Deity Difficulty

[CrazyG]

Yeah that's an important point. Conquering a single city that gets settled to close to you doesn't make a person a warmonger. Totally peaceful is a very harsh restriction, war is part of the game, and tradition is allowed to participate in it.

It can be achieved but I imagine its out of the scope of this project.

I'm giving it my best shot!
https://forums.civfanatics.com/threads/3-1-deity-peaceful-photojournal.655819/

 

[looorg]

I am slightly baffled here really. I share data gathered from the game, via the in-game tooltips. But that apparently isn't good enough cause it doesn't come in the proper formatted csv logfiles. Even if I enabled them now this late in my current game I'm somewhat sure it would be dismissed cause I didn't have logs from the beginning of the game. So I said I would run them in my next game. But from how it's received I'm not even sure I would bother at this stage.

None of the things I have said have been based on "feelz", unless you somehow think that the data shown in the game is from the feelz and is inaccurate. Or don't you trust me to read a tooltip in game, write it down and then post it here with some extra text to explain what happened. The data I did provide is not "feelz", it's data from the game in a format that just happens to not be the form of the preferred logfiles. It might not give the full picture but it should be enough of a picture.

I'm still waiting on the reason y'all can't just play immortal or emperor.

Would that somehow make Deity less broken? If it is broken and this just isn't the new intended way it's supposed to be -- in which case please do state that and I'll adapt to it. I was only asking and provided some specific feedback on somethings I thought was odd and was creating the outliers. If they say this is how it's supposed to be now. All is working as intended. Learn to like it. Then I'll do that. I'll adapt and overcome. I'm just less sure if it's winnable or for that matter fun as a game concept.

Where are your logfiles or notes? I don't see much else in the form of sharing here. Lots of people taking the piss out of other posters but doesn't appear to provide much in the way of information themselves.

When was the last time I won a game on deity? No problems winning game after game with the 2-9-4 version, it was harder then the months before but still doable. Then with 2-18-3 version I just had a lot of CTD around the T100 marks so no games with that version. Then all my games so far of the 3-1 version create these overpower outliers that are so far from the rest of the game it's quite odd. I don't really need to finish those games since I know that they are unwinnable after a certain stage. There is nothing that can be done ingame at that stage to close the gap that is created. Eventually I might catch up techwise but I can never close the culture-gap. My thinking now is that I need very specific starts and civs to make it even remotely possible to win or even keep up with the AI. But sure all games are not viable games, that is just the way it is.

If this is just down to AI improvement how come it only creates one, or sometimes two outliers, and how come the other AI:s doesn't seem to enjoy the bonus shower. My assumption based upon the in-game historical events notifications and tooltip data is that it's sort of a first come first serve type deal. The once that snowball first wins and the rest can just go and suck it. Go big or go home or something. When the AI enters a new eras, you get a notification in game. The following turns are now wonder turns. So the bonus is large enough to without issue pump out several wonders in short turn order. Which in turn gives you even more bonuses. From the events presented and the bonuses granted, based on the data in the wiki about difficulty levels that was linked here, I assume that certain AI:s are now more or less in a state of perpetual golden age. Am I wrong? So it's not an insignificant amount of yields given. Naturally all these yields will snowball over the game and in the end become unreachable. Which is what I wanted to show. If the outlier player generates 100k+ more culture then anyone else in the game over a 45 turn period, where the second best AI managed 44k in total which is still better then my 33k it's something weird. This is not a one-time occurrence since it has happened in game after game. There is a pattern. I could have cared more but I don't adjust my entire game around winning a city-state quest for 3 happyfaces in the capital. But these are not "feelz"-values but updated turn by turn in the tooltips shown in game.

Perhaps this is as Gazebo said previously just improvements in the early AI and has nothing to do with any bonuses, or that they are of an insignificant nature and have been in the game for some time now. I'm not ruling that out. I'm saying I think the bonuses appear to be larger and not of an insignificant nature and perhaps should be controlled a bit. Someone that is eras ahead of everyone perhaps doesn't need more bonus showers. Or perhaps with the improvements in the AI the bonuses just means more, in which case perhaps they are to large and not needed anymore. Or can't the AI perform these early improvement unless they have the bonuses? But if they only need them early to setup their game why do they keep getting them in the late game where it just compounds the outcome and feeds the beast so to speak. That I was told was rubber banding and evil.

