as powerful because the city-states produce much smaller quantities of :c5science:. In the 70s four allies were only adding 15 :c5science: per turn. That isn't terrible, but Research Agreements win at that kind of output level.

I'm thinking you stick with Stonehenge if you're trying to win on Culture on Emperor. That GE point is huge.

Settings don't matter, although Pangaea is going to yield more opportunities for city-state allies in the early going than any other setting.
 
Sorry about asking. But when I settle in a luxury I didnt get any bonus and I couldnt even sell it to an AI in the start of the game.
You still need to research the tech. So, if you settle gold, research mining first. If you settle a calendar luxury, go pottery->calendar first.
 
Sorry about asking. But when I settle in a luxury I didnt get any bonus and I couldnt even sell it to an AI in the start of the game.


You still need the proper tech to have access to that luxury, even when settling on it.

@Martin: Great work, thanks alot.
 
15 per ally of the 4 were only giving you a total of 15 :c5science:?

The four combined were yielding 15:c5science: per turn. It didn't really start getting better until close to turn 200, and by then I had Radio and Telegraph already. (This was a Cultural game, which ran to turn 272 because I am still a noob at Culture wins and made a lot of errors.)
 
The four combined were yielding 15:c5science: per turn. It didn't really start getting better until close to turn 200, and by then I had Radio and Telegraph already. (This was a Cultural game, which ran to turn 272 because I am still a noob at Culture wins and made a lot of errors.)
Yeah, because 272 turns for cultural really stinks. ;)

That said, I'm finding that there are a lot more things to screw up in culture games compared to my normal diplomatic games. Great Person production isn't as simple as "produce lots of scientists". Great Artists are the obvious goal and Great Engineers fill some niche, though I am still trying to figure out where they fit in. The wonders that seem obvious turn out to be less obvious when you look closely at the numbers (outside of Cristo which really is necessary for a fast win). There are a handful of techs that are both priorities and hard to block for (which is where GS comes into play in diplo games).
 
I have to build the Lib or rush buy?
I almost never have the money yet for rushing buy.
Are you playing on Deity? Ironically, for the strategies in this thread, deity is faster than immortal is faster than emperor... It's all about how much cash the AI has, so you can sell them luxuries and buy research agreements. At emperor and above, raising the cash to buy the library shouldn't be an issue.
 
At the risk of hijacking my own thread, some quick observations on Cultural (that was the first time I've won on Culture):

- Neglecting Scouts is a bad idea.

- Most Stonehenge turns are multiplier-free, so I'm not totally convinced it's worth it as compared to rushing to Free Speech and CR.

- Stonehenge is worth a LOT more turns than the Oracle, though.

- Early Landed Elite looks like a good idea. It isn't.

- :c5production: is king as usual.
 
I will try Immortal/Deity by the night when I arrive from the work.

The NC -> Scholarship just put you ahead in tech in Emperor. I dont know about Immortal.
 
At the risk of hijacking my own thread, some quick observations on Cultural (that was the first time I've won on Culture):

- Neglecting Scouts is a bad idea.

- Most Stonehenge turns are multiplier-free, so I'm not totally convinced it's worth it as compared to rushing to Free Speech and CR.

- Stonehenge is worth a LOT more turns than the Oracle, though.

- Early Landed Elite looks like a good idea. It isn't.

- :c5production: is king as usual.

Yep. Yep. Yep. Yep. Yep.

To make reasonable comparisons between wonders and social policies, I've tried to convert everything into culture. You can then convert culture into turns based on the cost of your last social policy (approximately 2500 in a OCC, give or take based on Piety tree, The Oracle, and Sydeny Opera House) and expected cpt generation during the crazy years when all the multipliers are in place (500 cpt is my general target for a OCC game).

On wonders, here is how I rate them in terms of priority:

1) Cristo. The reduction isn't huge, but when you build it you will still have at least 10 very expensive SPs to get. The culture savings over time is huge. Fortunately, you're unlikely to have any competition for it. You'll get it easily (assuming you aren't falling really behind in tech) even if you have to hard build it. Rushing it with a GE isn't really necessary as taking 10 turns longer to build is not a huge culture cost (<1000 culture). So don't save/produce a GE specially for this wonder.

2) Stonehenge. While many (not most as you should get Constitution SP less than halfway into the game) are without a multiplier, I still figure SH is worth 3750 culture or so. Plus, it makes getting a GE for the next wonder a bit easier. Building SH seems to be the wise Meritocracy play, assuming it still available when you hit Meritocracy.

3) The Louvre. Sure, it's a mid-game wonder. But 2 GAs mid-game are still a ton of culture. Suppose you get The Louvre at turn 160 and stop your culture focus at turn 230 (when you fill your SP tree and switch to production to build Utopia). Most of those turns are with a 4x or 4.33x multiplier. So a reasonable case is around 2 GAx6 cptx4 multiplierx70 turns = 3360 culture. That's a good haul

4) The Oracle. A free SP that doesn't change the cost of future SPs. Well, that's worth 2700 or so culture (in a OCC game - it's higher in a multi-city game). In deity games, I tend to ignore this. In lower games, I tend to hard-build this if I have reasonable hammers early and have nothing more pressing to build.

5) Sistine Chapel. +33% culture looks good, until you realize it applies to base culture. This was worth 2500 culture in a 252 turn deity win. That's almost identical to the free SP wonders. It's worth more if you have more base culture (monastery or multiple cities). Nice wonder, but hardly as essential as it appears, though more useful in multi-city cultural games.

6) Hagia Sofia. An extra GA is nice. Perhaps 2400 culture nice. Value goes down a bit if you are on a river and can build a garden anyway.

7) Sydney Opera House. Only available on the coast, and most of the best starts are not on the coast. Plus, you need 5 additional techs to get this over where I would normally stop with the RAs (Globalization vs. Radio/Telegraph). So you have to balance spending more money to go for this versus simply using that money on another cultural CS alliance. I almost never get this.
 
Well thought out and presented strats. Your point about :hammers: is of course solid. I have been trying a variant of your first strat using Ramesses to solve that particular issue (at least for wonder production). However, I've been trying to generate GS's instead of GE's and hard building the 'necesssary' wonders.

Please note, this is on emperor level (ala the HOF gauntlet). I am able to consistently go under 300 turns, even with sub par starts. But it seems to me that RA's far outpace the CS :c5science: bonus at this level and I struggle to acquire enough :c5gold: to fund both reliably because the AI's just do not have the funds on emperor.

I'm certain, I'm not playing optimally but I seem to have stronger games with Ramesses at this level?
 
Interesting about the first strategy (I congratulate Martin on all of them, but I`m not much interested in the Civ especific ones, because I don`t even enjoy playing the top Civs) is that I never though about skipping collective rule in order to go for Meritocracy. I usually use an early RA to get to medieval soon. But certainly not as soon you do with Meritocracy.

But I do wonder if you won`t miss the free settler too much in a non Arabia-OCC game. Wouldn`t a delay of about 10 turns be worth it, using the RA instead of taking meritocracy?

Also, Martin, after the scholasticism phase, what`s your build order like? I mean, between stuff like Unis, Workshops and Markets in your cities?
 
Top Bottom