Reconsidering Oracle for EL games

[Elephant]

Early Landing games are very different from standard Civ games, and that has prompted me to rethink some of the "standard doctrine" of Civ as I have seen the differences in action through Solo's Comparison Games. One of the observations I have reflected upon in my games (as well as half a dozen or so partial reruns of older games) is that there is an opportunity to reconsider the utility and lifespan of the Oracle in EL games. Usually it is one of the shorter-lived Wonders because I shoot for MonoTheism and Theology early in order to build my cities bigger. But both techs are off-path for an Early Landing so I have not pursued them myself (or the precursor tech, PolyTheism). I have left it to the AI, even trying to encourage any civ that gets PT by gifting Philosophy and holding off on more tech gifts (beyond perhaps Writing, Currency and University to encourage better research) if they start researching MT (and are not the KeyCiv!).

The problem is that most AI civs will not try to research MT until well into the game, well past when I could have used a boost in city size and had to resort to Colosseums (at 100 shields and 4 gold maintenence each). If I have passed RailRoad they are demanding my most recent techs to keep them happy, and often trying to research military offshoots like MobileWar rather than MT.

Another observation I have been thinking about is that the AI seems to skew its research when you have built a Wonder that has a shutoff technology, like Communism for Marco Polo and RailRoad for Hanging Gardens. I am usually the first to finish RR anyway, but I do prefer to get Econ before so I can go direct from Indust to Corp. Communism is nice to get in trade later on, especially when I am between SpaceFlight and Fusion and need a filler tech, because I like going for Espionage right after Fusion. But in general, especially in non-EL games, the AI will prioritize techs that kill any limited-term wonders I build.

I am wondering about the idea of deliberately building the Oracle right after the SSC wonders are done, or in between if caravans are popping out fairly well and gold is not in crisis mode like it usually is between Shakes and Newtons. There are two intentions to this path: I already have the tech (Mysticism) and Temples in every city besides the SSC, so this gives me two more citizens in every city for the "shield cost" of three Colosseums (with no maintenence), and I think it will "push" the AI to research MT, Feudalism, and Theology, about when I am not looking for them to contribute any EL-path research anyway but could use MT in particular to increase the size of my helper cities. I do not expect they will be able to finish Theology by the time I am ready to launch, and can always tech-bomb them if they seem to be getting close, or trade for it and build JSB. But I think this could increase the contribution of the helper cities in the midgame phase, when the SSC is struggling to make enough beakers to hit one tech per turn.

Your thoughts?

[SlowThinker]

You know, I am always for Michelangelo+Bach.

[Scouse Gits]

Elephant, although EL games are not for me, your treatise is well argued and seems to have a great deal of merit - as I recall Ribannah in some of her incredible OCC games chose to take the Oracle early with enormous benefit.
I would certainly not dismiss your idea out of hand and would strongly advocate a 'trial by fire'.

Good thinking,

SG[1]

[solo]

Elephant,

First, your observation about the AI keying on techs that expire my own wonders is right on! I also noticed that they seem to get an extra "boost" in their researching ability at these times! They can take forever learning an on path tech I urgently need, and then suddenly become like little Einsteins when learning a tech that expires a wonder I have such as MPE or the Colossus!

I have considered using the Oracle wonder, but have a number of reasons for not trying it:

1) I do not feel the need to go after MT (but will build it if the AI happen to learn Monotheism), so do NOT want the AI wasting their research along this path, by luring them there with the expiring Oracle.

2) By doubling temples, the Oracle allows cities to stay happy without luxuries at size 5, but I generally want helpers and colonies to reach size 7 or 8, which can be done with colosseums. Once the Oracle is used to get to size 5, there is not a cheap way of getting up to 7 or 8.

3) Although colosseums cost more and use up gold in maintenance, once they are added, they are a permanent solution to happiness problems. No off path techs are needed for colosseums, either, and at the time of the game when they are needed, trade is going so well they can be easily rushed.

4) At 300 shields, the Oracle is a substantial investment and I almost always have a higher priority wonder to build in its place at the time of the game when I think it could become useful. If it were only 200 shields, like the HG, then that would probably swing me over to your side and I would advvocate its use.

Of course my arguments should be taken in light of my current preference of using the minimal approach in EL games. Players using strategies involving more cities are apt to benefit more by using happiness wonders such as the Oracle and MT, and early on, the HG.

So, while my current opinion is against using the Oracle, I am ready to be convinced if someone can prove it works better!

[SlowThinker]

I will promote another wonder: Pyramids.

Food caravans may be a very effective way how to grow without celebrating and so without losing gold, especially if you have a shield+gold abundance.

[Elephant]

Originally posted by solo
Elephant,

First, your observation about the AI keying on techs that expire my own wonders is right on! I also noticed that they seem to get an extra "boost" in their researching ability at these times! They can take forever learning an on path tech I urgently need, and then suddenly become like little Einsteins when learning a tech that expires a wonder I have such as MPE or the Colossus!

I had not seen any extra "boost" in their research, but a definite change in "priorities", especially when their Mil/Civ characteristics would not lead me to expect them to pursue something, like the Babylonians researching Feudalism.

1) I do not feel the need to go after MT (but will build it if the AI happen to learn Monotheism), so do NOT want the AI wasting their research along this path, by luring them there with the expiring Oracle.