 

[Cokolwiek]

I don't really need to finish those games since I know that they are unwinnable after a certain stage. There is nothing that can be done ingame at that stage to close the gap that is created. Eventually I might catch up techwise but I can never close the culture-gap. My thinking now is that I need very specific starts and civs to make it even remotely possible to win or even keep up with the AI.

My thoughts exactly. And valuable input. Which is called "whining" around here.
In my two last games, one on deity, one on immortal Bismarck was the initial runaway from ancient, and stayed at the top (or second spot on immortal), increasing his gap between all others throughout the game. I entered atomic first, on deity, but it was just for show I can. I wasn't able to grab ANY wonder since classical with tradition.

I am sure THEY WERE pretty substantial bonuses on settling cities, introduced in 2-18 and was a blacklash against them then.

There is a gap between warmonger and peacefull due to "room (or space, not sure which is used)"
If you can win deity with warmonger 30% times you can never win it with peacefull.
If you can win deity 80% times then you can win it with peacefull too but then deity is too easy, there should be a difficulty above it for warmongers.
Should it be like that? Frickin obviously not, but that gap is just the nature of CIV5 (and most 4x games)

You written all this such messy I don't know what are you saying exactly.
The gap in my opinion is mostly because of design choices in VP, which tremendously benefit warmongering or at least as wide as possible approach, and I, along with many others like @BiteInTheMark pointed that out many times. Inflation of yields from industrial on, factories scaling on their number, corporations benefits, literally no drawbacks to hapiness and conquering everything around.
There is no efficiency penalty to yields and no "moral" penalty to conquering, foreign citizens work as efficiently and happily to you and make your soldiers the same rate as your native citizens, instead of hating you for conquering them.

 

[Balint2]

My thoughts exactly. And valuable input. Which is called "whining" around here.
In my two last games, one on deity, one on immortal Bismarck was the initial runaway from ancient, and stayed at the top (or second spot on immortal), increasing his gap between all others throughout the game. I entered atomic first, on deity, but it was just for show I can. I wasn't able to grab ANY wonder since classical with tradition.

I am sure THEY WERE pretty substantial bonuses on settling cities, introduced in 2-18 and was a blacklash against them then.



You written all this such messy I don't know what are you saying exactly.
The gap in my opinion is mostly because of design choices in VP, which tremendously benefit warmongering or at least as wide as possible approach, and I, along with many others like @BiteInTheMark pointed that out many times. Inflation of yields from industrial on, factories scaling on their number, corporations benefits, literally no drawbacks to hapiness and conquering everything around.
There is no efficiency penalty to yields and no "moral" penalty to conquering, foreign citizens work as efficiently and happily to you and make your soldiers the same rate as your native citizens, instead of hating you for conquering them.

Just lol
I give up

 

[Cokolwiek]

Just lol
I give up

You wasn't contributing too much anyway, so okay for me.

 

[crdvis16]

Lots of people posting about how deity should be challenging to the best players and then posting how they think deity is broken and too hard. Perhaps they are missing the most obvious conclusion- that they aren't in the category of best players and should consider dropping to immortal? I was a deity player previously but I'm playing immortal for now until I can move back up (if ever). I'm quick to drop down because I do put self imposed restrictions on myself because I enjoy the game more with those "rules".

Also, perhaps even the best players would need more than just a week to adjust to a difficulty level that can challenge them properly? If deity was too easy and then was adjusted to a proper difficulty level I would fully expect a transition time before true deity players could start winning at reasonable rates again. If anything, it might be a good sign that it feels too hard right now- it forces you to shed lazy play that's probably developed while being unchallenged.

The two above reasons are why I think a lot of the posts lately sound super whiny. It's the reason CrazyG was asking how many games people have even played on the new patch- a totally valid question and point to make.

Having said that, I also think there is an additional wrinkle to all of this. There is a big difference between deity with a runaway AI and deity where that runaway doesn't happen and the AIs are more clustered. I think the deity runaway is like deity++ and has more to do with luck than anything else. I think those games are the ones that feel helpless whereas if no single AI snowballs out of control then the entire pack is more manageable and the difficulty feels more normalized. I think this can happen at any difficulty level probably- there is just a large variance currently that can make games at the same difficulty level feel drastically easier or harder and it's mostly just luck I think.

