Version 0.41 Discussion Thread

[glider1]

@Gatling
Don't forget giant vortexes, difficult currents and fear about Moby Dick and monsters. Seriously clearing those up is cultural knowledge no? Hey, I like Yakk's verbal explanation of this exception case, but you can explain anything if you try hard enough yes!
Cheers.

 

[PieceOfMind]

Hey guys I've been away for a while but now I'm back I'll be trying out 0.41.

I've already noticed at one point the gpp bar wasn't displaying but it appears to be working now. Not sure if it was reproducible or not.

Can't wait to see air combat!

 

[scu98rkr]

Hey Guys,

Im playing with this version and I cant seem to get Uranium to come up on my list of resources.

Do I need a certain tech ? I already have the tech to make Uranium appear and a Mine has been built there.

The Uranium is actually in the land for 2 vassals but doesnt appear in there resource list either.

 

[PieceOfMind]

Hey Guys,

Im playing with this version and I cant seem to get Uranium to come up on my list of resources.

Do I need a certain tech ? I already have the tech to make Uranium appear and a Mine has been built there.

The Uranium is actually in the land for 2 vassals but doesnt appear in there resource list either.

You need to research Fission. That tech enables you to have "access" to the resource. Physics just lets you find and mine the resource (for the bonus hammers).

 

[scu98rkr]

ta 4567890

 

[Psyringe]

Regarding worker behavior: My automated workers seemed to put a rather high priority on putting a fort on an oil resource. I'm not sure whether this was a smart decision by them. The resource was in an unthreatened area of my mainland, I couldn't use it anyway (no combustion), and I still had unimproved tiles around other cities. I'd probably rather improved those instead. But maybe I'm just not understanding the importance of fortifying resources, and perhaps the other cities that needed improvements were too far away - I don't know how far the worker AI can see when it's looking for a new task.

Edit: Also, unrelated, another observation: I've seen many "orphaned" units in my game, especially on small islands. Usually these are previous invasion forces which didn't succeed and, after peace was declared, were pushed out of the enemy's borders, possibly onto a small island. And there they stand - until the game ends. I've seen some swordsmen or macemen standing around on remote islands for hundreds of turns (marathon game). Question: Can the AI be made to pick those forces up again? If not, wouldn't it be better to simply destroy these units, or doesn't the AI pay maintenance for them?

 

[armand453]

Edit: Also, unrelated, another observation: I've seen many "orphaned" units in my game, especially on small islands. Usually these are previous invasion forces which didn't succeed and, after peace was declared, were pushed out of the enemy's borders, possibly onto a small island. And there they stand - until the game ends. I've seen some swordsmen or macemen standing around on remote islands for hundreds of turns (marathon game). Question: Can the AI be made to pick those forces up again? If not, wouldn't it be better to simply destroy these units, or doesn't the AI pay maintenance for them?

Yes me too I've already notice the problem here http://forums.civfanatics.com/showthread.php?t=290036&page=2, it only happen on island where you have less than three city, so in clear where these area aren't considered as primary area by the AI, its very annoying because I have to open the world editor to erase the city because most of the time AI consider he has a good number of unit and don't create new one, did you find what was the problem Jdog?

Armand.

 

[Yakk]

That argument would make sense if the tiles in question were coastal instead of open ocean... ... coastal waters have navigable features... open ocean simply has big waves which can damage small boats.

They also have currents. Probably there are times of the year where it is safer to cross. And maybe there are some little known islands in the gulf that can be used for resupply.

And maybe there are tricks to building sea worthy boats that last long enough.

In order to cross the gulf, you need to know the currents, you need to know when it is safe, you need a local seafaring population that knows how to make boats that are good enough to cross.

 

[PieceOfMind]

I'm not that great a bug finder - usually the only problems I notice are with how the city governor assigns citizens. Naturally, I've observed issues once again with the citizen automation.

I understand that the automation code is not the easiest thing in the world to tinker with due to complexity of the algorithms and the sacrifices that must be made for speed and space, but I think the problem here is unusual.

Basically I have noticed cities working ocean tiles before coast tiles. In this game, playing as Victoria (financial leader), I was losing 2 base commerce for every ocean tile the automator chose to work instead of coast tiles.

Here's a screenie of the city Coventry. It pretty much doesn't matter what emphasise buttons are pressed, the governor still picks that ocean square to work. Picking emph food or commerce both cause the automator to take the citizen from the iron mine onto the ocean square (making matters even worse), and all else left the same. Emph production does a better job unsurprisingly.

I would hope that this is a bug because I don't remember the automator being this poor - previously it only had issues when the picking of optimal tiles was more complicated.

Having said all that, keep in mind I am running 100% culture. I notice that if I increase either tax, science or esp up by 10%, the citizens are taken from ocean tiles to fill up the coast tiles first. So is this working as intended? Why would the automator pick tiles with less commerce only when it's all going into culture? Is there a something in the code that checks for net increases in beaker or gold income (when checking for optimal tiles) but neglects to consider culture income? <-- That is what I believe it will be.

Note the exact same effect can be observed in York - going from 100% to 90% culture takes a citizen off an ocean tile (when there was a coast tile available) and puts it on a plains town.

For reference, I'm using the latest version of Better BTS AI (ver 0.41B I believe). No other mods.

