A vision for civ

[Hail]

first post updated!

I urge you to read it before posting!

[GermanJoey]

Are you planning on implementing this game?

[antisocialmunky]

I tried that once and then I counted the resources I was willing to put into it. 4x is too expensive to indie easily.

[Sirian]

Hail:

Like Aeson, I think the corruption model had some strong points that Soren's maintenance model lacked. It's a Hobson's Choice, with the least-wrong answer differing depending on the player's tastes and priorities. Soren's maintenance system does a good job of limiting growth early and gradually easing the boundaries, while the corruption system permits REX. Maintenance enforces a fixed city count, while corruption lets the player play the map.

The two factors that tip me into preferring corruption are:
1) In Civ4, the "extra" cities obtainable once the bounds loosen must be captured not self-grown, because the self-grown ones are coming too late to make much of an impact.
2) I chafe hard (harder than Sulla and others, apparently) at the fixed city count. I think it strips out a number of factors from the corruption model that enhanced my enjoyment of the game, including playing tetris with my available terrain to try to get the most out of it, instead of purposely wasting terrain to try to get the most out of each of a limited number of cities.

I see Civ5's tight economic bounds as a progression from Civ4's tight economic bounds. Jon just dialed down the number of cities (or tried to) that can be operated within the artificial constraints. The corruption model was much more free, letting the details of the map shine instead of the lead designer's chosen number of cities for you. Jon just followed Soren's example on imposing his own will on how the game would play out, and chose tighter boundaries. I think Sid's original system, flawed that is, is less flawed than the arbitrary city count. Of course, I admit that corrupted cities are not as much fun to play as uncorrupted ones, because they can't do much, and fully corrupted ones are just as worthless as late-founded cities under Soren's model, but at least having them does not ruin your game.

The problem is, unchecked snowballing ruins the game. Without either preventing cities beyond a certain number, or imposing harsh diminishing returns with each city added, the game becomes unworthy of being played. The winner is determined not by strategy or conflict, but a pure math game on land grab: he who has the most, wins. He who first takes over another civ will run away with the game. That's a worse outcome than either corruption or maintenance, as there *must* be boundaries.

The best boundaries I ever played under were those of MOO1: the fuel range on your ships. Each civ would REX until the map was full, and then it was a contest of whether you could hold on to what you'd claimed. Some of the worlds having hostile environments you couldn't immediately settle, combined with the gaps in the map these created, blocking you from expanding farther in certain directions until you either could settle those worlds or expand your ships' range, was brilliant. But how to apply that model (or something similar) to the tile-based maps of Civ, I am uncertain.


On the other hand, square tiles and the fat cross are painfully outdated. The hex tiles and the organic growth of cities one plot at a time raised the bar permanently. The old "expand once from 3x3 to fat cross" is older for me than it is for the rest of you (now seven years out of date instead of four) but I could not conceive playing under that system again. I fired up Civ4 about two years ago and felt a deep shock at how "primitive" the fat cross felt after being neck-deep in the hex-grid design and programming and the organic city expansion through Civ5's development. ... If that doesn't persuade you, then let me add that both Sid and Soren concur. Sid originally wanted to do hex maps but was already trying enough radical things in Civ1 and had to pick his battles. Soren's been working with hexes for a while now.

I know that Civ5 has some "deal breaker" design choices in it, in regard to most RB players, but it also moved the ball forward in a lot of other, subtle but important ways, and the hex maps are one of those items. The exact details of the organic city expansion, one plot at a time, hold room for further refinement, but the fat cross would be like Star Trek Continues in that original fans may love it for its authenticity, but it is limited to being a hobby for a handful of people. Its rustic trueness can only appeal to the original fans, so it's dead-on-arrival as far as a new commercial for-profit adventure goes. Niche material at best, even if fun for the diehards (and worth your time if you are such a diehard).

If you actually made a new empire game with square tiles and the fat cross, that would carve a big fat-cross-shaped hole in your sales.


- Sirian

[GermanJoey]

There is no fixed city count in Civ4. Just look at recent multiplayer games, PB18 for example - expansion is not really limited at all other than by other players.

[T-hawk]

I know that Civ5 has some "deal breaker" design choices in it, in regard to most RB players, but it also moved the ball forward in a lot of other, subtle but important ways, and the hex maps are one of those items.

This is very much true. Firing up Civ 4 feels like taking a serious step backwards now after a couple dozen games of Civ 5. Some examples are deterministic Great Person spawns, more organic religious spread, less swingy barbarians able to harass but not destroy, and of course the natural gradual city expansions. And there's quite a number of other subtle smooth edges in Civ 5 where 4 was still a bit rough.

We think alike in many ways. I never minded the corruption model from Civ 3. It was a perfectly valid way of applying diminishing returns to expansion. It just failed in the visceral experience. It trolled the player by showing right on the city screen in big red icons what they "should" be producing but the game tauntingly withheld to deliberately screw the player. I can look past that but most Civ 3 players couldn't.