 

[Kim Dong Un]

There is no efficiency penalty to yields and no "moral" penalty to conquering, foreign citizens work as efficiently and happily to you and make your soldiers the same rate as your native citizens, instead of hating you for conquering them.

I understand some of where you're coming from with all this, but this is wrong:

• Science / Culture yields are increased by 7% for every non-puppet city
• "Moral" penalties for conquering manifests through the diplomacy between you and other civs
• There are "resistance" periods once you conquer a city, in which that city is unable to perform action, while adding it's entire population onto your unhappiness until a courthouse is constructed
• Local unhappiness exists regardless of the city; if a city is unhappy it will produce military at a slower rate, and this is fine for simplification

 

[Balint2]

You wasn't contributing too much anyway, so okay for me.

Again, if you just plain nerfed Wide (player and AI) it would make Wide AI worse compared to Tall AI
What you need to do is add more "space" to tall play, because building orders tech orders, policy orders are not enough. Or improve wide AI logic.
(Also its either "so messy" or "in such a mess" and no, I dont think it was messy)

 

[Recursive]

I am slightly baffled here really. I share data gathered from the game, via the in-game tooltips. But that apparently isn't good enough cause it doesn't come in the proper formatted csv logfiles. Even if I enabled them now this late in my current game I'm somewhat sure it would be dismissed cause I didn't have logs from the beginning of the game. So I said I would run them in my next game. But from how it's received I'm not even sure I would bother at this stage.

Logs nullify observation and confirmation biases. Civ is a complicated game with many different ways to get ahead, and even very knowledgeable players might misinterpret the cause of something.

None of the things I have said have been based on "feelz", unless you somehow think that the data shown in the game is from the feelz and is inaccurate. Or don't you trust me to read a tooltip in game, write it down and then post it here with some extra text to explain what happened. The data I did provide is not "feelz", it's data from the game in a format that just happens to not be the form of the preferred logfiles. It might not give the full picture but it should be enough of a picture.

No one is accusing you of lying - but it's unclear what exactly the cause of the AI getting so far ahead is, and you cannot get that data without logging. If it is the AI's periodic yield bonuses, the logs identify what event is triggering the bonus. If it isn't, it could be a lot of City-State quests, or very skilled play, or even a bug, or some combination of factors. No one can act on the information if it isn't accompanied by data.

 

[Cokolwiek]

(Also its either "so messy" or "in such a mess")

Thank you, I am not native English speaker, I make mistakes quite often.

Again, if you just plain nerfed Wide (player and AI) it would make Wide AI worse compared to Tall AI
What you need to do is add more "space" to tall play, because building orders tech orders, policy orders are not enough. Or improve wide AI logic.

How would you suggest to add that "space" to tall? I think some more % increases from wonders to yields only in the city would be a good start.

• Science / Culture yields are increased by 7% for every non-puppet city

• "Moral" penalties for conquering manifests through the diplomacy between you and other civs

• There are "resistance" periods once you conquer a city, in which that city is unable to perform action, while adding it's entire population onto your unhappiness until a courthouse is constructed

• Local unhappiness exists regardless of the city; if a city is unhappy it will produce military at a slower rate, and this is fine for simplification

I am more than familiar with those, I am mostly authority warmongering player. But you know what all is it? That's right. It's not enough. Because of so much production, population, vassals, everything, you build such a snowball it isn't competition anymore. You can thwarth everything with so much production, be a leader in science, in culture, in city-state and world congress simultaneously. There's a reason that wide and warmongering authority are considered the easiest setting by immortal or deity players.
Many times I feel like science/culture yields penalty is WAY more important and hindering to tradition when they overstretch a little too far with fourth city for example instead of three.
"Moral diplomacy penalties" aren't really nothing when you are so strong they won't dare to act and it is you who are choosing next targets. And they will trade with you anyway. Even equation for the vassals to rise up against you makes them 100% safe for the rest of the game.

I think it would be fun to try a version, in which wide conquering is harder to mantain, if even by a little. Some penalties to production, or science and culture from city that is obviously foreign and hostile until it integrates two era earlier, would be reasonable.

 

[Balint2]

How would you suggest to add that "space" to tall? I think some more % increases from wonders to yields only in the city would be a good start.

Maybe "space" is not the word
Anyways, you would add "space" to tall by making tall play more complex and thus the human smarter compared to the AI in tall play.
Because in wide play the human is much smarter than the AI. In tall, not.

