1987

Again, while very limited in what AI behaviors he can really adjust with out having the actual source code, this dude, Delnar, has mad some decent progress in his AI Cleanup mod (like getting the AI to expand more, be better with upgrading, and having it more likely disband obsolete units it can’t upgrade):

v6 - 2016-Oct-30

  • Fixed a bug that was causing the AI to not move settlers around properly

v5 - 2016-Oct-29

  • Added the Combat AIType to Scouts and Rangers so that they can participate in operation slots that only call for a unit with some combat strength (e.g. settler escort)
  • Made the cap removal on settling operations properly work by assigning it a significantly high cap of 16 instead of deleting the cap entry, which would cause a cap of 0

v4 - 2016-Oct-29

  • Altered the way Settler flavor is increased so that it works properly now hopefully (done through a permanent favored list entry instead of through the pseudoyield’s default value).

v3 - 2016-Oct-29

  • Added a special thanks section to the mod’s description
  • Fixed a bug that stopped the behavior tree edit for upgrading units from working
  • Increased Settler flavor by 50%. This is done modifying settler pseudoyield’s default value, because apparently “pseudoyield” is what flavors are now called

v2 - 2016-Oct-29

  • Fixed a lot of the mod’s changes not loading properly (thanks to RushSecond and qqqbbb, kind of shows how rusty I am with non-DLL modding)
  • Reverted the addition of Cavalry UnitAIType tag from Indian Varu because they don’t have high movement (thanks to notque for pointing this out)
  • Altered the priorities of certain unit move types to make the AI pillage districts and traders more, attack civilians less, attack encampments less, use air attacks more frequently, and (hopefully) properly make armies and corps.
  • Altered city attack operations so that they are less saturated with melee units
  • Altered simple city attack operation to allow for a lot more ranged units (raised maximum cap from 3 to 12)
  • Altered the Upgrade Units behavior list so that the AI can disband units that it would normally wish to upgrade
  • Allowed the AI to run one extra city defense operation for each war they are in to help defend against simultaneous attacks targetting different cities
  • Removed the cap on settling operations so that the AI can have as many settlers out simultaneously as it wants
  • Added a new operation that should function like an annoying harassment operation: one or two cavalry units should run around in enemy territory, pillaging districts primarily
  • Note: if you want to see an example behavior tree that’s commented, this is what you should check out

v1 - 2016-Oct-21

AI-only changes:

  • Added the Explore UnitAIType to Spies because they can be used for scouting
  • Added the Ranged UnitAIType to Rangers because they have a range 1 ranged attack (like Slingers)
  • Added the Civilian UnitAIType to Naturalists because they have no combat strength
  • Added the Explore UnitAIType to Modern AT because they were the only melee land unit missing the entry (this just means the AI can use them as scouts if it wants to)
  • Removed the Explore UnitAIType from Modern Armor and replaced it with the Melee UnitAIType to make them consistent with all other Heavy Cavalry entries
  • Added the Civilian UnitAIType to Military Engineers because they have no combat strength
  • Added the Build UnitAIType to Legions because they can build the same improvements Military Engineers can, so the AI should be able to use them as builders
  • Added the Cavalry UnitAIType to Barbarian Horse Archers, Egyption Chariot Archers, Scythian Horse Archers, American Rough Riders, Russian Cossacks, Indian Varu, and Arabian Mamluks because they were all missing despite having cavalry movespeeds
  • Added the Explore UnitAIType to Barbarian Horse Archers and Scythian Horse Archers to make them consistent with the light cavalry roles in which they could act
  • Added a Ranged Promotion Class build preference for Tomyrisi because Scythian Horse Archers have that promotion class (not the light cavalry one)
  • Added a new, generic CitySpecialization to the city specializations list that is meant to act as a stand-in for split food-production preference
  • (Hopefully) altered the gold spending priorities of the default savings list so that splurge spending is rated less important than unit/upgrade spending
  • Tweaked flavors in the default flavor list to slightly value expansion over offense and value active defense over passive defense
  • Altered the preferred yield focus of the “build trade units” specialization to be production instead of gold because that would end up producing the trade unit(s) faster
  • Reduced current task bias from 80 to 50 to hopefully reduce AI railroading as much
  • Reduced turns to wait between trade offers from 10 to 3 to make it consistent with turns to wait between peace offers
  • (Hopefully) Reduced delay between denounce and DoW to allow the AI to truly execute surprise attacks

Changes with some non-AI effects:

  • Added the Ranged TypeTag to Rangers because they have a range 1 ranged attack
  • Added the Melee TypeTag to Garde Imperiale because they were somehow missing it
  • Added the Naval Ranged TypeTag to Privateers, English Sea Dogs, Submarines, German U-Boats, and Nuclear Submarines because they all have ranged attacks
  • Increased the fertility rating of bonus resources from 1 to 2 to correspond to their usefulness vs. other fertility ratings (strategic resource = 3, river = 3, coast = 1)
1988

@LordGek Have you tried it? Is it any better?

