AI Cheats

Urtica dioica said:

Actually I would like a discussion about what qualifies something as a cheat, just not in this thread.

Click to expand...

When you start talking about cheats, remember it's a two-edged sword:

1) Triremes travel freely across open ocean and are never lost.

2) Caravans instantly reach their destinations without having to travel there. AIs may have trade routes with people they've never contacted.

3) Surplus production is not forfeited. When you produce a caravan in a city with 8 units of production, it takes you seven turns. You produce 56 shields. Those extra 6 are lost. The AIs get to keep those extra 6 shields.

4) Wonders of the world are simply awarded. AIs don't work for them.

There are probably many others.

• Mathematically favorable formulas for science and city growth at higher levels
• Initiate diplomatic relations with any unit

Don't know if these count as "cheats", but #1 throws off the power balance as effectively as cheating.

Whenever I'm buildng a BC wonder, someone beats me to it and I haven't saved for 00's of years.

BTW, I tend to find barbarian attackers to be less potent. Illusion? or anti-cheat?

lndm said:

Whenever I'm buildng a BC wonder

Click to expand...

Never ever telegraph your wonder intentions early in the game. They always have more advances than you. Of course they win their wonders in a cosmic lottery in which you have no part. When I decide which city is going to be my wonder city I usually build a temple or something, and never change the build selection afterwards until there are enough shields to produce the desired wonder.

The FAQ has a list of AI cheats. I'm not sure all of them are correct; the FAQ isn't always right.

Thinking of AI cheats, I wonder if anyone considers abusing them a cheat...

For teleporting caravans, if you steal a city that's used them, you get not only a free trade route, but the MAP display pinpoints one of the world's biggest cities, often outside the discovered world. One might take this further by giving Trade to a target Civ before destroying them.

One time I played a game where I started with Map Making. While my chariots were rolling toward my first victim, they started a dialog and asked to trade techs. They took Map Making and were given Lighthouse within a couple turns. I soon destroyed them and got a nice wonder early and free. Some people revert to a save to keep the AI from taking a wonder, so why not revert over and over until the AI gets a wonder?

As for the AI's unfair production overflow, as I see it, the more improvements it gets, the more money I make from selling them.

@lndm

In my experience, trying to beat the AI to a wonder is futile. Our only consolation is that it needs to have the proper tech before a wonder can appear. That's no use for Colossus, since the AI puts a high priority on Bronze Working, but it's key to most other wonders.

If it seems barbarians are unnaturally weak, it might be based on difficulty. On Emperor, I feel like they're supernaturally strong. Then again, I swear I've seen a diplomat fend of barbarian cavalry before...

Dack said:

Never ever telegraph your wonder intentions early in the game. They always have more advances than you. Of course they win their wonders in a cosmic lottery in which you have no part. When I decide which city is going to be my wonder city I usually build a temple or something, and never change the build selection afterwards until there are enough shields to produce the desired wonder.

Click to expand...

Hi Dack, that's interesting. Do you think the AI really bases its decision on whether you are already building a particular wonder? The thought had occurred to me, but I now suspect that the algorithm is based solely on what the AI thinks it needs/wants... thoughts?

Dack said:

They always have more advances than you.

Click to expand...

They do? Do you mean they always start with more free techs, or...?

void said:

Do you think the AI really bases its decision on whether you are already building a particular wonder? The thought had occurred to me, but I now suspect that the algorithm is based solely on what the AI thinks it needs/wants... thoughts?

Click to expand...

I always assumed the formula was something like the following:

Where prob(N) is some probability factor, probably something like (rand MOD 64) and may be level dependent.

In plain English, I hypothesize that, given availability of required advancements, the CPU has a 1 in N chance per wonder per city of building an available wonder. I estimate the per wonder probability is around 1 in 64. Calculations abort once a wonder has been built--that is, I don't think it's possible for one CPU city to build two wonders within a turn.

I am more confident about per city chances than the probability that each available wonder is given a chance to build within each city--that is, I'm not sure whether each CPU city gets a chance to build each available wonder or just has one shot at an available wonder, either random or targeted. Only statistical analysis or reverse engineering would reveal this data.

void said:

Do you think the AI really bases its decision on whether you are already building a particular wonder?

Click to expand...

Whelkman has attempted to develop a formula for the cosmic lottery. I don’t know what criteria are used, but my experience leads me to surmise that the randomness is influenced by the wonders you are building. My short answer: YES

void said:

They do? Do you mean they always start with more free techs, or...?

Click to expand...

Perhaps I overstated. In general, collectively it seems that at the early stages at least some of the other civilizations will have an advance that you don’t. What this implies to me is that building an early wonder the odds are that some other civilization will have that advance also.

Dack said:

My experience leads me to surmise that the randomness is influenced by the wonders you are building.

Click to expand...

This can be exploited. Start building a wonder you have no use for like the great wall and switch to what you really want just before it completes. (Of course you have to keep a close eye on it)

This reminds me - we aren't bound by the "must offer peace" benefit of the great wall. That's actually a cheat in our favor.

BTW, my usual strategy is to stockpile caravans

Urtica dioica said:

The FAQ has a list of AI cheats. I'm not sure all of them are correct; the FAQ isn't always right.

Click to expand...

Thanks Urtica dioica, I'd forgotten that list was there. For the record, a few corrections:

c) Improvements (at Emperor level) for the computer are at 1/3 normal cost.

