Civilization 5 Announced

[Txurce]

There are many substantive critiques of the game here - most recently posts about the economy, the trade-off between balanced and exciting starts, or the relative unimportance of strategic resources besides iron. But frequently, understandable disappointment and frustration with its failings is expressed with snideness and hyperbole. Sometimes it spills into what feels like “hate it or leave it.” More often, accurate criticisms lead to overblown conclusions. I have a hard time finding any aspect of the game to be wholly bad, and my own reaction to what I see as hyperbole is to question a conclusion. That tends to happen when someone writes in absolutes, like “the AI diplomacy makes no sense,” or “the combat AI is worthless.”

I’m never confused by what the AI does diplomatically, once it does it… just like with a too-straightforward human player. I agree that its about-faces are sometimes jarring, and wish they weren’t. But in a nutshell, I can get what I want – trade, joint research and sometimes war - until either someone sees me as dinner, or they think I’m going to make them dinner. This can happen early or very late in a game. The AI expresses all this in a fuzzier manner than it did in Civ4. And because everyone at RB is a good player, the AI playing to win often comes off as making sure you don’t. That doesn’t work for everyone, but it does for me.

Likewise, I find the combat AI very disappointing, but far from worthless or even incompetent. Keeping in mind Shafer’s decision tree, I replayed a section of an Emperor game where Songhai used catapults behind strong melee units to plow through a bunch of CS. The second time Songhai had no siege weapons, and the attack was the usual grinding bloodbath. This indicates that the AI can sometimes get things right, and that this would be plenty good enough vs the average player, when playing at a higher level. I see reason for optimism based on the first Songhai example. Similarly, size-4 ICS empires seeming most optimal in October doesn’t mean that the game hasn’t opened the door for a small civ to be successful in everything but conquest.

[Sullla]

Well this is sure to make a lot of folks happy: http://www.garath.net/Sullla/Civ5/liberteordre.html

I'm starting to wonder how much longer people will continue to claim that the AI is "playing to win", which appears to be a catch-all phrase which explains anything and everything that it might do. All I see the AI do is war endlessly with its neighbors over and over again, which isn't so much a winning strategy as a series of dice rolls telling it when to attack.

[Brian Shanahan]

Well this is sure to make a lot of folks happy: http://www.garath.net/Sullla/Civ5/liberteordre.html

I'm starting to wonder how much longer people will continue to claim that the AI is "playing to win", which appears to be a catch-all phrase which explains anything and everything that it might do. All I see the AI do is war endlessly with its neighbors over and over again, which isn't so much a winning strategy as a series of dice rolls telling it when to attack.

Thanks for that, was checking on and off all day (quarter to midnight over here), but in the meantime I've been reading some of your "lost" Civ3 reports using the way-back machine at web.archive.org (about half the pictures missing though).

[Krill]

Pity, couldn't you have made teh reports less...insulting?

[Brian Shanahan]

Pity, couldn't you have made teh reports less...insulting?

Yes all that politeness and niceness and goodness gets on my nerves.

But seriously between here and the reports make me mighty glad I did not spend my scant birthday money on the game, and even if they massively improve it I will not touch it simply because the way they release it.

Oh yeah CaptSAH while I don't work in games dev. like you do, from the other game I really play (Football Manager) I know that bugs can often be caught but not dealt with straight away, but the developers there are open enough to admit that and will always patch the ones they catch out asap (often with a patch coming out on release day). But not alone are the bugs not being dealt with properly, it looks like core mechanics of the game have not being tested for brokenness, e.g. ICS and maratime/cultural CS boni. That suggests to me a developer/publisher looking at enhancing the bottom line 3 months from now to the detriment of the health of the two companies a couple of years from now (actually a mentality prevalent in modern business, but that's another rant). I don't mean to aggrivate you but it's something I felt I needed to say.

[luddite]

Well this is sure to make a lot of folks happy: http://www.garath.net/Sullla/Civ5/liberteordre.html

I'm starting to wonder how much longer people will continue to claim that the AI is "playing to win", which appears to be a catch-all phrase which explains anything and everything that it might do. All I see the AI do is war endlessly with its neighbors over and over again, which isn't so much a winning strategy as a series of dice rolls telling it when to attack.

Excellent write up! I'm good at finding strategies that are brokenly powerful, but I'm not so good at writing them down and explaining why they're good to others. Your report does a great of showing why building small cities everywhere completely breaks this game. Hopefully Firaxis will implement some sort of change to fix it if enough people start complaining about how easy this is.

I think I know why they made units so expensive in this game. There's supposed to be a big emphasis on warfare tactics, right? So you're supposed to have just a small army of 4-8 units, and position them very carefully. If it takes 20 turns to build a new unit, and also lose one every 20 turns then your army will stay about the same size throughout the whole game. They just didn't expect us to completely break the economic system and produce unlimited numbers of units . When you've got a unit on every single tile on the front line, tactics are almost irrelevant.

I know exactly what you mean about wanting to build trading posts on the resource tiles too. I guess the local wildlife are interfering with construction can't build a strip mall with sheep wandering around everywhere, and you can't kill them off because some environmental group will protest.

The only thing I might perhaps criticize about your game is that you did a horseman rush at the beginning. Given how bad the AI is at warfare, and how stupidly powerful the horsemen unit is, this is almost as abusive as stealing workers from a city state. I don't think it matters much though, since I'm sure you would have still crushed him in any war even without using horses.

If you want to do a succession game on deity I'm down. I feel like as long as I don't die to an early attack (which can happen on deity) the AIs have no chance of winning. It's gotten to the point where I just found my first cities in the best possible defensive locations and ignore economic considerations, because I know that if I can just survive the first 150 turns or so then I'll jump ahead once the city spam kicks in. It is rather tedious though, killing so many AI units.

[Tredje]

Well this is sure to make a lot of folks happy: http://www.garath.net/Sullla/Civ5/liberteordre.html

I'm starting to wonder how much longer people will continue to claim that the AI is "playing to win", which appears to be a catch-all phrase which explains anything and everything that it might do. All I see the AI do is war endlessly with its neighbors over and over again, which isn't so much a winning strategy as a series of dice rolls telling it when to attack.

Splendid write-up once again I must say. And although I don't think you will manage to silence those who don't consider the core mechanisms of Civ5 broken, I am certain that you in your reports have touched upon all the important issues that point towards the fact that they are broken.

[De Bartman]

Well this is sure to make a lot of folks happy: http://www.garath.net/Sullla/Civ5/liberteordre.html

Sullla, In your report you give an overview of the terrain and you explain why you put a tradingpost on everything you can. Reading it gave me the sense that it doesn't matter where you settle your cities. Even your starting city. Sure rivers and happiness resources are nice and speed up your ICS but essentially it doesn't change the strategy. And as a result the game breaks. Just mindlessly ICS and win the game.

Am I right here or am I missing the point?

[v8mark]

I think that's pretty much what Sullla was trying to convey, reading the report. And it makes sense, too. I just get no impression that any kind of thought has gone into this on a small-scale level with regard to tile improvements.

It might even be playable if the AI copied this basic strategy of ICSing. The fact that it doesn't shows that the devs had no idea that this would be the dominant strategy, which again shows the ineptitude of the play-testing. It wouldn't be much fun, mind; just an inferior version of Civ3 (which wasn't much cop anyway)...

[Zed-F]

I think the intent is not to get the AI to copy ICS and thereby enable it, but get the developers to make some changes so this sort of ICS is no longer possible.

Sulla, I am sorry to hear that Civ5 disappoints you so badly, but I agree that taking a step away from the game until it can be improved to the point where you may be able to enjoy it is the right approach.