Technology: how do I keep up with AI?

[Mahler]

I would say first priority should be placing on the resources. When these good places are filled, one can place cities on other places. I don't build settlers from the beginning. I build granary first in the most villages.

Does the AI build farms near cities? I have landed with the freight to AI's territory and have not seen farms
Yes, I finished the game - no farms, and no terraforming made by AI.

[nef]

IMO, it is a question of strategy, ruleset and experience. In my games I follow (almost) none of the advice above (including that from Ignatus). But to start with I am using a modified version of civ1. (Some of the modifications are published, some not, but they are ALL designed to improve fidelity with Civ I). Before you ask, I always play cheating, and always with the maximum number of AI allowed by Civ I.

So to some specifics:

1. Number of cities - 6 makes for a quick and easy game, but 2 or 3 is enough. I end up with zillions through conquest almost all of which I disband. This number (6) applied when I was playing with other rulesets such as classic and civ2civ3.
2. Because of the (no) auto railroad on city centre I avoid specials (even just grassland if I have a choice). After I get railroad this changes because I can build railway first. (Note despotism means no advantage on plains in civ1/ Civ I.) The key criteria I use is to maximise the number of arable tiles for each city.
3. I always stay in despotism except while actually creating trade routes.
4. I select one city to be the SSC (super science city) preferably, but often not, the first city so building a new palace is ultimately needed to reduce corruption.
5. One high production city with barracks for, err, I forget. Once I get steal and advanced flight I need two or three.
6. The rest for caravans.
7. In the opening stages of the game I build sufficient militia (warriors) to claim (and defend) sufficient TERRITORY. Only then do I attempt to fully occupy the territory with cities. There is a strategic balance here - you need cities to build (AND SUPPORT) the militia. As I have said before, the imperative is to maximise the number of tiles worked. (Not the number of cities.)
8. I make full use of exploits. Compared to Civ I the fc AI are BRAIN DEAD. Learn the exploits. The game becomes trivial. The civ1 ruleset is especially bad in this respect even after the hitpoint / firepower = 2 fix.
9. I like to have one settler for every two or three cities for gardening. This keeps up with natural growth. I do NOT use granaries until just before I build an aqueduct. Just before railroad I try to double the number of settlers. (In civ1/Civ I railroad increases ALL tile produce by 50% so irrigated grassland gives an extra food bushel and this will support the extra settlers. In other rulesets you have workers.

If you are just starting out, the strategy above will END IN TEARS. One of key attractions to Civ I was that each new game was a new game. You LEARN new skills (and I still do). You have to LEARN the ruleset. You have to LEARN strategic balance. And, you have to learn the idiosyncrasies of the AI. I have said this before but it is worth repeating: the difficulty levels in fc are insignificant. (In fact it took me a year to figure out whether the level applied to me or to the AI, and then it was ONLY because of the way the server command worked.) Because of this beginners will have a tough time getting started, so take it easy; try to last longer in each new game, until you can eventually start actually winning. The key is to learn as you go, and there is no harm restarting a game from an old save (10,20 turns back or so) in order to try variations to your strategy.

One last point, fc mapgen is a pig. The quality of the starting position varies enormously so some will be ridiculously easy; some ridiculously hard. With the latter, don't bother - start again. (Because of an (unpublished) modification I no longer have this problem, but it requires adaptive strategy.)

[PatTheCatMcDonald]

Early game, to keep up with tech, I'll go for harbours, marketplaces and libraries in my coastal cities as a priority. It doesn't give the same short term boost of having lots of small cities but it pays off in the long term.

It's a big slog really, but you get way more trade in the early game from the sea than from the land. Plus, pottery is a very good starting tech, it leads onto triremes and harbours and gives the granary improvement once you get it.

It makes sense to build marketplaces before libraries, to give you the gold to pay for improvement upkeep.