Civ 4 solved this by hiding the expansion brakes under several layers of mathematics and spreadsheet screens where most players would never venture. Civ 5 solved this better for the mainstream by promoting the brake to a core gameplay mechanic in plain sight that you have the options to manage. Civ 5 really applies diminishing returns by raising the effective cost of each settler by way of happiness. But playing along with that feels like good strategic play for the mainstream. "I need happiness, build a colosseum" is good strategy for casual gamers even if it's obviously trivial for us.

Incidentally, that whole corruption system was an artifact of Cold War politics back in the time of Civ 1. The more autocratic government types imposed tighter corruption restrictions, until you reached Democracy with no corruption at all. It was an idealistic commentary reflective of the times that Civ 1 was birthed and the system stuck around through SMAC to Civ 3.

The best boundaries I ever played under were those of MOO1: the fuel range on your ships. ... But how to apply that model (or something similar) to the tile-based maps of Civ, I am uncertain.

You could literally slap a max range on settlers. Maybe a soft cap, allow paying to extend it once the settler is on the move. Has ramifications for island maps, of course. Just throwing out ideas.

On the other hand, square tiles and the fat cross are painfully outdated.

The idea was that culture beyond the fat cross was supposed to matter a lot more than it turned out to. You were supposed to put effort into 3rd and 4th ring expansions to claim isolated resources, or to compete for resources on border pushes with a neighbor. This failed once players figured out how to economically expand everywhere so no need for longer reach, and moved to higher difficulty levels where the AI easily outpushes you on borders. Civ 5 does a much better job of making culture always matter and yes going back to Civ 4 feels primitive by comparision.

[Huinesoron]

The best boundaries I ever played under were those of MOO1: the fuel range on your ships. ... But how to apply that model (or something similar) to the tile-based maps of Civ, I am uncertain.

You could literally slap a max range on settlers. Maybe a soft cap, allow paying to extend it once the settler is on the move. Has ramifications for island maps, of course. Just throwing out ideas.

I wrote up a long idea along these lines, then realised the problem: it seems likely to have pretty much the opposite of the desired effect. With reference to this:

The problem is, unchecked snowballing ruins the game. Without either preventing cities beyond a certain number, or imposing harsh diminishing returns with each city added, the game becomes unworthy of being played. The winner is determined not by strategy or conflict, but a pure math game on land grab: he who has the most, wins. He who first takes over another civ will run away with the game. That's a worse outcome than either corruption or maintenance, as there *must* be boundaries.

Limits on settler range - I was thinking something along the lines of food supplies - reduce your initial expansion to the local area - but have no impact on conquered territory. In the absence of corruption/maintenance/happiness/something distance- or city-count-limiting, conquering an enemy civ - even a small one, even a single city - becomes absolutely imperative. Plug out a super-early army, go take the capital of your nearest enemy, and bang: suddenly you have twice the ground to slowly expand into. Heck, if you wander past the first capital and take out your second-nearest opponent, you can build a pair of empires in tandem, back towards the one in the middle, crushing him in the cradle and giving yourself triple your 'allotted' land for very little cost. Even though your new cities are close-in... you can plant a lot more of them.

Plus, of course, limited settler range is like an open invitation for ICS... hey, they can't go anywhere good with this range, so better just to pack them in as tight as I can. More cities = more production, after all.

[Bobchillingworth]

The way AoW 3 handles ICS is interesting. Settlers cost a substantial chunk of a city's population, similar to Civ 3 (except more severe). Cities whose population falls under a certain threshold are "underpopulated", and have their gold, hammer, research etc. production significantly cut back until they regrow. The larger a city's population (represented by increasing a city's "level", from "outpost" to "metropolis"), the greater its yields. Cities are also "constrained" when their cultural borders touch- the percentage of a city's border touching another border is applied to its growth rate as a negative modifier. Thus a dozen small cities crammed in together will struggle to meet the economic output of four large ones.


It's not a perfect system- some abilities give cities a flat gold other bonus just for existing, which encourages spam, and 12 mediocre cities will do a much better job producing an army of cheap units than 4 super sites. But at least "fewer, but better" is competitive with "many, but individually weak".

[Krill]

2) I chafe hard (harder than Sulla and others, apparently) at the fixed city count.

That's just not true, unless you are specifically talking about high difficulty levels in SP play...and at that point I'm not exactly sure where the difference is with civ 5, apart from civ 5 having an AI that sucks more which necessarily means there are more viable options.

On the other hand, square tiles and the fat cross are painfully outdated. The hex tiles and the organic growth of cities one plot at a time raised the bar permanently. The old "expand once from 3x3 to fat cross" is older for me than it is for the rest of you (now seven years out of date instead of four) but I could not conceive playing under that system again.

