New Version - October 9th (10/9)

[pineappledan]

With the Tabya now being +1 prod on every river, doesn't Songhai look way too close to the Iroquois? It's just rivers instead of woods.

+1 prod on every (river/forest) tile, units have double move in (rivers/forests), and (rivers/forests) can act as city connections. They even get their UBs at the same tech level

Edit: also combat/mobility-relevant promotion on all units for (water/forest)

 

[ElliotS]

Late to the party, but very appreciative of the CIV balance changes. The only one that surprised me was Arabia, because most reports I've read is that they have a very solid CV game right now, but I guess it was kind of curious when Markets got buffed, Bazaar lost some appeal.

Played a game as Songhai recently, and with the gold from the early UU / easy culture / early UB - it definitely seemed like godmode. I generally play King, but have been playing Emperor a bit lately and I was really surprised at how they snowballed. I was especially left impressed with the mobility of their armies, which is something I had discounted many times before when considering choosing them. I think the change to production is good nerf, that is balanced by the synergy with the UB.

I played a game with Japan at Emperor for comparison sake, and really struggled. As luck would have it, spawned next to Songhai - it took well into Medieval to even attempt to fight them, and had to wait for them to declare on someone else first. I'll let others weigh in here, but Japan came off feeling rather weak. Perhaps I should have bee-lined Dojos, but I'm addicted to Oracle. (perhaps a personal problem)

Ethiopia getting a buff, albeit a rather small one is not optimal - the extra techs they get always seem to place them in the top tier. I've also started running some AI v AI games recently and I've seen the same results there. They never struggle.

I love the bump for the Ottomans, TRs finishing requires such a delicate balance with diplomacy and losing a trade route to barbs usually ended up with a ragequit on my part. I'd like to play a game with them soon and update on whether the 50% is overboard. I'll also throw them in an AI v AI game and see what happens.

Also, really excited for the Spain buff, I think they've got a fun design, and the naval buff should push them into the top tier. I play multiplayer a fair bit, and this is a favorite dark horse pick.

Sweden/Venice/Russia/Persia I can't really comment on, although I've been considering a Venice singleplayer game recently, so maybe now is a good time. Multiplayer game is just too passive with them, you end up twiddling thumbs most of the game while other people are playing. Can anyone make the case for why Sweden/Russia/Persia are fun to play, I've never been able to convince myself to try them?

Japan is actually really good. If anything they're closer to OP right now. The combination of getting Faith and culture early, and tons of culture and science with Dojo is insane. Those are the best yields, and japan gets them at great timing. Add in the host of other goodies and they're just really awesome.

 

[Gazebo]

With the Tabya now being +1 prod on every river, doesn't Songhai look way too close to the Iroquois? It's just rivers instead of woods.

+1 prod on every (river/forest) tile, units have double move in (rivers/forests), and (rivers/forests) can act as city connections. They even get their UBs at the same tech level

They play completely differently, so no.

G

 

[black213]

I don't know of any other situations that would be comparable. What nullifies statecraft?

Nevermind silly me, I don't think there are enough players playing without city-states or other similar disableable(?) major features to warrant conditionally reworking all respective bonuses/effects, whereas barbarians are often disabled for multiplayer, so 1 culture per city from Authority with disabled barbs makes perfect sense to implement.

 

[pineappledan]

They play completely differently, so no.

The fact that forests are destructible features and rivers are not is a major one, but on paper it looks kinda funny. They're both expansionist/production civs now with ancient era UBs. Was there ever talk about moving the tabya to replace workshop instead of stoneworks? That could preserve the culture focus, keep them more unique in their competency while solving the issue with them powering through early policies, and would fit more with their medieval historic power-spike.

 

[Ziad]

I would say while both are expansionist, Iroquois is more defensive while Songhai is more offensive.

 

[pineappledan]

I would say while both are expansionist, Iroquois is more defensive while Songhai is more offensive.

Oh for sure. The fact that their bonus is much stronger within their own borders makes Iroquois a more defensive civ, focused on thick play style (progress, industry), while Songhai was usually played hard domination (authority, imperialism). This patch, however makes me wonder if Songhai aren't equally capable, or even better, at Iroquois' thick infrastructure play? If they have the requisite quarry and rivers, of course

 

[Moi Magnus]

Thing is ability to choose what to construct can't compensate it. Too many buildings are lost when city gets conqured. And even if buildings stay - here is a simple maths: if you annex a city - you get 40% bonus science and culture from that city, but it reduces science and culture of all of your cities. It means that starting from a starting with a certain number of cities puppeting will always be better, the exact number of cities depends on a map size

If I remember correctly, the science and culture malus is linear, not exponential.
If that's the case, your calculations are wrong:

We suppose for the example that you discover a new tech every 10 turns. Founding/annexing a new city increase the cost of techs by 'X'. If founding/annexing the new city increase your research output by 'X/10', then you still discover a tech every 10 turns. If the city produce more, it will increase your research rate, and if less, it will decrease it.
Saying that cost are linear means that 'X' is the same whatever your number of cities, so that "is this city worth annexing" is independent of your number of cities. A city good enough is always good enough.
However, what is true is 'X' increase with time, so it is more and more difficult trough time to have the city pay for itself. So past a certain point of the game, unless the city you acquire has full-infrastructure, you no longer want any new city on your empire for science victories.

