1-23 Hotfix (1/23)

[Tubbilicious]

Do human players actually buy resources like Iron and Horses for money? I can see buying something like Coal so I can build factories, but if I buy ten horses and build ten cavalry I'm in danger of the deal ending and having ten useless units.

If humans don't do it, the AI shouldn't try either.

I do. Control of the oceans can be very important, and that can require considerable amounts of iron.

 

[pandasnail]

Do human players actually buy resources like Iron and Horses for money? I can see buying something like Coal so I can build factories, but if I buy ten horses and build ten cavalry I'm in danger of the deal ending and having ten useless units.

If humans don't do it, the AI shouldn't try either.

I do. I very much like engaging in prewall early warfare. If I can't improve horses myself, I'll buy some horses, make some horsemen, and conquer a horse city. Same if I somehow don't have iron by trebs. I mean, I don't often care that much about going from 5 to 8, but I find going from 0 to 2 quite useful.

One other thing is I never steal workers if I'm not actually at war. Just feels unimmersive to me personally to enslave without consequence. So buying horses even when I have them lets me focus my limited worker turns on farms, city connections, and even getting my starting monopoly up instead. I mean, they aren't expensive usually.

 

[Enrico Swagolo]

I really mostly like the new building gold costs - it's much more fair now and actually worthwhile to get them. The production you get to buildings is probably better than on units but you only get parts of them.

I don't know if it's due to scaling by amount of cities and eras but the price becomes a bit high later on? It's a bit weird a Caravansary costs me 500 gold to get half of it, granary costed 90 but now is about 180 and harbour is 650 or 670. I look at later era buildings (industrial iirc) and they usually cost like 1500 each? I play on Epic btw (as always).

By the way, is it just me or is the AI less likely to expand much in this version? It's late medieval and I have like 9 cities (4 conquistador'd on some weird fishy islands which I probably wouldn't have conquistasettled if I didn't get God of the Sea), Selassie has 4, Pachacuti has 4 and Boudicca has 4. I play on Emperor.

 

[kaspergm]

Compare the yields on those tiles to the yields of your specialists, and you'll see that - in many cases – the specialists are more valuable (yields + GPP). If you want more production...put it on a production focus. Otherwise, the AI will try to balance all of its yields, which it seems to be doing just fine. Unlike vanilla civ, 'default' focus tries to get all city yields to be roughly equivalent (as much as possible) while staying food-positive.

Also, it's a big city, and your food surplus is quite nice even at that size, thus specialists are even more likely.

No 'sweetspot' was missed – every city is going to behave slightly different based on the formulas in the DLL.

G

Sorry, but I disagree. Even when I do use whowards dll to separate Great Merchants from the other Great people, those two citizens working as GM specialists are completely wasted instead of working those two gold tiles. The two specialists give me 4 Gold + 6 GM points, the two gold tiles will give me 4 Gold + 8 Production which is hands down better at default focus imo. I dread how people would feel about this if those GM points also would work against their GS progress once a GM spawns.

 

[kaspergm]

Just to throw a bit more light on the above, I did a bit of testing around and found some more weird issues. For instance, I switched to manual specialist assignment which gave me this - notice how a citizen works the Cotton southwest of town:

Spoiler :

Now what I did was turn to food focus and then back to default focus as in the first screenshot, but now the game chooses a different selection - notice how the Cotton tile is no longer worked, and instead a farm on the floodplains east of the town is worked:

Spoiler :

That seems just ... random?

 

[Gazebo]

Just to throw a bit more light on the above, I did a bit of testing around and found some more weird issues. For instance, I switched to manual specialist assignment which gave me this - notice how a citizen works the Cotton southwest of town:

Spoiler :

Now what I did was turn to food focus and then back to default focus as in the first screenshot, but now the game chooses a different selection - notice how the Cotton tile is no longer worked, and instead a farm on the floodplains east of the town is worked:

Spoiler :

That seems just ... random?

It isn't random, it's a factor of the iterator plot system – plots are iterated in a ring, if certain food requirements are hit by a certain number of plots, the model starts to look for other plots.

I'll look at reducing specialist values a bit, but keep in mind that the exact opposite complaint was levied about two months ago (that cities on default weren't using enough specialists). So don't expect much.

