How Difficulty Bonuses work for AI

On the note of AI bonuses, I find the massive bonus against barbarians really stupid. Pathfinders attacking fortified spearmen and clearing the camp is just silly. And it doesn't really make the game that much harder either, it just forces you tip toe when damaging camps

 

I agree - the AI is competent enough now that it doesn't need that vanilla handholding.

G

 

It also might contribute to overvaluing of authority.

 

that doesnt make sense to me, disliking a poor function for being poor is perfectly legit. its bad enough instant bonuses currently do nothing for happiness (already a solid reason to avoid them where you have a choice), i dont see why we need a smorgasboard array of yield modifiers across the board that do nothing for instants either. I agree it wasnt good to have big ones on few key must-have buildings like vanilla did but their ubiquity now isn't a good thing either. Why should core basic infrastructure get %'s on them? the only reason i can see is for ease of coding and using existing functionality, but its not like there arent plenty of ways to add rewarding mechanics that wouldnt utilize snowball inducing, strategy specific bonuses like the modifiers. and its a lot harder to get control over yield creep with these %'s everywhere

I agree they have place in the game, i just dont think that place is on basic buildings everyone gets (baths, opera house), monopoly modifiers, or so many social policies before tenets. Again, I think 2 people taking the same policy shouldnt get different rewards just because one person may have struggled with barbs a bit up to that point or while the other was building infrastructure. the struggle already created discrepancy, giving % based reward just enforces it. Ditto poor terrain starts. things like the synagogue or stele dont need to be removed but could probably just be lowered, and faith was added to strat resources for Ethiopia so they can afford to lose a little honestly.

ya know if baths gave a flat instant yield on entering a golden age people aren't gonna stop building them or feeling they are rewarding to build. and I know opera house has been changed a bunch of times already, in fact most of the things now that have % on them didnt at some point in the last year (roughly?), its almost like at a certain point you just started added % instead of doing other things to balance stuff out. i just think that trend went overboard a bit, fealty is a good example. castles dont need to give %10 food, if it was +5 food i'd still be happy to take the policy and build them.

 

just remembered corporations add a ton of % also, wern't on the list and hadnt been mentioned already but i knew i was forgetting more.. and still am probably. at least they come late

 

Nope. I added them in primarily in later-game buildings (Baths being a niche earlier building) to promote tall play. Too many flat yields benefits wide play over tall. As it stands, your arguments are largely 'this doesn't feel the way I want it to feel,' which is fine (you're welcome to your opinion), but it's hardly a way to run a business balance this game. I thrive on numbers, not 'supposeds.'

G

 

I literally just said % modifiers dont apply to instants, how is that a feel? its true. Id say my points were all valid and unchecked so far as ive seen but youve only just now presented a relevant counterargument which is that more flat yields would somehow benefit wide more than tall, which honestly i didnt think was the case but if thats what you say id take your word on it over mine, fair enough. and yeah short of us just acknowledging theres around ~%200+ modifiers available for gold, food, prod, science and culture i have no magic numbers for you

 

Right, but in very limited situations would you have the sum total of all of those modifiers running at once across the board. They're very often exclusive or at the opposite ends of viable strategies/builds.

G

 

The game speeds up in mid-turns. That is why we need quadratic bonuses to AI. CIties no longer generate yields in tens, but in hundreds. Everything is inflated.
It is not because one building or one policy is op. Everything stacks up. Plus base yields are much bigger - tiles produce more, buildings produce more, specialists produce more.
There’s no easy way to stop that or limit at least. Just systematic nerfing and maybe strict rules. Example. No single modifier bigger than 15%. 5 - standard building or policy, 10 - wonders, super-policy, 15 - uniques, single cases like spaceship factory.
Limit dynamic modifiers to +25. Influence, tourism, etc.

 

I think thats moving in the right direction.

just curious, why does progress have a %20 production modifier in it? its an ancient era policy designed to benefit wide, and G was just saying how % are there to help tall. but Expertise is so powerful it is literally the reason people are even looking at taking 2x ancient trees vs the medieval ones... and as we are just discussing how to bring parity to unit and building purchases, you realize thats impossible for anyone who took progress with Expertise still existing as it does.

 

Can’t chisel with a hammer.

 

If you didn't steal that from a song, then you ought to write the rest of it.

 

I am not familiar with that expression, not native speaker here.

 

my impression was he was attempting to express your suggestion would be too big of a change. although i've seen him say before he'd rather make big changes first and then adjust from there so people see/feel the impact *shrug*

 

Most of modifiers are already in range of 10-20. There are few offenders, with 25 and more impact. So it is not a big change actually, depending on the rule itself. Maybe +20 as max is ok, idk, i was giving an example.
I think it is important that a rule exists at all.

 

Gazebo is saying that his coding tools are too imprecise to accomplish what you suggested.

 

Nuclear plants give %50 but because they come so late I actually see them as a good thing and not causing any problems. Some rule or guideline is cool but it'd have to incorporate timing as well, i.e. no %20 before renaissance or something

really? now im curious to see him explain the statement lol

 

You can't use a coarse tool for a precise task.

Big, specific changes, not sweeping changes for all modifiers of a type. You and I both know what is being discussed, twisting words isn't helpful.

I don't think a hard limit is necessary - besides, as you note, we already do follow a guideline of sorts.

Nope, that's not what I meant.

G

 

Since we defined it almost identically ("Gazebo is saying that his coding tools are too imprecise to accomplish what you suggested"), I assume you're saying that you may indeed have the tools, but the method Infixo suggested wouldn't do the trick.

If not, what did you mean?

 

I'm saying that Infixo's broad rework is not needed when we can make surgical changes to problematic modifiers.

G