>Can anybody think of good examples from history or literature where a country is made up of a small number of very specialised cities?

USA, USSR had many cities but they were specialized. And even if no, one of the best games I played in several years waas Puzzle Quest were battle is represented by moving diamonds in a table.

The point I feel I need to make is that there is an important difference between citizens and population. Population was in the game long before citizens and I don’t think anyone wants rid of it. Originally population was just a means of obtaining levels. We had merchants and workshops you could build one of in every city, and so the only way to increase production of gold or materials was to get more cities.

This lead to city spam. Far worse than it is now. It was recognised that there was a need to make bigger cities much more powerful than smaller ones and encourage players to develop what they had instead of just spamming cities as close together as they could. In fact, for big cities (rather than numerous ones to be optimal, a city with 2x the population needs to be MORE than 2x as productive, but we’ll get onto that later).

There are two ways to increase production for bigger cities (instead of forcing you to build lots). You either increase production with population, or by levels. Ultimately you want a bit of both of these things. But since population increases all the time, you don’t want to be constantly worrying about it (this is what happened with citizens). The benefits of population should be automatic. The micro-management stuff is better broken up into stages so that players can deal with it all at once.

With gold production, Stardock actually got it REALLY right IMHO. Taxes earn you 0.1 gold per person per turn. This production increases automatically as population goes up, but you don’t have to run around building more buildings all the time as your pop goes up, each of which generates one gold per turn. It’s all handled automatically. The only gold buildings that a player has to build are at level up (ie, the % bonus buildings). So a city with 2x as much pop earns MORE than 2x as much because it has both more population AND better infrastructure.

What we need to do is migrate other types of production onto this sort of system. More pop = more production & more levels = more buildings. Population should determine both these things. But it makes sense to control the number of buildings by level rather than population itself, because that way all the town planning can be done after level up, when the player is notified about it. Plus, the devs are able to specify exactly what sort of buildings & production a settlement of a given size has. The maximum number of buildings for a max level city can be set BELOW the tile limit so the player never reaches it, but the player still gets the benefit of ever increasing population because his buildings produce a certain amount per citizen.

So at each level a settlement has a certain number of slots for buildings. A level 1 settlement might get 3. One produces 0.05 gold per person, another produces 0.1 research per person, another 0.1 materials per person and another 0.1 spell research. At the next level you can add another such building and so on. This gives the player all the current choices between specialisation, based on population, but in a far more controlled manner. Plus, the player gets to get all his buildings out of the way at the level up stage (assuming he has the building materials of course). This also demonstrates the redundancy of charging gold maint / turn per study in the current system. You need 5 guys and 1 gold to run a study, and those 5 guys generate a certain amount of gold per turn. So you’re effectively just saying the same thing twice. Far better surely just to produce less gold in the first place and ditch the maintenance. It would make the maths easier if nothing else. The only thing a player has to worry about when choosing his new production building on level up is what he would like more of (a positive decision), rather than whether or not he can afford to run it (a negative one). The real expenses would be in running armies, building things and running special buildings like barracks, bonus production buildings and other specials. The production within a city would be largely self sufficient, so it would be much more difficult to suddenly go into the red when you conquered a city that focused on materials rather than income.

This is only half the question though. Population growth is the rest. But this kind of system would help turn settlement development into something that just “happened,” rather than the key to winning the game. That’s what the war of magic is for.

That companies within the same industry tend to cluster in areas instead of spreading out is a pretty well-studied area in economics.

http://en.wikipedia.org/wiki/Business_cluster 

>Can anybody think of good examples from history or literature where a country is made up of a small number of very specialised cities?

I think you would be hard pressed to find a single country in western history that didn't have some level of specialization in their cities. Of course, you can't think in terms of "tech research" and "arcane research". Think in terms of "is there a market", "is there a harbor", "is there education", "how is it governed", "what food do they eat" and most importantly "what do they produce?". Now I'm not saying that there was one city that produced all the cotton clothing, and another that produced all the iron. But was there a tangible difference between each city? Absolutely.

Agree entirely. But I'd argue that all the differences exist on the level below that one which Ele tries to model. If we had more resources (wood, coal, iron, steal, wheat, fish, cotten, stone, marble, paper, wool) etc then specialisation would work really well. Not that I'm arguing that Ele should introduce all these concepts. That's not the aim of the game. I was just suggesting that some thought should be given to what level of abstraction we want in our model, what strategies make sense, and how they relate to the feel of the game and it's world.