Making tall more complex would mean new techs, policies, buildings, diplo stuff mechanics, etc, ..
And not smarter tall AI that can use these new complex features.

This is what I said is probably out of the scope of this project.

Also what you are saying is not bad, its creative, I like it (if someone is willing to implement it)it just wouldnt achive what ur saying it would, tall being viable for players equally as wide.

 

[Cokolwiek]

Making tall more complex would mean new techs, policies, buildings, diplo stuff mechanics, etc, ..

Now, I see what you have in mind. And while I agree that this would be extremely though and is not what the team focuses on, I think it is a way to go. I, myself, am contructing a mod that will rebalance infrastructure a bit, mainly increase the cost of buildings, lower their yields, and rebalance or rather delete most national wonders which are a little awkward right now and make them normal buildings, just very expensive, and very situational, terrain or policy depending. Also make some normal buildings more wonder-like like military academy which basically every nation on earth has/had one. That would nerf warmongering, while providing more realism.

 

[Cokolwiek]

Also what you are saying is not bad, its creative, I like it

That is the first time I hear something like that on these forum. Thank you. It's quite touching to be honest, considering how simple those words are, in light of the reception of me on this forum (which is to say it politely: non-favourable or ignored at all). I may have tendency to explain myself in harsh words, which is somewhat unintended, but people around here are many times getting overly sensitive and in turn taking me as super confrontational, which mostly I am not. It is refreshing to me that even if we had misunderstanding due to my poor language skills, we were able to work it out and find common ground.

 

[CrazyG]

There is a big difference between deity with a runaway AI and deity where that runaway doesn't happen and the AIs are more clustered.

I agree. There is a suggestion in another thread to adjust AI bonuses for completing wonders, which I think is worth discussing. I haven't yet dealt with a snowballing civ, and truthfully that is something that certain strategies don't have a response to. It is often the same small group of AI civs who start grabbing all the wonders and get out of control as well.

But I still think the early game bonuses are manageable if you have a good build order, take the optimal social policies, and take the right beliefs.

 

[Balint2]

Just to note, you can find the exact bonuses the Deity AI gets on this page:
https://civ-5-cbp.fandom.com/wiki/AI_and_Difficulty

Quick question about AITrainPercent and AIPerEraModifier: do I understand it correctly that this means in Information era deityAI builds all regular buildings and units in 1 turn because everything costs it (70-7x10=0) 0% of its original production cost, so 0 production?

If so Ill get back to it.

 

[Recursive]

Quick question about AITrainPercent and AIPerEraModifier: do I understand it correctly that this means in Information era deityAI builds all regular buildings and units in 1 turn because everything costs it (70-7x10=0) 0% of its original production cost, so 0 production?

If so Ill get back to it.

It starts at 70 then is reduced by (era * pereramod), so 7 * 10 = 70

30% of 70 is 21, so in the info era on Deity the AI's production cost is 21% of normal

 

[looorg]

So I enabled logging. It's slightly annoying when you run many and multiple games since they share logfile and there is no real distinction between them except civname and turn numbers.

Anyhow. Restart. Random stuff. Deity. Epic speed. Normal sized map. 8 civs 16 city-states etc. The normal stuff. First 100 turns (+- a turn). For convenience I have summarized the DifficultyHandcapLog.csv by player and event.

It might be a matter of opinion and wording here but it's a fairly substantial amount of bonuses being dished out if you add the numbers up. But then how much is to much or enough is hard to say, it's like how long is a string. Who knows.

Settling the Capital, 73 gold, gap, food and production. The gold might not matter all to much at the start, but it shortens the time it takes to buy a unit or make a downpayment for some nice building in your new shiny capital. 73 gap is 12% on the way to the first golden age assuming they cost 600 for the AI -- I forgot to check if they get some bonus there to. 73 food guarantee that you'll have pop growth for the next two or three turns. 73 production should complete either the shrine or the monument or a free scout, second warrior or possibly even a worker.

Getting 73 free food is like every AI civ being Spain, if you are Spain you get a large double dip here. Some other civs will do better then others for sure.

Settling a city. 73 gold, gap, food, prod, cult and science. So it's even better then getting your capital on.

Building a wonder and Generating a great person is 73 gold and gap.