If you aren't happy with the default auto-AI system, I strongly recommend manual specialists and/or plot assignments. Most users adjust plot assignments manually anyways.

G

 

[Wario Mario]

Something I've noticed with trade deals:

Say an AI offers you 4 gpt for a luxury. I've found that they will always go for 5 gpt, or even 6 gpt. There's no reason to ever accept an AIs offer, they will always go for more. Anyone else noticed this?

That's really annoying.
It reminds me of Civ 3 where the AI had an exact amount it would accept for every trade - but you had to manually enter every value one at a timeuntil you found the magic spot where they'd say no (ex: will they pay 10 gold? ok, 11? ok, 12? ok, 13? ok, 14? No? 13). It would make a lot more sense if they'd just given the max value by default.

On the topic of Paper, Paper is a pretty hard to get resource that you need to keep city-states on your side, I definitely don't think they are undervaluing it.

The AI really loves that freaking paper. I've seen AI offer technology, other strategic resources in favorable ratios (I've been offered 11 iron for 3 paper) and the grand prize: DOW on my least favorite Civ for 4 paper. I often forgo using paper for its intended purpose and just use it in trading.

Wait a minute - does that mean paper works for gaining influence on major civs as well as minor? Well played, big G.

Now if only the AI wasn't psychic when it came to world maps. First to discover "the new world" on a terra map? Every civilization up to and including your arch enemies will be asking to buy that map, despite the fact that they shouldn't know what's on it.

 

[kaspergm]

I'll look at reducing specialist values a bit, but keep in mind that the exact opposite complaint was levied about two months ago (that cities on default weren't using enough specialists). So don't expect much.

G

Thanx for giving it a try at least. I was not aware of such an issue being put up - I was rather happy with the default AI myself in this area, but when that's the case, obviously you're in a tight spot to satisfy all of us. When that's said, however, it seems like it would be a good idea to treat Economist specialists different than the rest if possible, given that the flat 2 gold yield is downright bad compared to what you'll get from working a tradepost, and on top of that Great Merchants are under normal circumstances undesirable, making it a double-loss to have a citizen working such a slot - but of course, I don't know how much work that would be.

 

[Gazebo]

That's really annoying.
It reminds me of Civ 3 where the AI had an exact amount it would accept for every trade - but you had to manually enter every value one at a timeuntil you found the magic spot where they'd say no (ex: will they pay 10 gold? ok, 11? ok, 12? ok, 13? ok, 14? No? 13). It would make a lot more sense if they'd just given the max value by default.

I've addressed this a hundred times, but here goes again (not a dismissal, but a prelude to a rehashed explanation from before).

They give the best possible value to themselves, but the 'flexible % value' (necessary for most deals to fire properly) means that there is usually a 1-2 GPT amount of 'wiggle room.' Removing that % means that the AI will only submit deals that are absolutely 100% even, which rarely happens. Considering the number of nascent bugs and/or incredibly dumb reasons the AI evaluated deals before (remember, the AI didn't know how to properly evaluate or offer third party war/peace, defensive pacts, strategic resources, luxuries, or open borders deals prior to the CP), a 1-2GPT variability in value is an acceptable consequence.

G

 

[Gazebo]

Thanx for giving it a try at least. I was not aware of such an issue being put up - I was rather happy with the default AI myself in this area, but when that's the case, obviously you're in a tight spot to satisfy all of us. When that's said, however, it seems like it would be a good idea to treat Economist specialists different than the rest if possible, given that the flat 2 gold yield is downright bad compared to what you'll get from working a tradepost, and on top of that Great Merchants are under normal circumstances undesirable make it a double-loss to have a citizen working such a slot - but of course, I don't know how much work that would be.

I've addressed this concern as well (not saying this as a dismissal, but to note that the current model isn't a bug, but rather intended).

In short, the AI did not consider the value of tile yields outside of the three 'basic' yields (Food, Gold, Production), and also science for specialists. Culture, faith, tourism, and the actual outcome of GPP from specialists (i.e. the fact that GPP merchant points lead to a great merchant) were not really considered by the AI. City AI just didn't really care about any of that, and was - essentially - surprised when a great person spawned as a result of specialists. In fact, specializations for 'low yields for a city' only really cared about food. I rewrote all of this so that all yields are evaluated by the AI in terms of deficiency/proficiency in a city, and GPP are considered based on strategies. This is why the AI uses all of the different city specializations now (instead of just default/production/gold, as before).

