They actually just don't start out automated at all. They show up in the turn's unit queue waiting for orders so I have to press R to automate them.
Why are you expecting them to?
They don't show up in any of the cities at all, and the automated cities then go ahead and build them. It may have happened after a recalculation, but I'm not sure.
Could be a legit bug then. I'd have to evaluate a save that shows the problem to find it.
Oh! I meant that they do this even though my civ isn't in contact with any other civilization. No use for espionage then, after having conquered the continent. (Although arguably useful if I contact another civ?) Actually, I notice the player governor will sometimes set cities to build espionage instead of going for any more infrastructure if I have it emphasize commerce. Emphasize seems kinda broken tbh, at least in terms of affecting the governor's build choices.
I see. There are some ... assumptions in the code that the AI is playing for an AI player. As a result, there are also some accepted less optimal choices that can present interesting challenges for the player. No AI player would be in play if a human player wasn't (kinda the old if a tree falls in the forest bit). So obviously, espionage would always be valuable to an AI player, even if they haven't met you yet - they know you're out there somewhere or the game wouldn't be running.
The city AI for automation assumes you are not a human player, and if you are then you should not be automating. Where's the fun in letting a game play itself for you? (Can you tell I'm NOT a fan of automation?)
I don't blame anyone who plays with automation but I'm highly reluctant to work on it. There are far more important things to attend to here and once I'm finally done with those things, I'll be hard pressed to care to improve automation rather than moving on to another project instead. We might eventually find a programmer who is more aligned with your desires for the mod to be able to play a strong game against itself with as little human input as possible. Koshling was the one to program the city automation as being a slightly different animal from normal city AI for the non-human players. Personally, with all the goals I have and a task list of hundreds of fixes and balance items and projects, I don't think I could be persuaded to spend any time on it. Nothing personal. I wish we had the manpower and talent to answer to all requests.
I do, however, want the AI to be able to play effectively and strongly against human players. I'd also want it to, by default, beat the hell out of players that automate the majority of what is happening.
By 'AI' here I meant the player governor as well actually. I think the amount it was willing to rush increased over the game but I can't remember. I've basically committed to a strategy of outrunning inflation- it only increases expenses by 1% per turn, and I can scrimp and save with maintenance to lower the effect further. I'd rather automate the cities than try to manage them at this point- probably why I made it this far! And with this strategy I'd rather the player governor go all out and spend all the gold I give it to keep the wealth generation going up, instead of just rushing 15-20 cities. It will make for an interesting late game...
A dangerous gambit that can really restrict your options later. There's been some interesting discussions about inflation lately. I have not had an architectural hand in inflation much at all so I don't really know how it works much (I've been hearing a lot though. Most of my development has been on earlier game stuff and military matters.) Joseph might be able to comment more on this. As far as coding, I wouldn't know what would be optimal for gold rushing so I haven't modified it at all from the way it was originally set and would have a hard time wanting to mess with it without a better understanding of it. By that I mean I wouldn't want to try teaching an AI engine the best practices when I have no idea what the actual best practices may be, and from what you say you're doing, I do not suspect that is a 'best practice'?
Just tried it, doesn't work, even if I only select two cities. Only does it on the city it's shift-clicked on, or the earliest founded city with only alt or ctrl.
Ok. AIAndy set up the filters and I don't know enough about their coding or about multi-city selection coding to suspect I could add that functionality. I don't even know if it's in the python or the dll for that matter. Maybe Toffer would be able to share an insight as to how it could be done. I can see the point of doing it at least. That said, it's also not a really pressing need in comparison to many others.
Maybe if I bother to learn Python or C++ one of these days I could help.
That's basically what I did when I encountered C2C in the early days. I realized they didn't have a C++ guy and I'd always wanted to learn some programming so I dove in to see what I could figure out to help with some really serious bugs in religion options. It progressed a LOT from there but there's still a lot I don't know how to work with. My personal frustrations that nobody was there to help with what mattered to me drove me to learn and join the team so perhaps as I say I'm rarely ever motivated to help with automation concerns, it might frustrate you to learn and dive in too. If you did, I'd do what I can to help you with that. I'm working with someone now.
One thing to know, however, it is my understanding that Steam base Civ IV installs cannot properly work with C++ coding from what I'm told. It's possible that this is being disproven now but I'm not sure.