 

[Jaybe]

Regarding worker behavior: My automated workers seemed to put a rather high priority on putting a fort on an oil resource. I'm not sure whether this was a smart decision by them. The resource was in an unthreatened area of my mainland, I couldn't use it anyway (no combustion), and I still had unimproved tiles around other cities. I'd probably rather improved those instead. But maybe I'm just not understanding the importance of fortifying resources, and perhaps the other cities that needed improvements were too far away - I don't know how far the worker AI can see when it's looking for a new task.

Would you like to have oil available as soon as you get Combustion? If so, you MUST have a fort on one of your oil sources by then, because oil wells are not available until the Combustion technology!

I personally ALWAYS do it when available, particularly because I play at Marathon speed.

 

[Cybah]

as you can see on the screenshot (AI autoplay), the AI is not using its gold ressources in an optimal way.

edit: the AI now has 6600 gold without any use of it... still increasing.

 

[Psyringe]

Actually I don't see a single gold resource on the whole screen. I also don't see any AI areas on the whole screen since the whole area is clearly Zululand, which (according to the scoreboard) you play yourself. Maybe the BetterAI programmers can read your mind better than I am able to, but unless they have such powers I'd recommend to make a bit clearer *where* in the screen the information you're referring to is located. There's a *lot* of information in that screenshot which is apparently irrelevant to the point you're trying to make, so a simple MS Paint job (such as drawing a couple of red lines) might work wonders.

Edit: Seeing the note about "AI autoplay" now. Still not seeing gold resources. From what you've said in your edit, do you perhaps mean gold *reserves*, i.e. the amount of money that a civ has?

 

[Cybah]

oh yes reserves sorry Im not English

 

[Psyringe]

No problem, money *is* a resource in the general sense, I was just led up the wrong track by the fact that Civ uses the term "resource" in a more specific way.

Getting back to the question whether the AI is handling its gold reserves suboptimally in the game you're watching: I'm actually not sure whether that's the case. The AI doesn't curb research (or anything else) in favor of money production, only 10% of commerce are used to produce money. The AI's units (as far as I can see them) are apparently upgraded. And the AI probably can't rush production yet since it's unlikely to have Universal Suffrage as a form of goverment at this point - or has it?).

So what would you regard as optimal behavior? Do you think he should run a deficit to speed up research by 10% more? I'm not an expert, but I don't think that would be smart, given that he'll want to upgrade his units again sooner or later. He could also use the money in trade arrangements but I think the AI is programmed to be careful with that in order to prevent exploits. So what course of action would you recommend for him?

As long as he does use his money to upgrade his units once he can, I don't see much of a problem. Doesn't mean that it can't be further optimized of course.

 

[Cybah]

And the AI probably can't rush production yet since it's unlikely to have Universal Suffrage as a form of goverment at this point - or has it?).

first: the AI should research with 100% (here 80%+10% espionage = 90%) for several rounds. the reserves are too much.

or the AI could switch to universal suffrage for 10 rounds to get rid of the money.

and: with that amount of gold the AI could probably get another AI to start a war.

 

[scu98rkr]

Again Cybah Im going to disagree with what you think is the optimal strategy and agree with Psyringe.

I think the AI having large amounts of money as a back up isnt such a bad idea. At least it means it has money for upgrading/random events/diplomacy when it needs it.

 

[Cybah]

for this reason human players always have the newest technologies (without tech trading)

 

[scu98rkr]

yes but just play on a higher level.

Although the AI needs to be better we should try and keep it as simple as possible.
This feature is not desperately needed, the AI can tech reasonably quickly if you use the right difficultly level for your style of play.

Every time more code is added in it makes the game slower, the code harder to debug and adds more chance for mistakes to be added.

I think the Better AI team needs to be very careful where it tries to add new features and should only add them where really needed.

For instance at the moment I would say the suicide stacks are where the biggest improvements in the AI can be made.

Making the AI a bit more efficient with gold spending / improving the Great General AI will add a more complexity for little advantage.

 

[Psyringe]

Just to clarify, I'm not saying that the AI *should* build reserves of the size reported. In fact I'm simply not sure about it. For me it's hard to grasp mathematically how the strategy "put everything into research to get a tech the fastest way possible, then make money for a while if you need to upgrade units or buy things etc." compares to the strategy "Always put a bit of commerce into money so that you have reserves when you need them".

Intuitively it seems very plausible that the first strategy is better, but for things like AI programming I'd like to look beyond the layer of intuition and get some more proof that a suggested algorithm is actually superior. That's what I was (am) asking for.

Besides that, there's also the more pragmatic question of whether the AI can effectively deal with something. If, for example, a reduction of the money reserves in favor of higher beaker output would man that the AI is then incapable of upgrading its units properly, or if it means that it would make the AI's decision algorithm of whether or not to build a money reserve inappropriately complicated and inefficient, then leaving things as they are may indeed be the better option. If, however, a simple change like capping the money reserves would improve the AI's performance, then I'd definitely say go for it. Presently I just don't know enough to have a well-founded opinion on this, so I'm merely raising questions.

 

[r_rolo1]

Well, there are definitely parts of of the game where stockpiling cash and then run some turns on 100% make sense, like after having libs or univs up ( if , OFC the AI makes them roughly at the same time, a thing it should do IMHO ). But in terms of the whole game...... I'm like Psyringe : I can't see clearly wich one is best, especially because of the reasons that AI stockpiles money, and why humans don't do it as much ( AI rely on upgrading to have a up-to-date army in a scale that humans never do and humans tend to not have much money in the vault to avoid AI money demands ).

The best thing to do there would make a test AI with those features and see how it goes......