 

[tu_79]

Can anyone make the case for why Sweden/Russia/Persia are fun to play, I've never been able to convince myself to try them?

Sweden is really fun if you like to conquer. You can go since the very beginning with some spearsman dealing more damage. Add some fast catapults and you are ready to expand aggressively in Classical. But unless other civs that spike only in that era, Sweden gets better and better. Have your Great General and your melee units, swordsmen for example, fight at +45% CS. That's huge. You can take cities barehands with that. Oh, you need iron, by the way. Now, when most civs are exhausted by their conquests, Sweden receives the Carolean, amazing unit when attached to a GG, giving +60%CS, and the March promotion (heal every turn). This unit does extra damage to enemies adjacent to your recent kills. You probably won't be finishing many enemies with your Caroleans (that's cavalry work), but if you take Authority (melee heals on kills), there is synergy. I'd say Sweden is unmatched in land combat. At least that was two versions ago. Now it seems that some combat bonus has been replaced by an XP bonus, and maybe March is gone.

Persia is ... interesting. It's a golden age warmonger civilization. The Immortals comes pretty soon and they defend pretty well, but lacks a bit punch when fighting, best to be used as tanks, backed by ranged and siege. You can do some early conquests with that. Then comes the Satrap building and the fun begins. Golden Ages come faster and stay longer, something you can abuse with your religion. And most importantly, extra movement. Think that +1 movement is little? Think twice. Your melee, ranged and siege units can move 2 tiles per turn in rough terrain, while most other civs move 1 tile per turn. That's doubling your enemy's. Given you have room to maneuver, you can outsmart any army. Only drawback is that you need to time your fighting. Now, Persia's been given some reason to focus on gold.

I don't have experience with Russia lately.

 

[randomnub]

March is still there for caroleans.

 

[matlajs]

By the way, am I the only one who always goes Industry in renaissance?

I am starting to think, that industry is a ,,must have'' for going wide. It will yield much more science than any other tree via ,,bank'' purchases. sometimes, when i swimm in a gold, i just purchase a most expensive stuff buildable for getting a science from it. But also it is probably same valuable for unit training, since workshop/windmill/factory give +5% flat production toward everything and gold to adition.

 

[Owlbebach]

Well no it doesn't reduce your city outputs. It just increases the amounts you would need from all cities to compensate.

Don't you think that it is the same thing mathematically? Question is how much turn it takes to research a tech, exact number of science per turn and cost of tech does not matter.

If I remember correctly, the science and culture malus is linear, not exponential.
If that's the case, your calculations are wrong:

The question is what % leads to a net benefit per city

I could see lowering the culture/science scaler by about 2-3% at each level

See some maths below. Long story short: currently expansionist civs are screwed. Dropping increase of costs per city by 3% might be a good thing.

@CrazyG Summoning maths teacher, check my mistakes, its been couple of years since i done this last time.

Okay i have some free time and i made some calculations.

Assumptions first. Lets assume your cities are equal, and increase in cost of tech/policy for every new non-puppet city is 10% (large map).

N is a number of your non-puppet cities. M is a number of your puppet cities. Here is a function that calculates research time of a tech depending on N and M: https://www.wolframalpha.com/input/?i=plot+y+=++(n+m*0.6)/(1+0.1*n)+from+0+to+20

Function looks like this:

Here is a plot:

Lines on the surface are lines of equal value of a function.

From the image you can see that if you have 20 cities - it is better to have 1 capital and 19 puppets. It does not seem right, don't it? But thats a very extreme case.

Now lets solve an inequality to find an optimal number: https://www.wolframalpha.com/input/?i={+d(n+m*0.6)/(1+0.1*n)/dn+>+d(n+m*0.6)/(1+0.1*n)/dm}

For those who do not know what derivative is - this inequality defines values of N and M for which non-puppet cities are better than puppet cities.

Here is the answer:

It means that the optimal number of non-puppet cities for large map is 6. Any non-puppet city after 6th will make more bad than good.

In reality your capital is always better that other cities. Cities settled earlier are usually better that cities settled later. Also Trade Routes and Guilds and some other stuff does not scale with number of cities, so real result will be shifted towards lower number of cities even stronger. That's it, guys.

 

[Workerspam]

Assumptions first. Lets assume your cities are equal, and increase in cost of tech/policy for every new non-puppet city is 10% (large map).

Is this a valid assumption? Will a puppet city provide as much science as one under your direct control where you can build/rush science buildings?

