The Perfect (strategy level) Engine (Team, and everyones) input requested

[Gazius]

What would you like to see in an ideal strategy game engine, with possible tactical battles. This engine should not only be able to respond to the ideal of the team's scope for EB, but for all history up till this morning.


Obviously AI is going to be a huge one, but what might contribute to a better AI, or what needs more depth modelling? My current idea is based around the fact that it's individuals who really do move the world, and thus everything associated with the nature of people will play out on a grander scale. My goal is also to include a detailed look at economies, and how it truly is trade that has lead to great nations. The rise of Rome, the rise of the Caliphate, the dominance of the West in the age of imperialism, and todays major players in the globe, all engaged in trade in markets. But what else plays a big role that gets downplayed for the sake of making your basic strategy game to appeal to the lowest common denominator. I assume we're all here because we want depth in our games, to some degree.

To this extent, I may have a follow up sort of concept that I'd also appreciate input on from those of you who've managed to read this far. It'll need a bit of revising on input from here, but I hope to be able to present to you a concept statement later on.


The quote that started me off;

Incidentally, it would be very interesting to me to see a campaign map system where you gave every army a destination as part of movement, as well as engagement parameters ("pursue and attack all enemies encountered/pursue and attack half strength enemies encountered/actively avoid all enemies if possible/force march/defend location X and the nearby vicinity from any enemy spotted"). When you hit "start the clock" every army on the map would advance to where they had been ordered by the player or the AI and battles would occur during an abstracted length of time (week/month/year) until you got back to the strategic approach again. However, the AI required to maintain a consistent grand strategy for dozens of enemy factions would be potentially crippling to code, if not confusing as hell for the player without some abstraction, so that's on my wishlist for stuff I hope to see in 10 years or so.

[bovi]

Actually, on a tactical scale the game Laser Squad Nemesis has these properties, to a degree. There are less options for contingency plans, but at least you can decide whether to abort any remaining orders in favour of shooting what turns up, and opportunity fire (shoot at whatever might be in this direction when you take aim, if anything).

[Perturabo]

Some form of research and development that is based upon environmental and trade stimulus would be grand.
Most strategy games have research that is disconnected, as in you can research and develop anything with enough time, regardless of what cultures you are (or aren't) in contact with, and regardless of what material resources are available.
Please note that this does not mean that I would like parthians with machine guns or something equally silly

All I ask for is more realistic development of technology.

[Parkev]

I thinking I'm going to push the definition of a strat-game and go a bit RPG, but IMHO the quest for complexity can only end in a smaller scope considering the finite capacity of the brain. I haven't thought much into recruitment as I realised half way through this. And yes I am willing to wait 20 years for something of this nature, largely in the AI department I think, but this could work as an online game.

1. Change focus from god-like control of factions to a single character (=family member) that exists within a politcal structure. There is a maximum number of characters that form political structures (of which there are a number of; Kingdom, Republic) through loyalty. Political structures have offices (Governor of Illyria, Satrap of Media, King of Averni) which are occupied by characters, that fufill the responsibilties of the office. All characters essentially operated as individual factions; with a relationship scores based on situation, past history, good deeds. All offices have a list of canditates and a selector (you ask selector, with which you have good relationship score to put you on list), once the office expires a new holder is chosen. Once your character dies, you begin play as his (or her) heir (of your choice) and proceed as a line through history.

2. Split the Campaign map into two levels
a. Strategic level which is not realtime and more of a conceptual map of the greater world based on intelligence availiable to the player's character. Communication (and orders to subordinates) could be sent to other characters, with an appropriate time delay.
b. Tactical level which is real time (with pause button on tactical level. Marches are ordered at certain speed with equivalant risk of ambush/exhaustion). When the characters army moves close enough to another army/settlement/whatever interface descends into an RTW battle map equivalent.

3. Factions and political structures are fluid. If an empire has a weak leader, enterprising characters will establish greater autonomy, new factions just appear in other words.

4. All tiles have ethnicity and age distribution scores which determine the number, quality and type of troops recruitable. Different modes of recruitment would have to be represented. (Plenty of RAM or big tiles, take your pick.)

5. Provinces are removed and settlements are made fluid, all tiles have probability of spawning a settlement and these settlements grow and decline based on economic and social factors. The Romans burn Carthage to the ground and found a trade port down the road in a small fishing village, why should Carthage still be on the map.

[Strategy]

And yes I am willing to wait 20 years for something of this nature, largely in the AI department I think, but this could work as an online game.

You'd have to, and by the time it came out, the game would be ancient by the standards of the technology of the day.

Been there, done that, got the T-shirt.