1989

[quote=“Eric_Majkut, post:1888, topic:78555, full:true”]
Where is Firaxis in all this? For example, Stellaris came out with some issues and the devs were all over that. Blog posts, streaming multiple times per week, all over social media talking about the problems and what they were doing to fix them. Firaxis can’t even be bothered or what? Like, how about instead of Reddit having to reverse engineer the game to find out how basic game mechanics work, they maybe just put up something somewhere explaining it? Maybe something saying “Hey, we know about x y and z bugs, we expect a patch next week”. Anything? Hello? Wtf?[/quote]

I’m sorry, but I bought Stellaris, and that’s bullshit.

Paradox lately is really good about talking about how they are going to do something to fix things.

With Stellaris, they’ve talked about it quite a lot. What we have actually received, on the other hand in the roughly six months since the game has been released is the option to buy another DLC, for a game that really is substandard. And if you buy that new DLC, you’ll have a game that they have more or less admitted still really has no mid-game or end-game content.

Don’t talk to me about Paradox.

1990

Yes there’s a DLC, there’s also been three very large free updates. You can say that you think the game still sucks, but saying the Stellaris dev team is just talk and no action is absurd.

Besides, I would be shocked if there’s no Civ6 DLC within six months as well, just like I’m sure they’ll have free patches.

1991

I am stumbling my way through my first game to figure things out, playing as Rome on Prince level. Rome is a powerhouse Civ, as all their abilities are very useful for any game style and give you a serious boost early on.

So far I think they have a solid Civ game here, but my biggest complaint is with the UI and especially the info screens. It is hard to find the information you need to make city development/production decisions–in many cases the information is presented poorly or is just nonexistent. There is serious detail lacking in the tooltips and Civopedia. These flaws are quite fixable, and I trust many of them will be mitigated in future patches, but it is frustrating when interface/report problems that were solved in Civ5 recur in Civ6.

I’m sure the AI and some of the game balance needs tweaking, but that is hard for me to discern on vanilla Prince level. I will go ahead and say that it would be nice to have some more options to deal with missionary flood.

1992

Large free updates that really haven’t addressed the fundamental problems with the game. And paid DLC before they addressed the fundamental problems with the game.

Whether Firaxis dumps paid DLC before fixing problems with AI, etc. is besides the point. That’s a little bit like my neighbor beating his kids making it okay for me to do it too.

All I’m saying is, I wouldn’t hold Paradox up as a model of virtue because of Stellaris.

1993

Thank you for the heads up on the AI mod. I thought I would paste in Delnar’s post regarding developing a good AI and why 2K (Or other developers) wont spend the $ on it since it gets discussed here so often.

Here is the post from Civ Fanatics Forum::

"An average videogame programmer earns roughly 100k USD a year and gets treated like dirt. An AI expert can easily ask for 200k per year, and if Firaxis/2K start giving them lip, they can easily jump ship to a company that doesn’t treat them terribly (Google, Facebook, and Intel are always on the lookout for this kind of talent); in real terms, this usually means that if said expert would have to work the same schedules as videogame programmers, they would actually ask for something around 300k per year at least (this includes overtime and crunch that is standard for videogame development). For that price, 2K can just hire two generic programmers to make a superficial OK-looking AI (people don’t really have high expectations), spend a bunch on a celebrity voice actors who players will immediately recognize to sell more copies, and still have money left over.

Most AI experts work best when they’re working with neural networking architectures (also called “deep learning” by marketing people). In fact, even when such architectures aren’t used, being able to rely on heuristics and a static ruleset makes all AI programmers’ lives easier; this is why CPP was able to get its decent AI through 4+ years’ of work from 2+N programmers (2 main as Gazebo and Ilteroi, N supplement/groundwork as ninakoru, notque, and myself to name a couple), since BNW had a well-established system for unit control. However, all of this goes out the window for a game in active development: fundamental rules can change on a weekly basis, possibly invalidating months’ work of neural net training and heuristic development. As such, I would estimate that even with good AI programmers, developing a good AI for a Civ game would require having at least two of said programmers work on the AI from the start of the development cycle. If we are to take Civ6 as an example, it has been in development roughly since the release of Civ5. That’s 6 years’ development x 2 programmers x 300k USD per year = 3.6M USD extra just spent on AI development… and that’s not counting the extra overhead and management costs for a dedicated AI development system. There are way better returns on a 3.6M USD investment for 2K than employing dedicated AI programmers."

1994

15 turns seems about right to me for all civs, assuming the city is around size 10 and has some kind of mine resources. As in previous civs, having some hills nearby is important because obviously it’s impossible to have any real industry without mines nearby. My capital usually takes closer to 10 turns for a district at that point.