As for special squares, I generally avoid building on them. I like the bonus shield from grassland. Sometimes plains can be better to give a match for even food production but that's more of a late game topic than an early game topic.

I have noticed the AI doing very well by building first city on a mountain with gold and a river. That's one helluva advantage from turn 1.

[John Campbell]

You also don't have to rely entirely on your own science production. Send out Diplomats, establish embassies with as many other civs as you can. That lets you trade techs with the friendly ones, and at least keep track of where the hostile ones are at. AI civs that like you (and usually they either love you or hate you) are generally willing to trade techs even at a disadvantage. When you develop a tech, you can often trade it to three or four other civs for different other techs that you don't have, and then trade your new acquisitions around even further. This lets you pretty quickly bootstrap yourself up to having every tech that any of your neighbors has, plus a few that are just your own.

The Great Library is obviously excellent for ensuring that a rising tide lifts your boat, but it's also a Wonder that the AIs actively pursue, and I find it's very difficult to get it before one of the AIs does. Marco Polo's Embassy is more subtle but very useful. It gives you contact and embassies with every other civilization, in particular the ones on the other side of the world that you haven't met yet and so who have no beef with you. This puts you in position to act as tech broker for the entire world. Be mindful if someone else has the Great Library, though... if you get a new tech from someone, and want to trade it to the civ with the Great Library, do it the turn you get it, because as soon as you end turn they'll get it from the Library and you'll have lost your chance to get something out of them for it.

(If you're playing with the new terrible default ruleset, ignore the bit about the Great Library... it sucks now, and provides most of its benefit to everybody. Marco Polo's Embassy has gotten even better, though, because there's a tech-leakage effect, so techs are cheaper the more civs that already have them that you have embassies with. Having embassies with everybody means that any tech that anyone else has is cheaper for you.)

[cazfi]

(If you're playing with the new terrible default ruleset

Some feedback about civ2civ3 ruleset! (I don't know if bard is around any more, but already some years ago he was unhappy that there has been no feedback, to guide the development, all that time that the ruleset has been available - at least since freeciv-2.5)

I know it was only a one sentence of feedback, but; would there be any point in introducing something like "classic+" ruleset that would be the very mature classic ruleset with just some relatively minor changes (but changes bigger than what we may want to classic itself) ?
Though we already have variant2 that is quite close to what I just described.

[cazfi]

Though we already have variant2 that is quite close to what I just described.

Some background for variant2: As long as freeciv has had the concept of rulesets (instead of all rules coded to the engine), i.e., over twenty years, I've been maintaining separate ruleset, originally forked from classic ruleset, for my own multiplayer games with some friends. That exact ruleset has not been published to wider audience (there's always some experiments in it that do not turn out so great in actual game), but variant2 closely follows it.

The risk of variant2 not being great for any stable freeciv version is that we play with main development branch, and the rules are not really adjusted, and especially not playtested, for any freeciv changes that go to stable branches once there's a newer version in main development branch.

[Elefant]

According to this http://forum.freeciv.org/f/viewtopic.ph ... df64ed2789 poll, classic is still very popular, though the number of respondents means that only limited insight can be gained.

[nef]

Some feedback about civ2civ3 ruleset! (I don't know if bard is around any more, but already some years ago he was unhappy that there has been no feedback, to guide the development, all that time that the ruleset has been available - at least since freeciv-2.5)

I know it was only a one sentence of feedback, but; would there be any point in introducing something like "classic+" ruleset that would be the very mature classic ruleset with just some relatively minor changes (but changes bigger than what we may want to classic itself)?

IMO, "classic-" should be considered. Maybe, civ1civ2. I think something simple for beginners, but civ1 is showing its age and Sid was not crash hot with history (or geography for that matter). Some features of experimental have little or no effect on gameplay, so those features could be considered for inclusion.

As for civ2civ3, I see it as just another ruleset. I saw no particular reason for it becoming the default. In fact, in terms of simplicity for beginners, I would suggest classic before civ2civ3.