The consequence of this is that allowing the AI to manage a city wholly means that it will try to balance yields as well as GPP output based on the city's deficiencies/proficiencies. This will make default-AI cities behave differently from vanilla civ, but it is in no way buggy. Just tuned differently.

G

 

[kaspergm]

I understand that and I am, honestly, in big admiration of the job you've done. I'd even say your approach makes perfect sense if not for the fact the developers botched up the whole great person system, and most specifically Great Merchants, by linking the Scientist, Engineer and Merchant counters. As you are probably well aware, this means that under normal (non-Venice) circumstances earning GM-points is actually a negative thing and something you want to avoid if you want to optimize your game.

Now I don't know if the game separates the different GP-points or just counts any GP-point as a GP-point, but if it does separate them, I feel you should value GM-points lower than the rest simply to accommodate for this botched GP-system. This should mean that GM slots (by and large) only are used if city is on gold focus or if you have GP-focus on.

 

[Atlas627]

I understand that and I am, honestly, in big admiration of the job you've done. I'd even say your approach makes perfect sense if not for the fact the developers botched up the whole great person system, and most specifically Great Merchants, by linking the Scientist, Engineer and Merchant counters. As you are probably well aware, this means that under normal (non-Venice) circumstances earning GM-points is actually a negative thing and something you want to avoid if you want to optimize your game.

Now I don't know if the game separates the different GP-points or just counts any GP-point as a GP-point, but if it does separate them, I feel you should value GM-points lower than the rest simply to accommodate for this botched GP-system. This should mean that GM slots (by and large) only are used if city is on gold focus or if you have GP-focus on.

I'm pretty sure the CBP has unlinked the counters. They seem to sort by GP type. Hence, as long as you will at some point get the GP (and the cost doesn't keep scaling faster than a city produces points), then specialists are always useful.

 

[Gamewizard]

The GP counters are separated now, so earning a GM does not affect the creation of a GS or GE. Furthermore, all specialists start with giving 3 yields and they all increase with tech.

Great Merchants are extremely useful now, since it allows you to tackle poverty in low income cities.

 

[Gazebo]

The GP counters are separated now, so earning a GM does not affect the creation of a GS or GE. Furthermore, all specialists start with giving 3 yields and they all increase with tech.

Great Merchants are extremely useful now, since it allows you to tackle poverty in low income cities.

I think kaspergm plays with the CP only (he hasn't been converted to the one true faith yet), so his complaint is valid in that regard.

G

 

[yanner39]

Hey all, I apologize in advance - I love the game and love the CBP, but I'm not too good with understanding all the components.

Anyways, I've downloaded the new fix, installed it and cleared my cache. I have the newest EUI. When I load up a game, I cannot click on my units, settler or warrior. So I can't settle a city. I'm not sure what I'm doing wrong.

Thanks.

 

[Funak]

Compare the yields on those tiles to the yields of your specialists, and you'll see that - in many cases – the specialists are more valuable (yields + GPP). If you want more production...put it on a production focus. Otherwise, the AI will try to balance all of its yields, which it seems to be doing just fine. Unlike vanilla civ, 'default' focus tries to get all city yields to be roughly equivalent (as much as possible) while staying food-positive.

Also, it's a big city, and your food surplus is quite nice even at that size, thus specialists are even more likely.

No 'sweetspot' was missed – every city is going to behave slightly different based on the formulas in the DLL.

G

I also think the cities place way too big of a focus on specialists over growth. If you settle a city with a pioneer for example it is going to work both the scientist slot and the engineer slot, even if you set the governor to food-focus.

In fact, that takes me back to another problem, specialists should probably not overrule the governor in the way that they do, a specialists is at the end of the day a food-drain and should pretty much never be worked over a 3 food tile in a city that is food-focused. This issue is currently very visible in most cities and extremely visible if you've picked up any of the half food cost on specialist policies. A capital on food-focus with Majesty is going to work all specialist slots available, usually decimating its growth-potential.

 

[Funak]

I do. Control of the oceans can be very important, and that can require considerable amounts of iron.

I do. I very much like engaging in prewall early warfare. If I can't improve horses myself, I'll buy some horses, make some horsemen, and conquer a horse city. Same if I somehow don't have iron by trebs. I mean, I don't often care that much about going from 5 to 8, but I find going from 0 to 2 quite useful.

