Gamesmanship and Strategy

[cal_01]

Originally posted by Unspeakable Horror


Do animals get into your territory?

I had a bear that was trapped between my borders and the sea, the poor thing wandered in the 3 tiles it had of neutral territory for a few years and eventually died

Er. Maybe I should rephrase: the bear(s) was just outside my cultural borders, so any one of my units trying to get out would promptly get slaughtered (Bear 3 vs Scout 1 + 100% vs animals still equals splat )

[DrSpike]

Originally posted by Velociryx
I use lots of high risk strategies. Usually, when they fail, they fail BIG TIME (ie - I've been running totally defenseless to speed build my second city, barbarian wanders by...poof.

That's game.

There's no recovering from that, as I just lost my only city before I had my defenses up.

Has happened twice.

I can't speak for anybody else and their proposed ideas, but for me (and specifically re: "settler first," failures generally only come in two flavors:

1) Something wicked finds you before you get your defenses up.

Result: End of game. Not because you're a bad sport, but because you've only got one city with no defense. Predictably, you lose, and start again. *Shrug* It happens (twice in over a hundred starts at varying degrees of difficulty, so I'd still call it a pretty "safe" notion, overall).

2) Something wicked finds your early settler/worker and eats it.

Result: A setback. Nothing about the strategy changes, really. You're either committed to it, or you aren't. Maybe you'll slow down and build an escort, or maybe you'll push all the harder to make up for the lost time, but whichever route you choose, the end result is (or should be) the same. Then again, I'm doggedly determined when I feel like I've got a winning idea/start on my hands.

-=Vel=-

Let's leave aside for a second that settler first is a strategy of limited power.

I want to ask what you would do in a comparison game. Call it a 2% chance of failure, and assume that although there is this 2% chance if it comes off you will have the best opening position.

Funny thing is I've seen lots of risky strategies in comparison games over the years (usually with risks greater than 2%), particualrly competitive ones, and they always seem to come off. The number of people saying ah I did x but it went wrong seem very scarce.

[Velociryx]

A comparison game would be, in my mind, like any other game....that is to say, entirely dependant on the civ I'm playing, my early game goals, and the starting map position.

I'll go settler first if and when it makes sense to do so, given the above. Otherwise, I'll either go with worker first on the opener, or grow first, then expand.

The general trend, for me, is something like:

Settler first: 25%
Worker first: 50%
Growth first: 25%

-=Vel=-

[DrSpike]

And presumably if it went wrong you'd be man enough to admit it?

[Velociryx]

Sure. Just as I've admitted that I've gotten toasted a couple times before I could get a build out.

No biggie.

It's gonna happen once in a while if you leave yourself open to a window of disaster in the extreme early game in exchange for a jumpstart.

-=Vel=-

[Jcg316]

I've found barbarians to be just about meaningless. With Inca/Mali they're just free experience, with horses its much more common to take a few barbarians cities, when they occasionally build some, than to actually lose even one unit. Just having warriors and finding quick barb archers you may have to give up a couple units in a bad situation but as they are far too predictable you can usually just find a forest in their path and do fine.

Seems that with Raging Barbarians selected (and occasionally otherwise) the progression goes Animal-Archer-Warrior-Warrior and Archer-Archer and Axeman, instead of warriors before archers. The archers can be fairly brutal sometimes if they show up very, very early but otherwise they aren't a big deal.

You can always just move one space at a time with a settler and be able to scamper away from any 1 movement animal unit that you find. It won't be an issue if you know where you're going and what you're doing with your scouting unit. If you have a good idea where the barbarian animals are going to appear that might not even be an issue. As predicatable as they are, once you've seen a barbarian you should know pretty well what they're going to do over their next five/ten turns if they're close to you. Only time its really iffy is if they are far from your cities or about even distance from two different civs.


In general, one animal resource is a decent start, one farm resource is alright, even just having stone or marble wouldn't be terrible. Having a lot of unusable ice or desert isn't an issue as you can always move a city radius or two in a different direction. There may be someone with a better start but civs are usually placed in locations that, overall, are similar in value to where other civs will be placed. On a great plains map not getting one resource, if the map script even allows that, or just getting a piddly little resource or two would be tougher as some other civ might have three/four animals, but certainly not impossible.

