New Beta Version - January 20th (1/20)

[Gazebo]

Hey all,

New beta inbound. Same rules as always - don't use if you don't want to test, and report back findings here.

Auto Installer: Download and install, and you should be good to go.
Manual Install: Remember to delete LUA folders (CP/CBP, if using full set) and clear cache.

Bugs/Enhancements:

• Moved as much savegame data as possible to the m_syncarchive (out of the raw kstream). Duplicates and poor attempts at 'savegame preservation' were leading to savegame bloat and large packet sizes for MP games. Reduces savegame size quite a bit and should help a lot with MP desyncs.
• Cleaned out unneeded EUI duplicates and fixed some modpack creation bugs in C4DF.
• Fixes for pathfinding warping and AI road-building logic
• A few other pathing and operational bugfixes from ilteroi
• Fixed most issues on github and those reported in bug reports thread
• Incorporated sound selection fix from 8h42
• Rewrote building/unit/process/project selection methodology to focus more on the object being made (instead of just AI flavor modeling)
• Worked on grandstrategy and policy AI to take into account buildings, policies, religions, and other elements of the empire as factors in choice weighting
• Note: yes, up to this point 99.9% of AI choices (i.e. units to build, wonders to get, etc.) were based solely on weighted flavors and a roll of the dice. Not anymore!
• Puppet AI overhauled a bit, should be a bit more rational, build more things, and will also switch to processes (usually wealth) if you are in negative GPT or have nothing else to build
• AI changes extensive, and still being tweaked/tested/balanced. Need feedback!

Balance:

• Arabian Nerfhammer, ho! Arabia UA now +1/+1 (was +3/+3) and 20% (was 25%)
• Siamese Wat now grants +4 Science to Temples/Shrines in the City
• English Steam Mill now +1 production per 3 citizens (was per 2)
• Reduced Krepost Maintenance to 1 (was 4), gains +3 Production base
• American MoMA now The Smithsonian - now grants Museums +5 Science (in addition to +5 Culture) and gains +5 Science if themed (instead of +5 Gold) - art a modified file from Janboruta

Not savegame compatible, at all. ALL files updated (1 thru 6a/6b and EUI), so grab them all. Auto Installer made for this version (as it is pretty stable).

Auto Installer (1/20b Version)

Or

Beta Folder

Things to watch for: odd AI unit/building/policy/GS choices and/or asynchronous behavior (i.e. clearly going science victory but spamming diplo units, etc.).

 

[JFD]

Worked on grandstrategy and policy AI to take into account buildings, policies, religions, and other elements of the empire as factors in choice weighting
Note: yes, up to this point 99.9% of AI choices (i.e. units to build, wonders to get, etc.) were based solely on weighted flavors and a roll of the dice. Not anymore!

Hm, this might cause problems for how I've implemented counter-Crime buildings in CID, which use general flavours to make buildings such as Jails, Constabularies, and Police Stations more favourable, as they each reduce Crime (which can affect all yields). Similarly, Loyalty (from Barracks, Armouries, and Military Acadmies) use the Happiness flavour (a bit roundabout, but Loyalty is generally only applicable to expansionist civs, so I think the best I have). Could you elaborate on how exactly the AI evaluates what it should build now?

Working on the Puppet AI is fantastic news, however

 

[Empariz]

Thanks for the work. I'm looking forward to try out MP with this version!

 

[Gazebo]

Hm, this might cause problems for how I've implemented counter-Crime buildings in CID, which use general flavours to make buildings such as Jails, Constabularies, and Police Stations more favourable, as they each reduce Crime (which can affect all yields). Similarly, Loyalty (from Barracks, Armouries, and Military Acadmies) use the Happiness flavour (a bit roundabout, but Loyalty is generally only applicable to expansionist civs, so I think the best I have). Could you elaborate on how exactly the AI evaluates what it should build now?

Working on the Puppet AI is fantastic news, however

Flavors all remain, but they are now the base upon which other elements are defined. Essentially, I've taken around 75% of the XML table elements (i.e. YieldPerPop(eYield) <FreeVotes>1</FreeVotes>, etc.) and used them to quantify additional weighted values on top of the flavors. So if two buildings have the same flavor, but one has higher base yields than the other, the one with higher base yields will have a higher priority.

So this is completely compatible with everything, and (depending on what functions you use) will actually make the AI understand what buildings and units 'do' a bit more than the basic flavors allow for.

G

 

[Funak]

Some initial thoughts on the Balance-changes (the only thing I really have time to look at right now).