1995

Yeah, that’s a nonsense post. Game “AI” has nothing to do with real AI. No one is ever going to use machine learning and neural nets to path units around a hexmap, or to manage civ diplomacy, or if they do they will fail because those techniques are not useful for the problem.

Apart from the local tweaked version of A*, conventional software techniques are all that’s required for game-board AI; it’s just you actually have to want to code a not-horrible machine player, and Firaxis never has.

1996

I chanced across a chart at the Octopus Overlords forum (4th post down by persself) showing how to optimize the placement of districts and wonders:

1997

I get that you don’t like the game. For me personally, I have more fun with Civ6 than Stellaris as well, despite its problems. But the former has a 85% Positive rating on Steam (both recent and overall) and a 78/80 score on Metacritic. I’m not trying to pull an argumentum ad populum, I’m just saying that while we grognards here on Qt3 find a lot of flaws in the game, lets not pretend that Stellaris was this disastrous release that they should have to do a year’s penance for or something. In any case, @Eric_Majkut was complaining about the lack of communication/visibility on Firaxis’ part, not comparing the games directly.

1998

So, I’ve been tinkering around with a mod (thanks to some really helpful examples online I could crib from), the intent of which is to slow down research/civics without the corresponding sluggishness of production that comes with playing on Marathon. It’s intended to allow you to play on Epic and give it a Marathon feel for research, but the production speeds of the former. It also tries to address silly stuff like running around with musket and cannon in 200BC by further scaling the costs of later tech groups.

The intent of the mod was to get the game to play the way I wanted, but if anyone’s interested feel free to PM me.

1999

[quote=“KevinC, post:1997, topic:78555, full:true”]I’m just saying that while we grognards here on Qt3 find a lot of flaws in the game, lets not pretend that Stellaris was this disastrous release that they should have to do a year’s penance for or something. In any case, @Eric_Majkut was complaining about the lack of communication/visibility on Firaxis’ part, not comparing the games directly.
[/quote]

Well, first off, I do think it was a pretty disastrous release (I’m sure we can now start quibbling over what constitutes “disastrous,” but I don’t think there’s much point to that) . Tom Chick fairly heavily lambasted it, and gave it 1 out of 5.

Second, my point is that “communication/visibility” does not mean much. I can “communicate” all day. If I don’t actually do or fix anything material, that communication is rather pointless (and in some ways, even more irritating, as it starts to look a lot like lipservice).

Communicating with me about how you’re going to fix something, not having fixed it within six months, and then putting out paid DLC is not really good in my opinion.

2000

He lambasted Civ5 as well, but everyone assures me the game was terrific. :) We just seem to have a very different opinion on the development work that has gone into Stellaris post release. And that’s cool, this place would be dreadfully dull if we all thought the same. I appreciate your take on the subject!

2001

I must be getting really unlucky* or I’m doing something terribly wrong then as only one or two of my cities playing other Civs have managed good build times.

*E.g. I’ve only ever found one relic from a goody hut. I get scouts though, which is kinda cool, they’re effective and keeping my area lit up and free from barbs.

2002

You may perhaps sometimes have to override the default citizen allocation. Like you may have a couple of 2-food/2-production spots, but the AI will prefer a 1-food/1-faith/3-gold spot instead because it has more total resources, despite the fact it leaves the city unable to grow or produce anything. But 10-12 turns for a district in the early-mid game doesn’t seem that much of a stretch to me. Of course sometimes I may deliberately place a city someplace with no production to get a particular resource in which case it will be screwed for a long time, but I’m talking about generally well-situated cities.

2003

I dunno, that sounds cheap to me if those advances are carried forward into future releases. It seems nuts to me that we’ve made incredible advances in natural language processing in the last five years, a notoriously tricky, nay impossible, problem, yet game AI has if anything regressed.

2004

I usually micromanage my cities (years of being a Civ player), but maybe I’m just sucking at city placement - definitely possible. Just to be sure, this on marathon/huge?

2005

Oh no, sorry about that. I was talking about default settings.

2006

Yeah, build times are slower in Civ VI, no doubt. I think they want production to be co-equal to technology, and I see the point of that. Technology without production capabilities shouldn’t get you too far.

As a result, you really want to get your industrial district up as soon as possible, choosing places where you can get good adjacency bonuses. (Germany with its early, cheap version is the easiest civ to play.) And set it up so those industrial districts are within 6 tiles of multiple cities, so come factory time, you get multiple bonuses on each city.

But even before that, mine those hills, and put saw mills on forests. Sawmills on forests on hills along rivers are incredible. And, if you’ve got to set up a city or two where there aren’t good tiles for production, set up an internal trade route or two from that city to other cities that can give production. (Build the trade routes elsewhere and transfer them to the low production city.) Then use the imported production to build an industrial district and workship.

But, in the end, Civ VI forces you to pay as much attention to production as you do to technology.