The actual terrain where you start may influence what you want to build first, second, and so on but is so much of a smaller factor than other issues with placement on the map, in regards to other civs/landmasses/etc, it won't mean that much.

[Blake]

Something I do want to talk about is risk vs reward, especially in Multiplayer.

It can be absolutely nessecary to take some risk if you want to do well*. For example say you're facing a player you KNOW is better than you. If you play conservatively you are GUARANTEED to lose. You could however try a high risk strategy, there might be a 30% chance you'll lose outright (usually with the resposne LOL, what the heck are you trying to do???), a 40% chance you'll lose eventually, but a 30% chance of actually winning. Some Oracle strats would probably fit in this category, you might miss the Oracle, you might get reamed by some Prats or Jaguars before your plan comes into fruitition. But it might just work spectaculary. (as in "Holy crap! How'd you get so many of those so fast!)

* You can also play adaptively. The challenge in doing so is that if you are playing reactively you have already lost, as you are like a puppet to the other player. You have to play formlessly, such that the other player cannot discern how you will react, and thus cannot manipulate you. The pinaccle of strategy is ofcourse, having the other player believe they are forcing you to play reactively, while they are in fact falling for your masterplan hook, line and sinker. It is unfortunate that AI's are so uninteresting from a meta-strategy perspective. They don't even know when you're playing circles around them.

[Arrian]

This is the sort of thing that AU courses are good for. They're comparison games, and I believe that those who put in DARs are honest about what transpires. There may be a few who try, use a risky strat, fail, and do not post anything, but there's no way to know.

I've gotten crushed in AU (and proto-AU) games before. Those that spring to mind are MiniTourneyIII (Germany-in-jungle) and AU102 "Give Peace a Chance." In MTIII, I tried a risky strat that I often used to great effect. In this particular instance (up a difficulty level for me - MTIII was an Emperor level game), it failed rather spectacularly. I conceded defeat and owned up to it. I learned from that game, too. I picked the wrong target, I waited too long, and I had yet to discover the power of worker pumps...

-Arrian

[Stuie]

Originally posted by Dactyl
I will check them out though I hope they will give examples of the game as shipped, rather than modding it first.

Don't worry - AU will start out with the game "as shipped" plus the first patch.

[Arrian]

An AU Mod (to the extent there is one) would, I think, be consistent with the CivIII AU Mod. The CivIII version took quite some time after game release to develop and was done with the express purpose of making small changes to improve the AI & boost strategic choice.

It will take us quite some time, I'd imagine, before we're ready to approach Civ4 that way.

-Arrian

[DeepO]

Originally posted by Arrian
It will take us quite some time, I'd imagine, before we're ready to approach Civ4 that way.

My advice: don't start this just yet... we're still in the patching phase of the game, and tweaking these things is also something Firaxis has always took interest in. IMO, Soren would be the ultimate AU-mod programmer

DeepO

[Arrian]

That's what I said, DeepO. It will be a while before the time is right to start on an AU Mod for Civ4. Both for the reasons you mentioned and because of the inexperience w/the game of most AU'ers.

-Arrian

[DeepO]

hum... I wanted to chime in... where did my English communication break down this time?

DeepO

[Arrian]

Sorry, the way your "my advice" post is written made me think you were advising me to wait on the AU mod thing, which struck me as odd, since I was trying to reassure others that AU would most likely do just that (not that I have any more say than the others involved in AU).

-Arrian

[DrSpike]

Originally posted by Blake
Something I do want to talk about is risk vs reward, especially in Multiplayer.

It can be absolutely nessecary to take some risk if you want to do well*. For example say you're facing a player you KNOW is better than you. If you play conservatively you are GUARANTEED to lose. You could however try a high risk strategy, there might be a 30% chance you'll lose outright (usually with the resposne LOL, what the heck are you trying to do???), a 40% chance you'll lose eventually, but a 30% chance of actually winning. Some Oracle strats would probably fit in this category, you might miss the Oracle, you might get reamed by some Prats or Jaguars before your plan comes into fruitition. But it might just work spectaculary. (as in "Holy crap! How'd you get so many of those so fast!)

Absolutely. In WC3 I would tower rush very high level elves every game because I knew I would lose playing 'properly'. Haven't got to that stage with Civ4 yet, but possibly will.