Actually, the production box has 8 columns instead of 10. This drops to 6 columns in 1 AD.

d) Technologies are achieved at silly rates with only a couple of cities.

They are probably using the caravans for science trick. In one game, I was continually sending in diplomats to see what the Babylonians were up to. Most of the time, they were building caravans and beaming them to London even though there were other AI cities available. (The Babylonians themselves had only one city.) When they discovered banking, they built a bank. Their "silly" discovery rate was put on hold until they could get back to building caravans again.

h) The computer sometimes builds multiple cities before its first settler is created. (However, when playing the Russians, human players sometimes get two settlers!)

They probably get 2 settlers using the same algorithm that we use. They can also visit huts and get advanced tribes - just like we do.

i) Also, production in cities at emperor level continues, even if the city is shy the necessary resources to maintain what it has.

They can buy shields any time similar to out rush production, but they can buy any number of shields, not just what it takes to complete something. I believe they get a little help balancing the treasury, but this is hard to prove. If you step on their horses squares, they can still lose units to lack of support.

Not quite the cheat thread I had in mind, but it'll do.

To call something a cheat, you first need to know what the rules are. In Civ 2, there are the GOTM rules, but those are the rules of the Civ 2 GOTM, not the rules of Civ 2. If Civ 1 had solid rules for its GOTM, you could cheat on the GOTM, but that's not the same as cheating at Civ 1.

I'm not saying that Civ 1 doesn't have rules; it's not like using a chess set to stack pieces and make pretty patterns. The game engine mandates battle win chances, unit movement, construction rate, tech advances, etc. So, maybe we should play this way, doing whatever the engine lets us do. It makes the rules, and they're the only rules we've got, right?

What, you don't like that?

It turns out doing whatever the engine allows leads to a long, slow game few enjoy. In fact, it becomes more of an art form or a puzzle than a game.

- - -

Years ago I decided to play the Legend of Zelda (original NES game) without a sword. You can't actually rescue the princess without one, but you can get to the final boss, so the new object of the game was to get killed by the final boss. It turned out to be a fun game, a fresh challenge.

- - -

Like I did with the Legend of Zelda, I make Civ 1 more interesting by adding restrictions. Since moving a piece infinite times in a single turn is slow and dull, I don't move my save file to an auto-save slot. Calculating probability and playing the odds are fun, so I don't revert to my last save just because something went wrong. Exploring the whole world in one turn with two boats and a militia takes the fun out of it, so I don't use unloading to waken sentried ships. You get the idea.

On the other hand, if an exploit doesn't ruin the game, I'll use it even if it seems unrealistic. I'll take a fun game over reality any day. And if an exploit is hard to pull off, I'll be impressed if you can do it.

If you think of it this way, the game engine is a tool to make your own game on top of, like a chess board can be used to play chess, checkers, reversi, or any variation. There's no limit to the ways you can play Civ 1, and that's a good thing.

Now if there's a contest, you need to know exactly what you can or can't do. Otherwise, do what you want and have fun. There is no wrong way to play. Using a "cheat" in a computer game is not immoral, and shouldn't offend anyone. And when you post results of your games, be sure to say what rules you used.

This forum should have its own version of the Godwin, except instead of Nazis the discussion turns to playing preferences. I'd call it Civwin, but that's already taken.

To expound upon Urtica dioica's idea, several popular console games grew communities which developed restrictive challenges to extend play value. For example, the original Final Fantasy has what is known as a "one class challenge" where one must beat the game entirely with all characters set as one of several available character classes, the idea being you'll pick the weaker configurations, like White Mages, for a greater challenge. Once one masters this, he can move on to the solo challenge, where he plays with only one character instead of four. Over time, these groups develop strategies and tactical workarounds totally unimagined by early players or, at times, the creators themselves.

While I find these challenges interesting intellectually (especially the all Calculator challenge for Final Fantasy Tactics), artificial restrictions never appealed to me. I always fell like I'm pulling punches. If exploiting all the rules makes for a boring game, then I move on to something else. Holding back to make the game more "fun" is like walking on eggshells to keep an old flame around; I'd rather just find a better date.

Maybe there's a middle ground. Playing Civdos using zero aggression is one challenge I'm sure many of us have tried and there are probably a few of the many possibilities that many have tried re government types etc. These of course aren't cheats though.

The computer AI doesn't seem to use railroad, they still use rail roads as it it was normal road. My guess is if the AI did use Rail road as us it could loop a unit to move forth and back on railroad and never stop and then the AI turn never end?

This is a reverse cheat, that benefits us players

IIRC the AI still moves as far as it wants.

lndm said:

IIRC the AI still moves as far as it wants.

Click to expand...

What is IIRC? something to do with Internet Relay Chat?
I might be wrong, but I never seen a computer AI use the railroad bonus, for me it looks like they lose steps for each square they move even if it is a rail road?

IIRC = "If I Recall/Remember Correctly" in my usage.

With the similarity of some acronyms (or "initialisms"), for example (as you mentioned) "Internet Relay Chat," it can be difficult sometimes to know what someone means!

Yes, my bad. I decided some time ago that I had to get used to some, but I do draw the line at the basics.

I like IIRC and IMO (in my opinion) because it allows me to talk more freely without being afraid of making the odd mistake.

I didn't intend to sound critical--sorry if it came across that way. I was trying to empathize with C1v1l1z@t10n's difficulty understanding the "IIRC." I use it too, along with some other commonly used abbreviations. It makes for less typing!

It's all good, jmas