Subsidies and Aggressive Trading Practices

Hm I gifted an AI 1gpt / turn but the available gpt did not go up. Does it only work when you gift 2 gpt? Or was it already at the max perhaps?

If their Available GPT was non-zero, they are probably already at their max. Sell them a resource and see what happens to their Available GPT.

If they were at zero, then they are in a defecit of unknown magnitude, and you'll have to give them enough GPT to go into the black (or green) before you can begin trading. Not recommended, being rather on the expensive side.


As for whether and how this was intended for the game, I would be perfectly fine if this was fixed up and/or replaced with a new and more comprehensive resource-trading model.

I can't say I've used this technique yet, but if it is deemed to be an exploit, I would at least hope they'd address the trading system in general. 3GPT for a resource that's worth 3X that is almost as bad as what's going on here (aside from the research stagnation inflicted on the AI).

It's not a guaranteed exploit. It relies heavily on what resources you have, what resources your target civ(s) has(have), and even then it requires someone having Currency (more likely the AI than the player will research it first). Plenty of game can go by before this is even an option. On Marathon, you could (and should) already be marching multiple stacks around in that time. There's still nothing as powerful as pointy stick economics.

And as for making any jump in level, it sure beats having to change your map settings in order to approach the difficulty level (3800BC and sooner Quecharush wins, anyone?). As I said, there's still plenty of game to play (unlike those aforementioned rushes).

Excellent post and good catch. I'd love to hear some observation on when to use this trick. It is always great to earn lots of gold, but the fact that an AI is getting 6 Happiness and 6 Health worries me.

Theory tells me that small civilizations are most vulnerable. They will generally have less resources, which means you can sell them more. Also, they will have less beniifit from the resources and be more crippled. On the other hand, I think (hope?) the AI scales its payment to the size of its empire.

A decent strategy should be finding your weak neighbour. Sell him lots of resources for maybe fifty turns. He should now be technologically ******ed - hopefully, you will fight cavalry vs muskets. Now, cancel all thsose deals and declare war. His cities should hopefully go into collective disorder, making the war yet simpler.

I would use it on everyone I could. But forced to choose, I would avoid using it on victims. Seeing a civilization with a positive GPT is somewhat unusual, so I would prefer not to cancel the deals, once they're sealed. Secondly, if you don't weaken your non-victims, then who will?

I think the best partner to aggressively trade with will be your best trading partner.

Lord Chambers said:

I think the best partner to aggressively trade with will be your best trading partner.

Click to expand...

I agree. From using this strategy the past few games, this is what i've found. The ideal partner for this is one with few resources, and a very high diplomacy rating towards you. Higher diplomacy rating means he'll pay more for each resource. And you don't want to ever cancel these deals and go to war with him. You want to milk him until the end of the game.

Very helpful- I've been searching for a way to bring in large profits to upgrade old military units. Very nice work!

I have one question on getting this to work-

Let's say hypothetically that based on friendship levels an AI player is willing to spend 10 per turn on a resource but currently only has 8 available. If I pull up the dialogue and credit them 2 per turn as a gift, should I immediately expect their available amount to go to 10, or would it only go to 10 on the subsequent turn? If the latter, it seems a little risky because I have no guarantee that the AI won't find its own use for my largess before I have an opportunity to make the trade at the higher amount.

As for being an exploit, there are enough dubious trade dealings in real life that this doesn't seem out of place. And as has already been pointed out, it's not that easy to consistently pull off. I don't see managing to finagle myself an extra 3-4 gold per turn a huge game breaker and I would expect that the OP's 100 per turn from trades is the exception rather than the rule.

Besides, would that make the following an exploit? Let's say I have 2 dye and 2 gold and a neighbor has neither resource. I gift the neighbor both of the dye to earn the gifting points and then I trade back one of the dye for a gold to get more points for fair dealings.

It will switch to 10 immediately.