Balance:
Arabian Nerfhammer, ho! Arabia UA now +1/+1 (was +3/+3) and 20% (was 25%)

I can understand that they were overperforming (having them as enemies I really had not noticed it, but whatever) but this isn't a nerf, this is butcher. +1/+1 just looks extremely boring, especially so because both the yields are yields that fall off like crazy in the midgame. If numbers like this is what it takes for the UA to stay balanced it is probably better to look for another solution.

Siamese Wat now grants +4 Science to Temples/Shrines in the City.

Fine I guess.

English Steam Mill now +1 production per 3 citizens (was per 2)

I really don't see the reason for this nerf. I don't think this building was a lot more powerful than the other lategame unique building, and both the ones of comparable tech got buffed as well. Even if the building is stronger than buildings on comparable techs England doesn't really have anything else going for them outside of naval warfare so it would hardly be a problem.

Reduced Krepost Maintenance to 1 (was 4), gains +3 Production base

Did you change the name back to Krepost?

American MoMA now The Smithsonian - now grants Museums +5 Science (in addition to +5 Culture) and gains +5 Science if themed (instead of +5 Gold) - art a modified file from Janboruta

A pretty big buff, truth be told, I really don't feel like Americas lategame was having much of a problem. That being said this change is probably fine, but I really don't see why you would hammer one lategame building and buff another one while they weren't exactly that far apart to start with.

 

[Kastanj]

The nerf on Arabia prevents the massive snowballing in the earliest eras and encourages building up towards an excellent capital in the late game, but may be a tad excessive seeing as investing in GP generation can leave you vulnerable.

Maybe a 1 or 2% to all GPP/turn per historic event (with a cap of +XX%) would make Arabia even more of a slow buildup towards a powerful late-game contender? I don't have any spreadsheets with yields over time prepared so I am merely spitballing.

 

[JFD]

Flavors all remain, but they are now the base upon which other elements are defined. Essentially, I've taken around 75% of the XML table elements (i.e. YieldPerPop(eYield) <FreeVotes>1</FreeVotes>, etc.) and used them to quantify additional weighted values on top of the flavors. So if two buildings have the same flavor, but one has higher base yields than the other, the one with higher base yields will have a higher priority.

So this is completely compatible with everything, and (depending on what functions you use) will actually make the AI understand what buildings and units 'do' a bit more than the basic flavors allow for.

G

Do you mean the AI factors specific yields? Or, in non-DLL-understanding terms, looks at when the buildings produces any yield (or yield modifier?) and uses that to determine that the building is superior. This is helpful for Loyalty, if so, which adds to YieldModifiers but obviously the Loyalty yield itself would be unrecognized, so should if I understand correctly make the XP buildings a bit more valueable (as they produce Loyalty mods.)

But what if a building has a negative YieldModifier but which actually has a positive effect, which is the case with Crime and the Espionage buildings? I'd wager the AI would ignore this or, at worst, consider this a bad thing, in which case perhaps we can discuss a new flavour that helps the AI understand Crime a bit better (i.e. if is Crime high, it should pursue buildings with Crime flavour, with priority determined by the leader's value for that flavour). If this is possible (I don't know at all how the AI looks at flavours, so I don't know if adding new ones is easy, so forgive me if I'm considering too much), I can open an issue later.

Somewhat related, does the AI factor flavours with a negative value? I know Health & Plague uses these to steer the AI away from Disease ridden Buildings which, flavour-wise, impact negatively upon Growth. I would probably do a similar thing with my Health & Disease, provided it would have an affect on the AI's decision-making.

I don't mean to detract from your work on the new release, of course! Always good to see a new release I'm just trying to understand the implications upon my somewhat hacky implementations

 

[zeta]

Arabia nerf seems like a very rough one.

 

[Zanteogo]

Thanks for the continued work!

 

[hokath]

Reduces savegame size quite a bit and should help a lot with MP desyncs

Will report back as soon as tested. This makes me feel all warm and fuzzy inside. Thank you so much!

Rewrote building/unit/process/project selection methodology to focus more on the object being made (instead of just AI flavor modeling)

He can't be stopped! I've noticed the AI seems "smarter" recently, if I went back to vanilla I'd probably have a hard time believing the difference.

American MoMA now

awwww yessssss... will it be Ranch replacing Agribusiness? Saloon replacing Constabulary? The I just saw the word Janboruta, omgomgomg

The Smithsonian

damn.
I know you don't think a Hermitage makes sense for America, but it doesn't really make sense for the Aztec either, if you get my point.

 

[Bombad]

+1/+1 just looks extremely boring, especially so because both the yields are yields that fall off like crazy in the midgame.

