Elemental, and what would work from it in GalCiv III

City building: The specialization/generalization balance is nicer. Maybe building upgrades have a place in this game, too? I think I've seen the concept bandied around a bit. Not only housing should get better as the game goes on, methinks.

Starbases: I'd honestly be all for 'military outposts' and 'trading posts' in this game. The former would replace putting shitty cities at choke points. The latter should enhance trade routes and maybe allow resource trading with some sort of global pool. That could be fun.

Trade goods: These were kind of out there in the original, but they might make more sense here - sort of 'uncovered relics' from the past that grant advantages but can be traded. Then again this might call for a diplomacy overhaul so as not to further mess up the system.

Anomalies: There are a lot of 'incremental advantage' sort of goodies in GalCiv that I feel might make sense and give us a lot more variety than the current 'PILE OF GOLD' 'OH LOOK ANOTHER PILE OF GOLD' 'MY GOODNESS IT'S A POINTY STICK, WE DON'T EVEN KNOW HOW TO MAKE THOSE' - like, 'You find a beautiful stone statue from before the cataclysm, still relatively intact and consign your peasants to return it to the nearest city. This should attract more visitors and migrants to our town. +0.1 Prestige.)

Research: I like the new system but it may be a little too 'linear'. A research tree should have more branches. Kinda like the GC one.

Military: I'm gonna skip over this one as combat is a heaping mess right now.

Special Abilities: They were too good, sure, but they gave a hell of a lot more flavour and differentiation than we have now. We need more to distinguish the different factions, else why have factions at all?

AI: This is an interesting point and I'm going to consider it some more, but here's the thing. The WoM AI currently doesn't even factor magic enough to make a damn difference. So effectively, it's just a terrestrial empire building, management, and combat AI. There are far better examples of this in existence. What went wrong?

There's a lot of good stuff in this thread, but I just wanted to quickly answer this point:

I'm not too concerned about the AI at the moment; it simply isn't complete yet. There isn't a proper AI in the game. That is to say, that the AI hasn't failed the game - it just doesn't really exist yet. There's some placeholder AI that is able to play the game, but isn't intelligent at all. I would give AI time - Brad is more than capable of writing good AI. He just can't do it in one night. Especially when the game mechanics change so much as the game is being developed.

Give it time.

i really like resources and really hope that the GC3 economy would be more fun if it was about achieving a balance of resources than messing around with those confusing and counter intuitive production sliders and focus buttons.

i really think this would would help with the atmosphere of the game. most great space games have a sense of the infinite expanse of the cosmos and the unthinkable possibilies contained therein. your evil opponents should have the threat of the unknown. the gal civ games have failed at this regard. a lot of it is because of the art style, writing and the cartoony presentation that stardock seems to like so much. the fact that solar systems are just squares apart doesn't help much: imho, we have a scroll wheel, so it should be possible to get a better sense of scale within the current combined space&system display, but if not consider separating the two.

btw, if i see the snathi in another stardock game i am going to scream. this stuff just screams "amateur." to me.

the battles do not need to become tactical, but they do need a lot more choices in ship design and fleet composition. we need fighters to be outmaneuvering big capital ships and the such like. the recently released indie game "gratuitous space battles" is a perfect example of how this should be done. the battles also need a better sense of scale, with car sized fighters buzzing around kilometre long battleships like flies.

the build your own settlement mechanic in elemental will be great when ironed out and this should have it's equivalent in GC3 in being able to colonize any planet. we should have binary star systems, habitable moons, rogue planets and all this stuff. every planet needs to feel completely unique and believable, rather than the one number wonders in GC2. they should be generated on some sort of logic, with larger planets having heavier gravity, far out planets having lower temperature and other at least pseudo-scientific stuff. i want to know the gravity, temperature and orbit period of every planet. as well as capacity we need to have a number to represent the habitability of the planet (taking into account all the above factors). a barren mercury-style ball of rock might have a large capacity for industrial buildings and make a great shipyard, but have low habitability because no one wants to live there. so the player takes the hit of low income for having a great shipyard. or contain a resource or precursor research bonus.

resources are the way forward. no more floating gems in space. they should be based on the planets and moons we colonize. remember the tibanna gas mines on cloud city? this kind of stuff. whether they actually work as requirements like in elemental is arguable. i'd like to see a system where they simply made things more or less expensive. so if you don't have high enough hydrocarbon supply, weapons on your ships are more expensive. if you don't have enough titanium, building is more expensive. etc. get rid of all the other random, arbitrary, gamey "+10%" bonuses from anomalies and techs. they're just distorting the mechanics. remove production for military and social and give us just a research slider. we should just pay for units and buildings straight up and make the industrial buildings reduce training times (and costs?)

Pure Win

Elemental's city system works pretty well without the need for secondary or tertiary settlement types. If you want a military base on a choke, set up a city with a hedge wall and don't upgrade it with buildings -- military base accomplished. If you want a colony, same thing except resource gathering building instead of hedge wall.

Really, the city system has so much right I can barely believe it. Using food as a limited city upgrade resource was a stroke of genius. 

Only a few changes necessary.

-add the ability to really upgrade fortress towns without respect to city level, just have a cost mechanism where higher level city pays less for defenses

-add the ability to fire on a single approaching stack to every city, natively. buildings improve damage and range and aren't limited by level, level also improves damage and range

-remove workshop/study/loremaster spam. replace with specialist pops that are local and work in more powerful structures, with baseline +3 research building 1/city that can be used right off the bat

-more city specialization, more ways to specialize cities

-remove unit queue, one queue for both buildings and units *VITAL* forces player to forgo economic development when rushing

-administrator/bureaucrat pop who is used globally for %bonus buildings/city maint buildings/etc

-luxury resource + food resource required to build specialist pop huts -- more things to fight over, more stuff on the map

-specialists of various types: aristocrats (food bonus buildings), merchants (gold bonus buildings), academics (research bonus buildings), etc

-ability to build "factories" that require materials and output luxuries (one factory of each type per civ), would help with materials overload

-remove magic research, dump all spells into magic tech, cut to three tech trees (magic, civ, lore/adventure), forces player to specialize and prevents players from going for both weapons and spells at the same time. also stops rewarding players who get a few spells and then focus entirely on tech/gildar production (which basically breaks the game, since it means you've got 1/3rd more tech/gildar power than civs that do a 3-way split)