Eqqman said:

Besides, would that make the following an exploit? Let's say I have 2 dye and 2 gold and a neighbor has neither resource. I gift the neighbor both of the dye to earn the gifting points and then I trade back one of the dye for a gold to get more points for fair dealings.

Click to expand...

The AI won't accept the 2 dyes. It will only accept a resource it doesn't already have, and will only accept one copy of it.

Eqqman said:

As for being an exploit, there are eniugh dubious trade dealings in real life that this doesn't seem out of place. And as has already been pointed out, it's not that easy to consistently pull off. I don't see managing to finagle myself an extra 3-4 gold per turn a huge game breaker and I would expect that the OP's 100 per turn from trades is the exception rather than the rule.

Click to expand...

It's not just an extra 3-4 gold per turn. Potentially, every duplicate resource (and even non-duplicates you don't need) could be making you at least 4 and upwards of 10-12 GPT. In a recent game, by the early middle ages (once all my resources were hooked up) I put together trades for over 50 GPT with 5 different AI's. Seeing as my total expenses were only about 100 or so, this was very significant. That would translate roughly into around 75 extra beakers per turn. When my total output is only ~400, you can see why this is so powerful. Without the "exploit" I proabably couldn't have been making more than around 20 GPT through trade. And, this was in a game where I was going for a cultural victory and purposely stopped grabbing territory and resources. If I had continued to grab extra resources, they all could have been turned into more cash.

It seems I can't cancel my gold subsidy to the AI, not even after ten turns?

Look in the diplomacy advisor.

I get the sense from other threads that some folks still feel this idea is an exploit, which shocks me. Since I don't want this practice to appear on the GotM banned list, I'll make a more expanded argument on why it cannot be considered an exploit (yes, that's 'cannot', not 'should not')!

[1] The most important point of all. To quote Leonard Nimoy (was Morgan Freeman unavailable?): 'Everything is worth what its purchaser will pay for it'. If Asoka is willing to pay 20 GPT for my bananas, I CANNOT have swindled him. He WANTS to pay me 20 GPT, so I would be an idiot to accept less. If he had 20 GPT on his own, would you charge him less for your bananas? Of course not. This point cannot be stressed enough and I will refer back to it often. If Asoka is happy to pay me 20 GPT, then the deal is FAIR by definition as both sides walk away satisfied.

[1a] I cannot get the best deals except from AIs that I have good relations with. Therefore I must have been treating them well over the course of the game (or, well compared to the others). I bring this up to further weaken trying to characterize the de-gifted trade as 'unfair'. I have expended effort to get decent relations and now I am just getting my payoff. Refer to [1].

[2] So, maybe I had to gift Asoka to get him up to 20 GPT. But, as we all know, the AI is incapable of consistently having enough GPT on hand to make trades that the human players will find acceptable. And what is a reasonable trade, anyway? I think we can probably all agree on amounts we consider a rip-off, but 'fair' amounts are totally arbitrary. We can waste time discussing what is a truly fair deal. And we can waste time coming up with brackets to what are fair price ranges, like saying military resources are worth more than happiness resources, which are worth more than health. Or vice versa, or any other scheme you want. However, Asoka has already decided that 20 GPT is a fair price: refer to [1]. So who am I to question my trade partner's judgement? And in all honesty, there is no real way to fix this. If Asoka needs three resources that he is willing to pay 20 GPT each for, is he supposed to keep 60 GPT on hand at all times in the vain hope that one or more players will give him what he wants? Of course not. We use the gifting to work around an inherent flaw in the trade model. If anything, I would go with the OP's opinion that the trade model should be changed. But once I got the hang of things, I like this current setup a lot. Anything different is as likely to be considered a worse system than a better one. I think the team at Firaxis has better things to do than revamping the whole trade model for a patch. Remember, I did not gift the AI to cheat him [1]. I gifted to help him have an amount he WANTED to pay in the first place and never would have otherwise (and under the circumstances, it is not reasonable to force him to).