It's better than it looks. Especially if you only look at the culture/science, which is only one part of the UA. You also get a huge amount of great people and +60% tourism* from historical events. Every fourth historical event gets you a great person, which definitely feels impactful. The UA actually gains traction later in the game when you get buildings and policies that increase great people production. I highly recommend that you try it for yourself.

*+20% per event, and events trigger 33% more often because of the great people they sometimes provide.

 

[hokath]

+60% tourism*

I thought no way, but it checks out

events trigger 33% more often because of the great people

Except isn't it 20% of the points now?

 

[Enrico Swagolo]

Arabia definitely won't be underpowered with the changes, you guys are forgetting how many free GP's it can generate on a good Tradition capital, especially if some nice city states also give you some heroic events. It's still reasonably strong, just not overpowering.

 

[Funak]

It's better than it looks.

Arabia definitely won't be underpowered with the changes, you guys are forgetting how many free GP's it can generate on a good Tradition capital, especially if some nice city states also give you some heroic events. It's still reasonably strong, just not overpowering.

Here's the issue, underpowered or not, the issue is that it feels poinless. The +1/+1 is never going to be impactful, the number might as well be 0 as it is never going to change anything. And after removing that number the UA just looks boring and empty, stronger HE is a thing, sure, but the bonus is pretty much only tourism-related at that point, and feels really weird.

 

[Bombad]

Except isn't it 20% of the points now?

Oh you're right, I thought the +25% -> +20% chance was the bonus tourism from historical events but it's indeed the random GPP progress.

My bad, it's actually +50% tourism from events now (25% more events from the bonus GPP progress and +20% per event)

Arabia definitely won't be underpowered with the changes, you guys are forgetting how many free GP's it can generate on a good Tradition capital, especially if some nice city states also give you some heroic events. It's still reasonably strong, just not overpowering.

I agree.

In my last game I got 80 historical events in 230 turns (which is where the game ended because I accidentaly won a cultural victory), which means that the UA got me 20 free great people. That's huge !

 

[Lynnes]

Thanks for the work. I'm looking forward to try out MP with this version!

I'm currently creating the modpack, so it will be available for download very soon!

 

[Enrico Swagolo]

Here's the issue, underpowered or not, the issue is that it feels poinless. The +1/+1 is never going to be impactful, the number might as well be 0 as it is never going to change anything. And after removing that number the UA just looks boring and empty, stronger HE is a thing, sure, but the bonus is pretty much only tourism-related at that point, and feels really weird.

Meh, with Tradition you're gonna be getting 10/10 culture/science by medieval era for sure. An early HE will speed up your acquisition of stuff too, if a CS wants you to destroy an easy Archer camp then it's +1C/+1S for free basically. Considering all the free GPs you will get on top of it and that tourism it seems more than fine. The UA is still very strong and the bonus is far from pointless.

 

[Gazebo]

Here's the issue, underpowered or not, the issue is that it feels poinless. The +1/+1 is never going to be impactful, the number might as well be 0 as it is never going to change anything. And after removing that number the UA just looks boring and empty, stronger HE is a thing, sure, but the bonus is pretty much only tourism-related at that point, and feels really weird.

How is a UA that gets stronger and stronger all game, has unique flavor elements, and generates a crap ton of GPs from the active event system pointless?

The +1/+1 of the UA isn't the meat of it. The GPPs are the show. The yields are there to help you through the dark times when you forget that you'll be getting fields of GPs in the near future.

G

 

[Funak]

The +1/+1 of the UA isn't the meat of it. The GPPs are the show. The yields are there to help you through the dark times when you forget that you'll be getting fields of GPs in the near future.

Then replace it with something that actually feels interesting. Scrapping the science chaning it to 2 culture and going back to the tourism something like 1 culture 1 tourism per turn would make it feel interesting. Alternative 1 would be interesting because 2 culture feels like something 1 doesn't. Alternative 2 would be interesting because flat tourism per turn is pretty rare early on.
This is kinda the reason why I didn't like the science on it in the first place, it feels a bit spread thin, even more so with the lowered yields.

 

[Gazebo]

Then replace it with something that actually feels interesting. Scrapping the science chaning it to 2 culture and going back to the tourism something like 1 culture 1 tourism per turn would make it feel interesting. Alternative 1 would be interesting because 2 culture feels like something 1 doesn't. Alternative 2 would be interesting because flat tourism per turn is pretty rare early on.
This is kinda the reason why I didn't like the science on it in the first place, it feels a bit spread thin, even more so with the lowered yields.

I wasn't agreeing with you that the +1/+1 isn't interesting, I think it's perfectly fine.

G