If not that changes the equation, potentially significantly depending on the disparity in science output between puppets and controlled cities.

 

[Enrico Swagolo]

Spain's pretty cool now. It has an actual strong start thanks to the food/faith proccing for capital, and the naval fleet for Faith makes every conquest rewarding even if the Faith costs for boats seem super high. Not a problem with Hero Worship and/or Veneration. Together with Zealotry UA made me happily spend Production for buildings/wonders only come medieval. Mission still sucks as you can't buy with Faith for puppets (meaning the majority of your cities), Conquistador's settling is situational unless you modify worldsizes file to reduce the penalty like I did so Cataphract's better in comparison. It's not OP but way better than it was.

However, internal routes seem useless now that the alternative can easily provide more Culture alone than they provide Food/Production, and that's not even taking into consideration all the gold/science they have. I think the trade routes should go back to what it was before this patch, it's way too many yields as it is. It wasn't underpowered or anything, it was still significant, just not overwhelming.
I also agree with guys above about puppeting/annexation. After all the improvements to how puppets choose their buildings, not even having a choice gives much as they're somewhat competent anyway, and what it does is not worth wasting so much Science/Culture over. Not even Faith bought buildings you can't get without annexation can really make it worth it. Puppets aren't even OP, it's just that the alternative is bad. I feel like annexation should be beneficial if you can handle the unhappiness before a courthouse is up, otherwise you waste several turns of production getting a Courthouse up only to hurt your Science/Culture anyway. I've changed the culture/science penalty per city to 5 (on Standard map size) in balance changes/worlds/worldsizes.sql while increasing Epic tech costs from 200% to 210/220% depending on my mood and all's been more balanced since.

 

[Owlbebach]

Is this a valid assumption? Will a puppet city provide as much science as one under your direct control where you can build/rush science buildings?

If not that changes the equation, potentially significantly depending on the disparity in science output between puppets and controlled cities.

It is good point. It is something that pushes the exact answer towards more cities. But as i said - keep in mind that later cities are usually way worse that earlier cities, also when you capture a city - you have big population and some buildings there. Also there are other things that push the answer towards less cities that are also not taken into accounts, like bonuses from Founder, Enhancer, Guilds, Trade Routes, City States, National Wonders etc.

One more GIANT thing that is not taken into accounts here is happiness

Overall I think the assumption is quite okay

 

[ElliotS]

Lots of maths.

Forgive me if I'm wrong, but isn't this missing the point for the most part? The cost increases are additive, so at some % adding new cities that are decent will literally never be a bad thing. For example if the increase was 2% every additional city only needs to produce 2% of the base cost for each tech/policy to be profitable/cost neutral. even if you have a bunch of sources that don't scale w/ number of cities, as long as the new cities cover their costs they're not a burden.

The goal should be to set the % to just under or at the expected increase from satellite cities depending on your balance wants.

Also is there a reason to scale % with map-size? At surface level it makes sense, but thinking about it there's a double-whammy in both more space to expand and reduced cost increases that benefits wide civs. If the standard equation is right, then it should be the same on all map sizes. Right?

Paging @CrazyG for math fight resolution.

 

[randomnub]

*checks his current game*





*counts 6 non-puppet cities*

 

[Owlbebach]

Forgive me if I'm wrong, but isn't this missing the point for the most part? The cost increases are additive, so at some % adding new cities that are decent will literally never be a bad thing. For example if the increase was 2% every additional city only needs to produce 2% of the base cost for each tech/policy to be profitable/cost neutral.

It will, in the long run. But question is whether it is better to add a puppet or a non-puppet city? And the answer is "adding the same city as a puppet is better".

EDIT: take a look at the plot, you can check different point by choosing N and M. Also you can play with the function with wolframalpha. For example: change 0.1*n to 0.07*n and you will get the value for 7% cost increase instead of 10%

 

[Y.O.]

With the Tabya now being +1 prod on every river, doesn't Songhai look way too close to the Iroquois? It's just rivers instead of woods.

+1 prod on every (river/forest) tile, units have double move in (rivers/forests), and (rivers/forests) can act as city connections. They even get their UBs at the same tech level

Edit: also combat/mobility-relevant promotion on all units for (water/forest)

IMO Songhai is all about expansion through force and the gold and production to support it. Tabaya also helps out satellite cities build infrastructure, which can be an issue when going wide through war. It's a different type of warmonger than say the Zulu, which just carpets the map, and the new and improved Sweden, which has very potent front line military abilities and units. I've only played as the Iroquois once, but I didn't play them the same way I did Songhai. Maybe one could tho.

EDIT: Also I always go industry with Songhai, unless my military is completely eclipsed by another Civ, then Imperialism. Don't think I ever tried rationalism. Industry always seem to fit more, and with the changes, the synergy is even better.

 

[kawyua]

I think pioneers with organization policy alleviates the culture costs for settling a new city quite well, but now how do I adjust for that...