[3] Now, when I cancel my gifts, the AI should be 'smart' enough to see that 20 GPT for my bananas is now a 'bad' deal. Okay, now we've reached an argument that appears reasonable on the surface. But scratch the surface and you'll find nothing there as this is based on a faulty premise. I think the premise that the deal is suddenly 'bad' stems from two ideas: one, that I ripped off the AI by having it pay 20 GPT for my bananas in the first place. Refer to [1] for this. Two, that since the AI is now likely running a deficit, I've 'crippled' it and it should be taking steps to get back to a positive cash flow. Well, so what? As we can see from looking at the diplomacy screen, the AI is frequently running under a deficit by itself and manages to crank it its spaceship out just fine all the same. It is fully capable of managing life under a deficit. The fact that this particular deficit was caused by me should be irrelevant. I am the player. It is my job to cause things to happen. So I influenced the AI to be in a deficit for a while. This is no more an exploit than tricking an AI into war to slow down its tech research or Wonder building. You are taking a step to help you win the game. That is what you are supposed to do. You have set back the AI's progress by getting it into a war, you have set back the AI's progress by stealing its Workers, you have set back its progress by causing a deficit. It is all part and parcel of mastering all facets of the game.

And I find the harmful effects of canceling the gifts highly debatable. I've had deals where after I cancel the gifts, the AI player was back up to a positive GPT the next turn. So clearly in this case no harm was done. Now put yourself in poor, victimized Asoka's shoes for a second. I'm running a positive cash flow and then suddenly I go negative. I have an advisor tugging at my sleeve telling me, 'Hey, we're paying 20 GPT for bananas! If we cancelled this deal, we'd be back in the black. I demand that you cancel this deal immediately!' But there was a reason I wanted those bananas in the first place. Let's say hypothetically that the bananas were giving me +1 happiness (I should have picked a happiness resource from the start, but we've got bananas. Work with me). Am I willing to have every city in my empire generate an extra unhappiness just to avoid a deficit for some period of time? Who is to say. And what if I decide that I can deal with the extra unhappiness. Again, who is to say that my partner will be willing to trade the the bananas once more in the future? The deal may be impossible to renew with any partner. The same advisor could also be tugging at my sleeve saying, 'Hey, we negotiated this banana deal when we were best buds with this guy/gal, but things have cooled a lot since then. We're only willing to pay 10 GPT for bananas now. Cancel this deal immediately!' Just as with the previous logic, I cannot assume that I can successfully renegogtiate a banana deal at what I consider a better price.

Malekithe brings up a case that looks like a genuine example of economic warfare. But I think this only reinforces my point. Arguing over the 'badness' of a single de-gifted deal is debatable. If I've made 3, 5, or 10 deals with the same player, the deleterious effects of canceling the deals are sure to be huge. I may be able to handle canceling one deal that provided happiness. Can I cancel three? Can I cancel the horse deal that is giving me Conquistadors? Can I cancel the aluminum deal that is giving me a spaceship? Can I cancel the coal deal that powers all my Coal Plants? Who are we to tell Isabella how to manage her empire. We cannot call all de-gifted deals 'bad' because this is highly subjective. And I don't buy the logic that if one or two out of my deals MIGHT be genuinely bad, I should not make ANY deals that are sure to leave the AI in a deficit out of some distorted sense of fair play. I can assure you that your AI opponents are not burdened with such scruples. And what if I only make deals when the AI has positive GPT on its own, and never gift it? Are my deals now bad because it happened to run into a deficit by itself? This sword cuts both ways. If we force the AI to prune its deals when it hits the red, it is going to cancel trades regardless of how they were derived. No thanks.

[4] Some folks (I'd call them total wackos, but I'll be charitable) will argue this 'has' to be an exploit since a crafty player will now have X more GPT than they normally would at some stage of the game, so it is now 'unbalanced'. Well, who are you to say how much money a player is 'supposed' to be earning? This same player might have gotten the same amount by using an Organized leader. Or Financial. Or built more cottages. Or optimized Palace/Forbidden Palace placement. Or did a better job optimizing Wall Street. Or got more use from Great Merchants. Or any combination. They picked trading. Big deal. It's all part of playing the game: you look for any way you can to increase your assets. And as Zombie69 pointed out, here's a great way to help you out at higher levels of play. I'd think that the peaceful types would be pleased since they have a new way to try and get an advantage that doesn't call for constant warmongering.