If I don't have strategic resources available in any of my cities I'm going to try and get a hold of them in any way possible, building a citadel, settling a crappy out of position city, or trading for it and then using the units produced from it to attack the closest city with strategic resources available.
Main reason I don't want to rely on the trade is because if the AI trading with me decides to stop trading or DoW me, I'm going to be in a big pile of trouble.

I really mostly like the new building gold costs - it's much more fair now and actually worthwhile to get them. The production you get to buildings is probably better than on units but you only get parts of them.

I don't know if it's due to scaling by amount of cities and eras but the price becomes a bit high later on? It's a bit weird a Caravansary costs me 500 gold to get half of it, granary costed 90 but now is about 180 and harbour is 650 or 670. I look at later era buildings (industrial iirc) and they usually cost like 1500 each? I play on Epic btw (as always).

I'm still undecided on this actually. I liked the super-low price for rushing buildings in my capital, actually used it as well, got gold from a ruin and invested in both a shrine and a monument (first time ever? ). It did however increase up to normal levels, and probably higher than normal levels as soon as I settled a few more cities. So mostly undecided, and probably a huge boost for Carthage.

By the way, is it just me or is the AI less likely to expand much in this version? It's late medieval and I have like 9 cities (4 conquistador'd on some weird fishy islands which I probably wouldn't have conquistasettled if I didn't get God of the Sea), Selassie has 4, Pachacuti has 4 and Boudicca has 4. I play on Emperor.

Yeah some AI seems to not really want to expand early on. I've run into a few who did however, Kamehameha for example settled 10 cities on my last map (small Pangaea). On the other hand Austria only settled their second city way later on, right next to my 7th city. Rome settled 6 cities, the ottomans 3 and Boudicca 3. The ottomans eventually settled 2 more cities in the industrial era.

 

[Enrico Swagolo]

I'm still undecided on this actually. I liked the super-low price for rushing buildings in my capital, actually used it as well, got gold from a ruin and invested in both a shrine and a monument (first time ever? ). It did however increase up to normal levels, and probably higher than normal levels as soon as I settled a few more cities. So mostly undecided, and probably a huge boost for Carthage.

Yeah I agree, I feel like the price should scale less harshly. Especially on late era buildings whose prices become really irrational. That's why I said I was mostly fine with it - at first it's great and incredibly cost effective, but then it becomes worse than it was. I feel bigger maps would be more disadvantaged from the system due to higher amount of cities you're expected to have to keep up (unless it's already map size scaled as well).

In my opinion the scaling should end at around the prices of previous patch (and you'd need more cities to get there than now), at which point getting another city no longer would affect costs. Basically previous patch prices = upper boundary reached at X cities depending on the map would in my opinion be the best solution

Yeah some AI seems to not really want to expand early on. I've run into a few who did however, Kamehameha for example settled 10 cities on my last map (small Pangaea). On the other hand Austria only settled their second city way later on, right next to my 7th city. Rome settled 6 cities, the ottomans 3 and Boudicca 3. The ottomans eventually settled 2 more cities in the industrial era.

Aye the biggest example of that I've encountered is Casimir. From an expansive guy who always became THE runaway in BNW now he can't into expansion, always sits at 1 city, spams wonders and then gets another city in medieval/ultra late classical

 

[kaspergm]

I'm pretty sure the CBP has unlinked the counters. They seem to sort by GP type. Hence, as long as you will at some point get the GP (and the cost doesn't keep scaling faster than a city produces points), then specialists are always useful.

I think kaspergm plays with the CP only (he hasn't been converted to the one true faith yet), so his complaint is valid in that regard.

G

Yeah, I play with my own balance patch but manually decouple the GP in the Custom Options of the CP, so it's not that GM are a negative for me under normal circumstances - but even at that, I do feel those gold tiles were the better choice for working.

 

[Gazebo]

Good news - the discussion prompted me to give it a look, and I found a rather...odd bug from firaxis related to plot and specialist scoring. In essence, yields, happiness, food, and GPP were interacting in such a way that cities that were unhappy and poor were less and less likely to make good decisions. Perhaps a mirror of real life, this is not functionally very fun. Fixed, and it seems to have mollified the issue somewhat.

G