The part of Civ series that gets me going is the simple map-centred mechanics. Military operations exist on tiles. Supply bases and fortifications exist on tiles. Terrain advantage is part of the tiles. You can set up a difficult fortress city 1000 years in advance by settling on the right tile.
A game that is more committed to a grand strategy basis of gameplay might do away with the tiles, to just having map zones and military presence that's just a score, but I will always look to Civ for my fun because I get to plan around battles for tiles.
To me, that is the core way that military and economy get to intersect for the gameplay. The fertility of the land influences your settlement decision, as well as land development. But terrain impacts defensibility and will motivate you to claim certain tiles with military presence, to have advance warning or just to deny positioning before an assault. This is from the defensive side of things; Amplitude's
Humankind has some sort of 'outpost' settlement, which plays the role of military forward supply , and so there is that angle to consider too. I think, despite my love for the military part of the game, and affinity for winning my games with that skill, that the "perfect 4X historical game" would be about society and its development - an approach which makes the emergence of an armed caste something that would be computed as such. I hunger for systems that start with the Civ3(2?) idea of persuading the republic to authorize a war, and then blow that out in scale. Again, it looks like
Humankind is ambitious about being the shaping of your culture as gameplay.
The thing about the tiles mechanical core is it makes the design of things seem easy in a way. We just have to allow the right elements to interact with each other and everything technical, like balance, can be configured in a data file. In a mod file. So, about the military part of the game, I want supply to be something you have to worry about, up to an extent. And I want militaries to be able to pillage or besiege cities as situated, extended locations on tiles - this, they already do, and in Civ6 you have the typed plunder bonuses for improvements and districts, so that's a plus.
I want the quality of life fix to the damnable 1 unit per tile limitation. Tactical games also have 1upt but they include mechanics that prevent "Absolute" chokes from emerging, such as an order to merge units to restore HPs. Actually, the game I want does have X units per tile, and the magic number is 10. Remember Civ V, which had 10hp units? Consider allowing up to ten units on a tile, and reprise the combat rule where every fight destroys at least one of the participants. But don't stop there. I think what became noxious about stacks was the way that the initiative belongs to the attacker, the player who gets to make all their moves on their turn, with no interruption, and with the benefit of chaining conditional decisionmaking upon the successes or failures of that turn. I believe that stack warfare should be programmed to engage the stack with the stack, instead of a series of unit-to-unit engagements. Keeping this simple would be an important priority, as we must remember we are only trying to make the players pressured to care about certain details, the details that I believe the player wants the (military) game to be about. Supply, and control of the fertile tiles for the settlements.
If we got careless about stack combat, we might end up creating a de facto unit workshop system, where we add bows to spears and forget about the mob we create a la carte. That isn't what I want. I want military engagements, on the strategic level, to have options of risk or delay or whatever that interplay with the international objectives for the player's civ, and also to interplay with the addition of new weapons (of course), and, this is the big one, to interplay with technological progress so that we can introduce the idea of Doctrine to the game. My friend is not a history buff but he tells me that about what happens in the into-Atomic period are advances not of any kind of materials science, but of military doctrine so that kinds of engagements are possible that were not conceivable before. So we have a basic idea for a military order to look like this:
- Players give orders for maneuvers to military stacks, which key off of terrain, defender standing orders, and doctrinal know-how for their effectiveness.
The unit-of-meaning for the player to decide, is how to defend or reposition at tiles, so that we have a kind of fighting that is still easy to grasp at a surface level, but allow players to make a few more kinds of decisions that concern the domestic side of gameplay, at all of its stages, settling, developing, and maximizing.
-----------------------------------------------------------------------------------------------
The other thing I feel strongly about is demographics visibility. I think it's true that the game needs to have some way that is public-public (i.e., everyone knows-that-you-know) of tracking the most successful player, so that you can see yourself losing as it happens (replays would be ideal but it seems memory intensive? but you have to be able to take a lost game and look at what others did right.) But at the same time, I don't like the way that, for example, in Civ 5 you can, depending on things, have exact knowledge of critical things like hammers to hammers for deciding on the midgame stretch. This is where espionage comes in. I pine for Civ4's espionage points system. If we could streamline the per-civ sliders (and fix the slider haywire glitch), this provides a system for observing rival player status which also interacts with an individual Spies system.
Do people like controlling spies in Civ6? I think they do, but they need to have the right kinds of missions for it to be fun. Civ5 was limited but was clean. However, there is a logic issue with Constabularies merely slowing down opportunities to counterspy. Civ4 and Civ6 claimed to be conscious of the poison problem, where you just sabotage and sabotage, throwing off everyone's careful math, producing more grief than enjoyment. I don't think I want espionage to be a subgame that is super powerful, but this goes toward my bigger philosophy for a "Perfect 4X game" to be an Ecosystem of Rivalries.
In my view, the perfect game has a few ways and means within it for you to try to prove you are number 1, and by going down some of those roads, you get to reap a benefit for outplaying your opponent there. But these multiple systems do not all make for decisive victory. You can play a game with a weak espionage element, and you would lose some things to an espionage specialist, but you get to recoup in the main through an excellence in another capacity. I don't know much else about bringing this vision into existence, but I would think that the WRONG way to do it is by having each subsystem run "refrigerated", with its own power source. I mean, once you unlock spies in Civ5, you have spies, so you use them! Once you have a source of +

, you start pooling it, so you have a religion game! The ecosystem vision would need to have more of where special subsystems are bought with a common resource, instead of just powering themselves. I could also note, a real time game can balance these things because of the common attention resource, but a turn based game instead must provide a direct, and deliberate, limitation on such "perfect optimization". You have to pay with giving up some other effort. However, instead of being solvable with an Excel spreadsheet and patience, it becomes more of a gambit interaction that depends on some of the hidden information of your opponent's choices - and their skill.