[5] Of course, there will be some of you that read this and think, 'that all sounds reasonable, but it still feels like an exploit. Well, to those folks I say, refer to point [1].

Just my 2 cents. Or if I gift you first, my 20 cents.

Eqqman said:

[1] The most important point of all. To quote Leonard Nimoy (was Morgan Freeman unavailable?): 'Everything is worth what its purchaser will pay for it'. If Asoka is willing to pay 20 GPT for my bananas, I CANNOT have swindled him.

Click to expand...

The entire definition of an exploit is something that takes advantage of the programming of the game to get an advantage that is not intended or appropriate.

Suppose that I discovered a "strategy" where, if I make a certain deal, then cancel it, then make some other deal, I can trigger a situation where Asoka is willing to pay 1000 gpt for those Bananas, or all of the money that he can generate from his entire economy by running at 0% research, in exchange for one resource. You're saying that that can never be considered an exploit, because the AI agrees to the deal?

My feeling is that this pretty much breaks the game, which is why I haven't been doing it (before or after it was posted here). You're arguing that because the game lets you do it, it isn't an exploit. But that applies to every other exploit that's ever been discovered in any version of Civ: they are all things that the game does let you do.

E.g., trading cities to the AI that are bad for it (but it wasn't smart enough to know that) was an exploit, yet your point #1 applies equally well to that.

DaviddesJ said:

The entire definition of an exploit is something that takes advantage of the programming of the game to get an advantage that is not intended or appropriate.

Click to expand...

I think you're proving my point. My contention is that the current evidence points to the fact that Asoka was properly willing to pay 20 GPT for my bananas all along. However, because of issues I've already mentioned it's never going to be likely that he has that much GPT available at any moment. The Firaxis team wanted him to pay 20 in an UNsubsidized deal all along, so it cannot be an exploit to employ a method that achieves this goal. I'll be honest and admit that because it does not make sense to force the AI to keep its wallet full, I don't know how they ever expected players to make a max trade with the AI. But because of the fact that a max trade exists, employing it can not be an exploit since it's a properly coded part of the game. And who is to say that subsidies are not in fact the intended method anyway?

Eqqman said:

My contention is that the current evidence points to the fact that Asoka was properly willing to pay 20 GPT for my bananas all along.

Click to expand...

What's "proper" is a matter of opinion. I think the designers coded in an upper limit for how much the AI would pay for a resource, with the idea in mind that it would never actually pay that much unless it happens to have the free cash flow. You're exploiting a loophole in the latter constraint.

Neither of us can prove our ideas, and I don't see any "evidence" one way or the other. The main question is what is the actual effect on the game. My own opinion is that it's quite powerful, to the point that players not using it won't be competitive with those who do. Since I see no upside to allowing it (making the game easier just means we need even larger handicaps to make the AI opponent challenging), I would prefer not to do so.

DaviddesJ said:

My own opinion is that it's quite powerful, to the point that players not using it won't be competitive with those who do. Since I see no upside to allowing it (making the game easier just means we need even larger handicaps to make the AI opponent challenging)

Click to expand...

Big deal. Some players are better than others, and expert players always know things beginners don't. That's why they are experts. The whole point of this forum is to put the advice of the experts in the hands of everybody. It seems like a silly argument to me to say that everybody should be equally competitive, that's not how life works. You sound like a scrub (which I say in a joking fashion, something you can't pick up from text).

I also don't seem to hear people on Deity complaining that the AI is not challenging enough. I can't imagine a player out there seriously handicapping themselves in a game, unless they are playing a lower level maybe. If so, simply move up a level.

