The AI and its obsession for food

[nbarclay]

Originally posted by alexman
I don't think the double entertainer value changes strategy that much for the human. I really don't. At the same time as the AI reached the point of the screenshot, my single city was size 7, full of improvements, and much more productive - mostly because of my use of the luxury slider. I never used a single entertainer. As Nathan pointed out, the food lost from assigning entertainers is a steep price to pay.

It's not just the food. It's also the production. A city with seven laborers working tiles is up to seven sixths as productive as a city with six laborers working tiles and an entertainer. And in my experience, in 4X games, production is king because that's the key to getting everything else that matters.

Nathan

[nbarclay]

Originally posted by lockstep


No, you aren't. Making entertainers more powerful is very likely to make building happiness improvements an option instead of a necessity, and I wouldn't like that at all. Seems like a violation of the core game design to me.

The luxury slider already provides a fair amount of flexibility in terms of when to build happiness improvements, at least in core cities, and eight luxuries and a marketplace (with the right trades) can keep a city from rioting past size 12. But with or without the change in entertainers, relying on entertainers rather than on happiness improvements or luxuries as a permanent solution is wasteful. I contend that the differences involved would be more a matter of degree than a matter of kind. Players could hold off building happiness improvements longer in some situations, but would still pay a price (albeit a smaller one) for doing so.

Nathan

[Pyrodrew]

Double power entertainers do make happy improvements & luxuries less valuable. On the other hand aren't the humans usually the ones with more happy improvements & luxuries than the AI? Does anyone see a logical reason to not use happy improvements or the luxury slider anymore because of double power entertainers?
Unfortunately, doubling entertainer power also makes 'unhappy losses/dangers' less of a threat. If everyone embargoes you away from your 8 luxuries & bombards your Cathedral to rubble suddenly you now have more citizens able to keep working for you as the loss of those luxuries & Cathedral are now not as great. It devalues them in that aspect as well.
However, isn't the AI usually the victim of that situation rather than the human?
The AI also uses entertainers far more than humans (the reason I've always been tempted to double their power). So where we use the luxury slider (trade (science/gold) cost) to gain more able working citizens, the AI uses entertainers (shields/food/trade cost). Double power entertainers, therefore, would make the AI more productive in all cities.
Therefore, from what I see it helps underdog victims & the AI.

At the very least I think it deserves a full game test.

[lockstep]

Originally posted by alexman
In this test, 1 entertainer gives 2 happy people and nothing else. 1 laborer gives you 2-3 food, 2-3 shields, and 2 commerce. Using the luxury slider, the 2 extra commerce from the laborer has exaclty the same effect as a double-value entertainer, but you get to keep the extra food and shields.

Hmm ... To get a total number of 7 food/shields/commerce per laborer, you need railroads, a trade bonus and solely bonus grassland/hills/floodplains (3 base F/S/C). Without railroads and a trade bonus and assuming 50% grassland/plains, the total number of food/shields/commerce per laborer is about 4.5. Still, you may be right that the luxury slider is more powerful than double-effect entertainers.

OTOH, you can't adjust your luxury rate per city, so you 'waste' some of your luxury spending most of the time (that is, besides improving your score somewhat). What if the combination of a small (10-30%) luxury rate and double-effect entertainers for some cities allows for an effective strategy to hold off building happiness improvements for a long time (let's say until size 12 cities) even if there aren't many different luxury ressources available? And holding off improvements would allow for a bigger early military and lower upkeep costs ... This might well create an entirely new game, as Dominae put it.

[alexman]

Originally posted by lockstep


Hmm ... To get a total number of 7 food/shields/commerce per laborer, you need railroads, a trade bonus and solely bonus grassland/hills/floodplains (3 base F/S/C).

You don't need 7 resources to beat double-happy entertainers. Even a 1/0/2 coastal tile without a harbor will do it if corruption is low (because of the single extra food).

[Dominae]

I got to thinking that maybe the best "fix" to this problem would be to actually reduce the effectiveness of the luxury slider. I think we can all agree that using it is better than using entertainers 9 times out of 10 (or more). I was thinking something along the lines of transforming the luxury slider so that it can only transform unhappy citizens into content ones, or only creating one happy face per two gold. However, as far as I can see, there is no way to do this in the editor.


Dominae

[Fitz]

Alexman, I want to apologize in advance, just in case you found my rather direct comment about the flawed experiment insulting. I was tired, thought I saw a problem, and stated it poorly last night.

Originally posted by alexman
As I said, maybe it doesn't represent a typical game situation, but it does reveal some areas where the AI needs some help.

I really respect the fact that you are trying experiments to try and determine the nature of the game. That's actually one of my favorite Firaxis games activities (actually the analysis of other's experiments ). Anyhow, I felt that in this case, the experiment might need some tweaking to make it more like an in-game situation.

As Pyrodrew said, the does AI build units, which also cost support. As for having only one city, you might have less income, but you also have less expenses (you have only one of each building to support). Actually, because of corruption, your income/expense ratio is greater when you have one city!

Actually, I will concede this point. It occured to me relatively shortly after posting last night. I think that the AI certainly has a food priority, but I think it is related to how many cities the AI has and how many optimal cities the map allows for.

Just theorizing out loud here, but I strongly suspect that the AI food priority will remain so long as it has a significant number of cities less than Optimal.

Even if it doesn't, do the codes that Soren used work around future income posibilities? Does it restrict spending based on the likelyhood of increasing income later? I would hope so, but unless Soren explains (or opens his codes) it may be hard to discover.

As I mentioned in the first post, I reduced the tech cost, so everyone was discovering new techs after 5-10 turns. I also tried reducing the optimal number of cities to 1, and got the same results.

Missed both of these parts, which potentially invalidates my theory. Another theory that I will posit is that in a single city environment, the only way to expand income is to expand the population. Until then, you have very limited income at your disposal, and if the coding looks at future income potential (via expansion), the limit will known to be very very low. If this is the case, expanding income to the maximum is a priority and buld options must be carefully considered. You say the AI built units, but it sounded like they were (primarily) defensive units.

Probably the most relevant question is income enhancing buildings. Does the AI build these? Also, if you add 3 luxuries (making a Marketplace more attractive), does the behavior change.

Anyhow, that's enough rambling for now. I'll let you get back to playing with various options. Just in case I'm not around enought to really participate in your experiments analysis, I wish you the best of luck.

[Pyrodrew]

What if the combination of a small (10-30%) luxury rate and double-effect entertainers for some cities allows for an effective strategy to hold off building happiness improvements for a long time (let's say until size 12 cities) even if there aren't many different luxury ressources available?

That seems risky to me.
1st, you lose culture tied to those happiness improvements, so one cannot complain later if cities start to flip because they did this.
2nd, that makes the city dependent on those entertainers. Entertainers who instead could be producing food(growth)/shields(production)/trade(science/gold). In the short run that may provide more science/gold because of a lower luxury bar, but in the long run I believe you would still be better off with happiness improvements than double power entertainer(s). I don't see that as an effective strategy over a long time, but it sounds like it could be over a short time... don't know only game play testing will prove it.

However, I have used entertainers to achieve WLTKDs, which I don't believe the AI knows how to do. Double power entertainers would make this tactic easier. Could that advantage surpass the AIs advantage from double power entertainers?

[FrustratedPoet]

Soren speaks on AI food production:

http://apolyton.net/forums/showthread.php?postid=1432941#post1432941