At the same time, I loathe the lack of control I have over a cities cultural progression. I loathe the removal of tactical city placement and how it fits into a strategic economic game. I've never thought that squares are better than hexes; I don't care which one is used in a game, the graphics are not something I care about in a TBS game. Game play wise, there isn't much difference either IMO, although there are specifically fewer axes of movement for units which can lead to a limit on unit combat. I feel that the lack of flipping tiles another contributor to this; that removal is a step back. In reality, I don't feel like the Civ 5 implementation is a step forward at all. It takes 2 good ideas that added to what already existed would have been a great leap forward and instead is the game design equivalent of the gold plated turd. I'm sorry if that offends you, but it really is the most apt way to describe it.

[Sirian]

The best boundaries I ever played under were those of MOO1: the fuel range on your ships. ... But how to apply that model (or something similar) to the tile-based maps of Civ, I am uncertain.

You could literally slap a max range on settlers. Maybe a soft cap, allow paying to extend it once the settler is on the move. Has ramifications for island maps, of course. Just throwing out ideas.

I wrote up a long idea along these lines, then realised the problem: it seems likely to have pretty much the opposite of the desired effect. With reference to this:

The problem is, unchecked snowballing ruins the game. Without either preventing cities beyond a certain number, or imposing harsh diminishing returns with each city added, the game becomes unworthy of being played. The winner is determined not by strategy or conflict, but a pure math game on land grab: he who has the most, wins. He who first takes over another civ will run away with the game. That's a worse outcome than either corruption or maintenance, as there *must* be boundaries.

Limits on settler range - I was thinking something along the lines of food supplies - reduce your initial expansion to the local area - but have no impact on conquered territory. In the absence of corruption/maintenance/happiness/something distance- or city-count-limiting, conquering an enemy civ - even a small one, even a single city - becomes absolutely imperative. Plug out a super-early army, go take the capital of your nearest enemy, and bang: suddenly you have twice the ground to slowly expand into. Heck, if you wander past the first capital and take out your second-nearest opponent, you can build a pair of empires in tandem, back towards the one in the middle, crushing him in the cradle and giving yourself triple your 'allotted' land for very little cost. Even though your new cities are close-in... you can plant a lot more of them.

Plus, of course, limited settler range is like an open invitation for ICS... hey, they can't go anywhere good with this range, so better just to pack them in as tight as I can. More cities = more production, after all.

The MOO1 range isn't just the colony ships, though. It's all ships. So a similar principle on civ would be unit range, not settler range. Then there would be no jumping out with an early attack force to capture a rival capital far outside your regular range.

The MOO1 range did have an extender, a "long range" module you could add to large and huge ships. It was very expensive and difficult to do anything with, but occasionally necessary, sometimes because your randomized tech tree might come with too many range-extending techs missing, to where your first range expansion might be from 3 to 6. Super painful to go that long with just the starting range available.


The lack of range in Civ could well lead back to ICS. But of course, the tiles themselves lead to ICS. And the anti-ICS measures put in place to prevent rampant abuse end up distorting the game in undesirable ways.

Civ5 suffers from a second problem: UCS. The Uber City Strategy. With no caps on "tall" expansion, the uber capital with 3x the intended production rate could claim any wonder at will (can't grab them all, but can have its pick), finish all the improvements rapidly and benefit from the mulitpliers while the old-style multi-city "wide" empires had few if any buildings done yet and not be getting yield multipliers. The uber city could churn out an early army faster, and more. There's an extensive array of measures in place to block UCS from being as broken as ICS, and those distort the game as much as the anti-ICS measures do.


Another area where Civ5 really shines is the shedding of the distance-based costs. Those always favored the most compact empires, to the point where in Civ3 people worked out the exact "perfect ring" formation, which was so powerful at exploiting "ties" in the corruption formula, it overrode nearly all terrain considerations in its results. Just place your cities at the exact right spots in the "perfect" formation and experience the least amount of distance corruption possible, overall.

Civ4 still has the distance component, which makes it burdensome to create a nation like Chile, even if the map demands it. Chile occupies the southern half of the west coast of South America, with the Andes at its back blocking any inland expansion. It's a long, narrow country that nevertheless holds together well because of its unique terrain situation. ... The distance-based costs in both the Civ3 corruption system and the Civ4 maintenance system fail in any situation of this sort, whether it be a long thin stretch of useful land or a need to span your civ across multiple landmasses -- the further apart, the worse the results for you. Civ5 can do the "New World" scenario successfully, where you start with cities in Europe and try to expand across the ocean to America. That's a garbage formation in Civ3 and Civ4: you just don't do it. The game economics operate on a set of assumptions that your civ is going to be contiguous on a single continent.

Civ5 falls down in other, new ways, in regard to the size and shape of your empire, but it does solve/fix (or at least progress toward a fix) for many of the old, inherited problems. Where Civ5 fails the most is where it tried to fix an old evil and imposed a new, even worse evil upon us in its place.


- Sirian