E.g., trading cities to the AI that are bad for it (but it wasn't smart enough to know that) was an exploit, yet your point #1 applies equally well to that.

Click to expand...

I'm using an economic themed argument to argue an economic themed case. Anybody can misapply an argument, but using a hammer as a screwdriver doesn't mean it is no good on nails. Maybe I was too witty using the tech quote, but I feel the essence of it is true.

There is zero reason to code a hidden limit when the current GPT for the start of the turn (excluding gifts) could have sufficed. It makes no sense to waste time coding a limit if you seriously do not expect the player to reach it. You give an example that seems clearly erroneous (if you could find a way to get 1000 per turn, for example). In practical terms I've yet to have more than 27 accepted for a resource. This seems like an 'appropriate' value to me, but as I pointed out, the 'true' value of a resource is so subjective that it does not merit serious discussion. Keep in mind that the limit is also clearly tied into your relation with the AI. Clearly meant to be a practical, reachable limit. A limit you expect players to never reach does not need to go up or down based on the player's relations.

Eqqman said:

I get the sense from other threads that some folks still feel this idea is an exploit, which shocks me. Since I don't want this practice to appear on the GotM banned list, I'll make a more expanded argument on why it cannot be considered an exploit (yes, that's 'cannot', not 'should not')!

Click to expand...

Like DaviddesJ said, it's an exploit because it wasn't intended by the programers. That's the very definition of an exploit and there's no way around it. Of course, that won't keep me from doing it, and indeed there are other exploits that i'm already using (for instance, the pop rush exploit described in my article). As for GotM and HoF, i couldn't care less what they allow and what they ban. I find them totally laughable anyway and have no interest in trying them out.

Eqqman said:

[1] The most important point of all. To quote Leonard Nimoy (was Morgan Freeman unavailable?): 'Everything is worth what its purchaser will pay for it'.

Click to expand...

If the AI had been coded properly, it wouldn't pay that much for it. A human will assign different value to a resource based on how much money he has. With the money you now have, buying a Mercedes (with a loan if you have to) may not be worth the money it costs. After you win a million, the price asked may now be a fair price t you. The actual value you give to a product depends how how much moeny you can afford to put on it.

"What its purchaser will pay for it", in the real world, depends on how much money said purchaser has. An AI should be willing to give you 20 gold for a resource that it needs when he has said money, but shouldn't be willing to give you 20 gold when he doesn't have it. It should then be worth a lot less to him. That's how the programers intended the AI to work, but a loophole works around this limit. This loophole is obviously an exploit.

Eqqman said:

[1a] I cannot get the best deals except from AIs that I have good relations with. Therefore I must have been treating them well over the course of the game (or, well compared to the others).

Click to expand...

Totally worthless argument, since the gifting itself makes the AI happy with you. Not only are you taking away all its money, but on top of it you're making him even more happy with you in the process. Another clear sign that this wasn't intended, and is therefore an exploit.

Eqqman said:

However, Asoka has already decided that 20 GPT is a fair price: refer to [1]. So who am I to question my trade partner's judgement?

Click to expand...

20 may be fair when you have the money, and be totally not worth it when you don't have the money. The AI was programed to refuse trades it didn't have the money for (like any sensible human would), but we found a way around this, through an exploit.

Eqqman said:

I also don't seem to hear people on Deity complaining that the AI is not challenging enough. I can't imagine a player out there seriously handicapping themselves in a game, unless they are playing a lower level maybe. If so, simply move up a level.

Click to expand...

I play Deity and don't find the AI challenging enough. It becomes challenging enough when i add "no tech trading". When i get better at the game, this may not be enough anymore to make it challenging.

Zombie69 said:

Look in the diplomacy advisor.

Click to expand...

I can see all the deals I have with them, by clicking them I get a message saying I cannot end this trade or I get the choice between ending it or not. This does not happen when I click one of my gold gifts.

Or is this the other diplomacy advisor?

There is only one diplomacy advisor, and it does work. Figure it out.