Background rationale, feel free to skip:
At higher difficulties, it's taken as a given that you will be (massively) behind the AI at first, then surpass them later (if you are to win).
Frequently I hear the inflection point is somewhere around the Renaissance (for several reasons, and it ranges from late-medieval to early-industrial).
This has some significant negative balance implications on play across different difficulties. If a medium skill player at an appropriate difficulty for them tends to have different game dynamics than a high skill player at an appropriate difficulty for them, they'll have different incentivization models. E.g., on the discord folks have been wondering if the balance between ancient and medieval social policies is off. It was mentioned that medieval trees tend to not help a player who's behind as much, so Deity players are frequently taking a second Ancient tree, while lower difficulty players tend more towards Medieval ones.
It seems to me that we're frontloading too much of the difficulty bonuses (especially as they accumulate at higher levels), and perhaps not making the lategame as challenging and dynamic as it could otherwise be to allow it to be possible to even reach there. This is in spite of C bonuses being designed specifically to allow for this use case.
One option would just be a general shift from A towards B or especially C bonuses.
However, one consideration is that you cannot give AI a bonus that affects the lategame without also affecting (to at least some degree) the earlygame - all 3 of the bonuses have identical scaling in Ancient/Classical eras, there's no way to 'delay' a bonus from starting at all until later in the game.
Proposal:
Change B bonuses to D bonuses, and C bonuses to E bonuses in the below tables. This simply zeroes their ancient/classical values. Modders retain full flexibility to retain current functionality by simply boosting A bonuses by a matching amount. As part of this, Recursive et al should make judgment calls as to what portion (if any, up to in full) to retain by doing so. Now, however, modders will have the ability to reward lategame AI triggers without being shackled to considering how those bonuses change the extremely sensitive start of game conditions.
Note: this breaks the standard 'by era' scaling players and modders are well used to. This is irrelevant to players because these values aren't visible in-game. Modders would admittedly have to adjust to any systemic change we make to ABC bonuses, but there are other probably other tweaks that could be made to sufficiently move the needle here without inconveniencing them in this way. The traditional reason of keeping Ancient and Classical bonuses the same (to avoid straitjacketing human builds by not incentivizing beelining classical) is irrelevant when this only affects the AI.
Current
Proposed
Alternatives:
- BC bonuses could further break from by era scaling by starting at a 1 multiplier in medieval, but that makes the medieval -> renaissance jump much bigger. My assumption is some of the classical -> medieval jump will be smoothed by augmented A bonuses to partially offset the removals.
- BC bonuses could further break from by era scaling by lowering both ancient and classical to 0, if there were some reason that ancient and classical should retain the same modifier.
At higher difficulties, it's taken as a given that you will be (massively) behind the AI at first, then surpass them later (if you are to win).
Frequently I hear the inflection point is somewhere around the Renaissance (for several reasons, and it ranges from late-medieval to early-industrial).
This has some significant negative balance implications on play across different difficulties. If a medium skill player at an appropriate difficulty for them tends to have different game dynamics than a high skill player at an appropriate difficulty for them, they'll have different incentivization models. E.g., on the discord folks have been wondering if the balance between ancient and medieval social policies is off. It was mentioned that medieval trees tend to not help a player who's behind as much, so Deity players are frequently taking a second Ancient tree, while lower difficulty players tend more towards Medieval ones.
It seems to me that we're frontloading too much of the difficulty bonuses (especially as they accumulate at higher levels), and perhaps not making the lategame as challenging and dynamic as it could otherwise be to allow it to be possible to even reach there. This is in spite of C bonuses being designed specifically to allow for this use case.
One option would just be a general shift from A towards B or especially C bonuses.
However, one consideration is that you cannot give AI a bonus that affects the lategame without also affecting (to at least some degree) the earlygame - all 3 of the bonuses have identical scaling in Ancient/Classical eras, there's no way to 'delay' a bonus from starting at all until later in the game.
Proposal:
Change B bonuses to D bonuses, and C bonuses to E bonuses in the below tables. This simply zeroes their ancient/classical values. Modders retain full flexibility to retain current functionality by simply boosting A bonuses by a matching amount. As part of this, Recursive et al should make judgment calls as to what portion (if any, up to in full) to retain by doing so. Now, however, modders will have the ability to reward lategame AI triggers without being shackled to considering how those bonuses change the extremely sensitive start of game conditions.
Note: this breaks the standard 'by era' scaling players and modders are well used to. This is irrelevant to players because these values aren't visible in-game. Modders would admittedly have to adjust to any systemic change we make to ABC bonuses, but there are other probably other tweaks that could be made to sufficiently move the needle here without inconveniencing them in this way. The traditional reason of keeping Ancient and Classical bonuses the same (to avoid straitjacketing human builds by not incentivizing beelining classical) is irrelevant when this only affects the AI.
Current
Ancient | Classical | Medieval | Renaissance | Industrial | Modern | Atomic | Information | |
A | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
B | 1 | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
C | 1 | 1 | 4 | 9 | 16 | 25 | 36 | 49 |
Ancient | Classical | Medieval | Renaissance | Industrial | Modern | Atomic | Information | |
D | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
E | 0 | 1 | 4 | 9 | 16 | 25 | 36 | 49 |
- BC bonuses could further break from by era scaling by starting at a 1 multiplier in medieval, but that makes the medieval -> renaissance jump much bigger. My assumption is some of the classical -> medieval jump will be smoothed by augmented A bonuses to partially offset the removals.
- BC bonuses could further break from by era scaling by lowering both ancient and classical to 0, if there were some reason that ancient and classical should retain the same modifier.
Last edited: