Discussion in 'General Balance' started by Stalker0, May 17, 2018.
I test with no bonuses on.
So in an actual game with a player, AI would have growth bonuses and per-yield followers will be even better for them than other followers.
I didn't mean to remove AI growth bonuses during AI testing, but in actual game.
Yes? Which is why I want to nerf them slightly...
It's not egregious, just consistent.
He means that makes their power varies widely across difficulties and between human or AI.
Something based on population balanced around 100% pop growth cost is very likely to be waaaaay better at 70% on Deity.
What are your thoughts on AI growth handicap? Honestly interested.
I think you misunderstood me; It wasn't a question, it was a statement, but never mind.
That's exactly my point, thanks ElliotS.
I think part of the challenge for humans at higher handicaps is realizing that some game elements will be stronger for the AI than humans. That's the point of the handicap. I don't balance against it, I count on it.
One thing to consider: If the AI growth handicap isn't matched with the production handicap, then the AI will end up with a higher yields per pop ratio, forcing the needs medians upwards.
Barely. And, even if it does...that's part of the handicap. Not sure you all truly understand that handicap = intentionally making something harder for yourself.
Intentionally is a keyword here, Gazebo. Intentional AI bonuses may lead for some unintentional results. However, if those results are also intentional, then everything is great. We don't know if something is truly intentional, so we ask to clarify it.
Let's take a hypothetical situation:
Handicap A Is exactly as it is now. Functional, but makes some strategies not viable and makes AI's power-level vary widely based on if they choose the now-OP beliefs/policies/etc.
Handicap B is made without discounted population growth. The other values are balanced in such a way that the end result is the same difficulty and has no gameplay impact as warping as extra population. This leaves all strategies as equally viable, but harder on this difficulty.
Which one is better? I think B is clearly better.
The reason I said it was hypothetical is that I don't know if it would make certain strategies or policies OP the same way extra food does. Progress could jump to a huge win percentage for example. I don't think OwlBeBach has done enough testing to have real empirical numbers on that.
I have also been playing with tuning the growth/production/etc bonuses for AI. So far, by returning those to original values (100%, or equal to a player’s values) and slightly bumping up the Difficuly Bonus A, B, and C, it seems more balanced yet still not any easier. Unless I’m going aggressive, I find it very hard to win even on Immortal (though I’ve blended it so much with Deity bonuses I’m not sure you can call it either Immortal or Deity). I don’t think AI NEEDS those discount bonuses to be competitive with the human. Those bonuses do mess with balance though I’m sure.
Not nearly as much as is claimed. I’ve tested and found the difference balance wise to be pretty minor. I feel as if it is tinkering for the sake of tinkering, which is fine, but that doesn’t mean its worth the change.
Yes, up to the point where it starts to really hurt the balance. I think i've made enough examples of that before. In fact the real reason, why i was proposing it was to fix misunderstanding of people on the forum when it comes to balance suggestions.
Problem is that people playing King think that lets say Goddes of Love is a decent Pantheon and you have the same result in your tests. And then when me or ElliotS play the game - we see that Goddes of Love is a peace of crap for human but AI dominates the world with it. And then me or ElliotS say that something is clearly imbalanced - but other people do not see that and you do not see that in your tests. (Or maybe you are just too much a democrat in heart and try to value all opinions equally)
@Gazebo you said that you test AIs without bonuses, so according to wikia, is like playing on Settler: http://civ-5-cbp.wikia.com/wiki/AI_and_Difficulty But I've read somewhere on this forum that VP is balanced mostly for Emperor difficulty. Do you agree with that statement? If not, then which difficulty should be considered the most balanced? My intuition says, it would be Settler, because you run tests in these conditions and possibly balance basing on that.
As time goes by, the more I wish the system was reworked to reduce the growth bonuses the AI has and replaced with others. Or at the very least made less front-loaded, so that they'd kick in (more) later and less at the start. As it stands now, the AI plants down a city and within 10-15 turns it has around 7 population...
I wonder, if the AI is not good at religion, why don't just give it religious advantages? Maybe +10% to faith generation and to passive pressure.
I think one of the bigger current problems is the founder AIs not being able to keep their cities following their religion (compared to the human player). I wonder if it would be prudent and possible to code that cities belonging to founder AIs/owner of a holy city cannot be targeted by missionaries of other religions (so like the Spanish UA) and/or have the foreign religious pressure in those cities decreased. I think that might help the AI to keep 4-5 viable religions, while leaving non-founders and city states as fertile grounds for spreading/converting.
If AI has a religion issue it's with choices. I've found their faith generation is fine. And while some AI have been slow to spread (I assume waiting to enhance first), the ones that do are still frighteningly proficient.
How about Veneration changes? Maybe XX Faith (scaling with era) for every built unit/building?
Or +Faith every time you make a Faith purchase, as well as another bonus.
Or -10/20% Faith costs.
Or +3 Faith for every city (if majority religion, doesn't matter if they're converted), and +5 Faith to Palace and/or Granddad Temple
Or reuse Stave Church graphics from Into the Renaissance, and just turn it into another Building. Maybe
+1 to all yields
-5% to all needs
+20% pressure, resistance
A few of us were debating a -10% faith cost idea in another thread. If there aren't any big abuses (would need to check through with Theodora) I think its a good idea.
Separate names with a comma.