WFYABTA continued

[Majromax]

GalCiv I for Windows was also quite cool; it just had a few misbalances that kept it from being long-term interesting for the RB community, I seem to remember.

[Yosh]

GalCiv I for Windows was also quite cool; it just had a few misbalances that kept it from being long-term interesting for the RB community, I seem to remember.

Yes, earlier in the WFYABTA discussion GalCiv I was used as an example of how unrestricted trading can be a "slippery slope", to use Sulla's words.

Apparently, they've done a lot of work (and taken a lot of cues from CIV) on this game - I'm excited to see the results!

Sorry for getting off-topic! I'll lay off the off-topic talking now!

To go on-topic, I would be interested in seeing the "only pick one tech the entire game" done sometime too; I think it would be a Variant that would force people to *very* carefully consider their tech trades, knowing that the WFYABTA door could slam at any time!

[aviator99_uk]

Tech trading does not SPEED the game, it SWALLOWS the game, so that there is nothing else left. I've had my fill of that and I think most others have too.

- Sirian

Hear, Hear.

Personally I think Civ4 Tech trading hasn't gone far enough and still has the potential to swallow the game, but realistically its far far better implimented than any previous Civ.

Well Done!

[microbe]

I really do hate having to broker part of the tech tree.

Then you turn off tech trading.

There are people who like tech trading and consider it an integral part of the game. It was essentially the only thing to enjoy in civ3 (other than wars). We've also played no-tech-trading as a variant too in the LK series and that was fun too, but don't force your personal preference on other players just because you have the position.

I can see why unlimited tech trading is bad, but not why tech trading is bad. How do you catch up on deity anyway if there is no trading?

[Sirian]

tech trading... was essentially the only thing to enjoy in civ3 (other than wars).

Right. That's the problem. Did you want to stay there, frozen, for all time?

You still have Civ3, you know. And tech trading is still there in Civ4.

How do you catch up on deity anyway if there is no trading?

Real difficulty arises from the gameplay. The only thing relevant is how many tiers there are and how much space between them. The labels are just labels.

In real terms, the old Deity is the new Emperor. It's harder to beat Civ4 on Emperor than it is to beat Civ3 on Deity. That's all that really matters. The fact that we have two tiers above that in Civ4 means that players who want it can dial up the difficulty to "beyond Deity" (in Civ3 terms) or to "Hurt Me I Want To Lose!"


- Sirian

[Majromax]

Also, the kind of "difficulty" offered by any Civ AI on its hardest level doesn't even compare to the kind of difficulty you'd get by playing against an experienced player.

The Civ AI, regardless of edition (although Civ4's far better than Civ3) still plays on a simplistic, formulaic level. The difficulty (above Noble) is achieved by a handicap, rather than a smarter AI.

It's like playing chess against your 8-year-old nephew, but giving him a 5-piece handicap. He might win, but it's not the same.

Being "unable to catch up" on Deity would actually be a good thing, because it would mean that the AI is smart enough that a fixed advantage can translate into an unbeatable position. (Consider: you play a game against one of the better players here, but give them a 50-turn advantage. You'd probably never be able to catch up and win.) The Civ3 AI failed this test; even when Sid difficulty came in, players still could beat it.

I'd wager that the Civ4 AI doesn't have as severe fundamental flaws as the Civ3 AI. Thus, I'd expect that some level of fixed advantage would lead to an unbeatable position for the AI collective; this suspicion is supported by the difficulty people are having in beating C4-Deity on open maps.

WFYABTA isn't a game-rule problem; it's a universally implemented AI strategy. With that in mind, it's good play for the AIs, because it increases the odds of an AI victory. It might also reduce the chances of a single AI being victorious, but on average if the AI wins more often a single AI will also win more often.

WFYABTA is occasional , but it's a good rule-of-thumb for the AI. I'd rather have my AIs just occasionally, myself, rather than all the time.

[Bezhukov]

How does one catch up on higher levels in CivIV? Diplomacy and astute vulturing. That's how I somehow managed to win my only foray into Immortal so far (I fell behind via a very early jag-on-quecha war not of my choosing - I was quecha).

But I could tell I was over my head there, so went back to Emperor and other variations for a while to get a better feel for the basics. I do suspect that diplo will continue to be the key for non-exploit high level play, tho I could be wrong. Looking forward to seeing how the mighty Sirian and Sulla fare someday on Immortal/Deity.

[microbe]

Also, the kind of "difficulty" offered by any Civ AI on its hardest level doesn't even compare to the kind of difficulty you'd get by playing against an experienced player.

The Civ AI, regardless of edition (although Civ4's far better than Civ3) still plays on a simplistic, formulaic level. The difficulty (above Noble) is achieved by a handicap, rather than a smarter AI.

Sure, but off-topic.

WFYABTA isn't a game-rule problem; it's a universally implemented AI strategy. With that in mind, it's good play for the AIs, because it increases the odds of an AI victory.

So does AI always declaring war on humans.

[Sirian]

...Sirian and Sulla fare someday on Immortal/Deity.

I've won on Immortal via the Sirian Doctrine. That's more or less deserving of an asterisk, though. The AI simply lacks capability in the area of fleets at sea. There is no counter for the Carrier Group except for a bigger, badder fleet, and the AI doesn't do that. So... it's a race. Can I get to Combustion and Flight, and Oil, in time to hit the leading AIs where they live, before they can launch. ... Not an option on all-land maps, which are therefore more difficult for me, on average.

Deity is only possible (for me) under the most favorable conditions, ala Moonsinger's gimmick. That doesn't mean it can't be fun, though. Sometimes the measure of a game lies in who loses by the lowest margin, who holds out the longest before going down in flames. 8)

There will be Deity events at RB.


- Sirian

[Majromax]

So does AI always declaring war on humans.

Let me refine, then:

In a competitive human game, it doesn't make sense for the human players to endlessly trade techs among themselves. Trading does give, in purely "economic" terms, mutual benefit -- but Civ is an adversarial game.

Humans will eventually reach a point where they realize that Player X is in danger of running away with the game. They'll then stop helping Player X, even at the expense of smaller harm to themselves. In Civ4 terms, they'd cancel resource deals and stop trading techs; if anyone's played Settlers of Catan (haven't myself), they'd stop trading all-important resources with the presumed leader.

The Civ* AI is bad at strategic decisions. Strategic decisions are, by and large, intuitive decisions among human players; that's why experience is the best teacher, rather than hard-and-fast rules. Intuitive decisions are the hardest to program, and it's to Firaxis' credit that the Civ4 AI conceals a lot of its strategic bone-headedness. But we should never mistake adequate tactics for good strategy; the Civ4 AI can't (correctly, anyway) assess when giving the player (or even another AI) a tech in trade would actually increase that AI's relative odds of losing.

Hence, WFYABTA. AI players aren't trade-happy themselves, so this limit amongst them isn't needed (yet. I'm sure, when the SDK comes out, someone will mod Civ3-style tech trading among the AI pack back in. Woe be unto them.) Since, as I've already said, the AI's can't reliably tell when the human player would get "too much" of an edge, the WFYABTA limit is an attempt at drawing a conservative line.

To that end, it works. I think its presentation is a bit suboptimal, but the basic idea is entirely worthwhile.