I do not research MT either, but I think building Oracle AFTER the key SSC wonders will make the AI pursue it, allowing me to trade for it and build Mikes with only 1 offpath tech. They still have to research two more techs to kill the Oracle, and I can easily make that so difficult that they never succeed (in time). I dont care about the AI "wasting" research at that point: it is usually after 20 techs anyway, when their research is too slow to produce anything in time for my needs. One difference between us, though, is that I tend to gift more techs to AIs than you do in order to update maps and keep them at least Receptive; gifting less techs as you do would still allow the AIs to be under 20 techs even when you are passing Flight.

2) By doubling temples, the Oracle allows cities to stay happy without luxuries at size 5, but I generally want helpers and colonies to reach size 7 or 8, which can be done with colosseums. Once the Oracle is used to get to size 5, there is not a cheap way of getting up to 7 or 8.

Temple gets you to size 3, Colosseum gets you to 6. How do you get to 8? And if the Oracle gets me to 5 earlier, when the SSC is still struggling before Auto, I can consider Xinning as an option too. It also may accelerate the AI researching MT, which I can then add on to get to 8 without Colosseums.

3) Although colosseums cost more and use up gold in maintenance, once they are added, they are a permanent solution to happiness problems. No off path techs are needed for colosseums, either, and at the time of the game when they are needed, trade is going so well they can be easily rushed.

Oracle does not involve an off-path tech either. I can time its building to use up accumulated low-pay/no-pay camels when it is convenient. If I can get it done earlier it will benefit longer, rather than waiting for the fatter trading time later in the game (after Corp?).

4) At 300 shields, the Oracle is a substantial investment and I almost always have a higher priority wonder to build in its place at the time of the game when I think it could become useful.

Two times it could be useful: very early, when the first camels are paying well but Ast and ToG are not done yet; and just after Newtons is built, just before Corp brings freights. Besides, for 8-10 cities, 300 shields is MUCH less than 800-1000, I can use up salty and woolly low-pay camels, and no maintenence. (And did I mention NO MAINTENENCE?)

Of course my arguments should be taken in light of my current preference of using the minimal approach in EL games. Players using strategies involving more cities are apt to benefit more by using happiness wonders such as the Oracle and MT, and early on, the HG.

I too am looking for ways to minimize expenses and get earlier boosts for EL games. My problem is that I am very slow in playing, so I am tossing these thoughts out to see what others think and perhaps try them in more games than I can. I've got another one cooking that I will write up today or tomorrow. Looks like I am just going to have to skip EL#4 and try to pick up with #5 next week.

[solo]

ST,

I am also considering the Pyramids for a future game, especially on large maps.

I have used food caravans for growth before, with good results, especially when the SSC runs out of its own food. However, caravans and freights are most often more useful for making trades or additional wonders, such as SETI and Darwin's.

Elephant,

These are all good counter arguments, but some tests are needed. I'm glad to see you will be trying this by joining game #5 in time to post a result.

The possibility of Xinning of size 5 cities did not immediately occur to me, too, which adds another major variable to consider.

Colosseums are boosted by the discovery of Electronics, allowing size 7 cities. I usually have Electronics by the time helpers and colonies reach size 7. Size 8 is most useful for the extra trade arrows, when making trades. In between trades, an entertainer keeps things happy, or if size 8 is desirable somewhere for a particular reason, a courthouse will do the trick and is cheaply added.

[Elephant]

Ah - I forgot about Electronics. I am thinking more of the tech timeframe between building Newtons and getting Corp, but I was also wondering about doing it earlier, like right after Shakes but before Copes. I often prioritize the higher-paying camels (Silk, Gems, etc) for limited shipping, so several Salts and Wools are often sitting around garrisoning empty cities. If Oracle could be built early those cities could join the SSC in its post-Shakes growth spurt. But then if it was that early the AI will have enough time to switch its research over.

So you add an 8th citizen and just keep her an entertainer? Doesnt that require a couple of irrigated grass tiles if you are also supporting an Engineer? And some Lux? I'm shooting for 0 Lux ASAP after Shakes goes in, but it is a struggle. And just getting the SSC tiles transformed is usually more than my Set/Eng staff can handle...

[Elephant]

Originally posted by SlowThinker
...especially if you have a shield+gold abundance.

That was a good laugh...
I rarely have a shield abundance; NEVER a gold abundance!
This is an EL game, after all...

[Elephant]

Originally posted by SlowThinker
You know, I am always for Michelangelo+Bach.

Me too, in "regular" games. But Mikes requires researching two off-path techs, and Bach a third (Feud, which I can sometimes skip if I can pick up Chiv in trade). I try to help the AI research MT, but am usually well into the endgame when they finally get it, if at all.

[solo]

Elephant,

Not all cities get to size 8, and if there is a food problem due to having an entertainer, I'll convert the extra citizen into an engineer, which as you noticed, come in handy.

However, in E$L games, most cities can operate under a food deficit long enough until the launch, if one has stored some in the food box before reaching size 8.

Another solution in a city without any food specials is to reassign its engineer to another having extra food, or even to the SSC when it reaches maximum size. My SSC often ends up owing a lot of food once it reaches full size, but the game ends before all its food can be used up.

Having too much gold is never a problem for me either!

[SlowThinker]

Elephant,
you consider ELG with the minimal approach only...

[Elephant]

Originally posted by SlowThinker
Elephant,
you consider ELG with the minimal approach only...

I'm not an expert on ICS and big civs like Cliff. ELG is enough micro-management with just 8-10 cities. You guys can try to beat us minimalist/perfectionists...