Era change. 219 gold, gap, food, prod, cult and science. This is as noted 3 times the normal yield here for everything.

T74 The most sophisticated culture already has 4 policies, everyone else have 3, I only have two but then I went shrines before monuments as I started as Spain and wanted to try the new "Goddess of the Home". It might not have been the optimal choice but I didn't know about my competition really. I didn't, or plan to in the future, log-cheat.

T88 First era change (Netherlands). With some level production bonuses etc that 219 production should be a free settler, or an ancient wonder or a few buildings and/or units. It should be a free tech to, or a serious discount on one.

Triggers, Name, Gold, GAP, Food, Prod, Cult, Science
9 Siam 803, 803, 584, 584, 511, 511
6 Poland 584 584 584 584 511 511
6 Netherlands 584 584 584 584 511 511
6 Maya 584 584 584 584 511 511
7 Celts 511 511 365 365 292 292
2 India 146 146 73 73 0 0
1 France 73 73 73 73 0 0

Siam already has 2 wonders (T41-Stonehenge and T72-Mausoleum), I have 1 (Pyramids, first wonder of the world), India has 1 (T78-Temple) and then the Celts got one (T68-Statue of Zeus).

Siam should already be in a golden age since the first one is just 600 points (for a player, possibly even less for the AI, I didn't check the full list). Poland, Netherlands, Maya and the Celts might be there if they got some GAP from another source. If not there yet they should be there soon.

As noted above Siam already triggered 9 handicap events (or 8 if we don't want to count the capital).

0 Siam capital
33 Siam city
41 Siam wonder (stonehenge)
57 Siam city
72 Siam wonder (mausoleum)
84 Siam city
95 Siam era change
97 Siam great person
98 siam city

And so far Siam has as noted gotten 803 free gold and gap, 584 bonus food and production, 511 free culture and science. Not bad. Poland, Netherlands and Maya are on the same level at the moment, Celts slightly behind but about to cash in to Siam like levels. I guess France and India will get there eventually. I would love a big army and just crunch them now while I can but I sadly picked differently while playing and didn't know the result until after turn 100 when I checked the logs.

Two world religions are already founded. Siam (given considering they got Stonehenge) and the Celts.

India and France have been having superbad times, they are my neighbors. They have been ravaged by barbarians and have not even managed a second city yet. I have not even screwed with them, yet. India gave up and built a wonder and I'm not sure what France is really doing until I watch the log but I'm guess building units desperately.

I don't know about the other civs in game yet since I have not found them, the map is a bit weird and there was to many barbarians about so I have not had a proper world scout as of yet. But I know how it looks around India and France. I could take the other things from logs. Netherlands, Poland and Siam are about to finish the first policy tree (they all went tradition). So I'm already two policies behind them. According to the logs they are all at 10 or 11 techs already and researching the first tech after their era changes in 8 turns instead of the normal 20+ turns.

India, 10 techs, 4 tradition, 9 pops, 1 city, 1 wonder.
France, 9 techs, 3 authority, 3 pops, 1 city, 0 wonders.
Spain (me), 9 techs, 3 progress, 17 pops, 4 cities, 1 wonder.

France recently built a settler but I have not seen it yet. Barbarians might have eaten it. India has not yet finished one. India built a trader they are sending to me instead.

I think we have very different ideas about what is insignificant or minor bonuses entail. I'm not really seeing the smart plays. I'm seeing the bonus rain fall over them. Which may or may not be what they need but what they get are not minor or insignificant amounts in my opinion.

I'll bet at least a gold that Siam will snowball hard in this game.

 

[Ragic]

Maybe they should just remove the difficulty labels and make it a slider. I get the feeling there are egos in play. Maybe not. I’m just sitting here eating popcorn.

 

[MIS]

Maybe they should just remove the difficulty labels and make it a slider. I get the feeling there are egos in play. Maybe not. I’m just sitting here eating popcorn.

I agree. I was going to make a whole post about Settler is too easy on this patch, but moving up to Chieftain isn't an option because I've always played Settler, and if I can't lose on Settler while making myself use all the exploits and resources the game has to offer, then they need to make it harder.

But I don't know if a slider would help - wouldn't people debate how hard the highest level should be? @CrazyG already won a Deity game on the 3/1 Beta, so I don't understand people saying it's too hard.

 

[Recursive]

A slider isn't really doable anyway AFAIK.