I would suggest that for gameplay and feel, it would be easier to adopt a capital + fiefdoms model, which gels with a lot of fantasy settings (King's Landings, Minus Tirith, and of course, Ankh Morpork!).

Or, as a wildcard, crazy-put-it-just-might-work, imagine a Setters of Catan style model for settlements. Very abstract, but would create interesting decisions. I doubt it would work but I seem to get going through a "board game mechanics are great" phase.

IMHO - Because EWOM is Frogboy's baby.  He won't (can't) abandon it until it grows up to be something he's happy with.  It doesn't matter at this point if only a handful of people care enough to keep playing it.  Call him obsessive, Type A, illogical, a workaholic, whatever.  Just don't ask why he won't stop.  It's not gonna happen.

Can anybody think of good examples from history or literature where a country is made up of a small number of very specialised cities?

Finance: New York. Government: Washington, DC. Food production: Chicago. Industrial production: Pittsburgh. Research: San Francisco.

It's a game abstraction, but it's a useful one -- otherwise, we'll be playing SimCity: War of Magic. Ancient Greece (pre-Phillip/Alexander) actually isn't a relevant model, because the city-states were independent rivals and fought amongst themselves (the Peloponnesian war in particular was long and bloody) until Phillip and Alexander came down from Macedon and conquered them all. And, of course, Alexander's empire fell apart quickly after his death.

On the other hand, I do agree with you that having some different city/state 'models' could help reduce the same-old-same-old factor that tends to plague EWOM.  Fall From Heaven 2 had one civilization that could only build a few (~4) 'real' cities but could spawn lots of colonies that stayed at an EWOM 'level 1' equivalent but could be used to grab resources and spread borders. On the other hand, I never liked playing that civ very much, so there you go.

I do agree withAlLanMandragoran above that a lot of city spamming (at least, a lot of mine) is to grab key resources. In particular, since one of the victory conditions in the game requires controlling all four types of shards, you have to be able to build (or conquer) cities with those four types in their ZOCs. Likewise, any food-producing resources are critical as well. One of the earlier Civ versions (2 and/or 3) had an option where you could build a road to a resource outside of your borders and then mine/exploit it, but since I already find the ongoing "bandits attacking caravans" issue one of the single most annoying aspects of EWOM -- seriously, folks, let us spend money and research to create armed caravans and be done with it -- I'm not sure this would work all that well (though it would certainly fit with the caravan-road-building model). 

Besides resources, my other major reason for colonization is to grab critical geographical choke-points. This is especially true in EWOM (vs., say Civ), because sea-transportation in EWOM is difficult and primitive. The far-too-cheap terraforming spells make it easier to create your own chokepoints, but heaven help us all if the AIs start figuring out how to use Lower Land, Raise Land, Destroy Land, Raise Mountain, Volcano, and so on.  ..bruce..

Hmm, modern U.S. examples:

• Detroit: Car production
• Seattle: Airplanes
• San Diego: Military bases
• Los Angeles: Movies and Entertainment
• Washington D.C.: Bureaucracy
• Oklahoma City: Dog and people food

Historical examples:

• Southern Italy: Tomatoes which were imported from the Americas
• Amsterdam: Drugs, Fish, Financial Services
• Sparta, Greece: Spartans, Duh
• Vienna: Music, Education, Filtered Coffee
• Jerusalem: Religion
• Beirut: Olives, Cedar exports, Education

If you research history you'll find a lot of regions specialized in different food products which were exported over a large area. Salt was a valuable commodity, as were high concentrations of various minerals such as iron, copper, gold. Most places that actually have cities developed where they are because of river or sea ports for exporting the products of the region and importing the products of other regions.

So yeah, I find it rather realistic. Does not mean however that it always makes for good gameplay. City placement was fun in beta 1 before they took away local resources and went on the made march to make everything "global" in favor of a bad RTS and TBS genre crossover.

Thats the reason why i stick with SD and Elemental. Its not only about making a game to make money. They love their products. In this case WoM. And you don´t throw something away, only because its not that what you hoped for. I think this does concern all sides Brad, some of the "WoM Team" and some players like me.  

aye, probably anything beyond shortswords (bows included) should require a type of specialty building in the city. (or city-level, or both)

just for the sake of argument, lets assume weapons come in 4 tiers (with (melee) wooden weapons being tier 1, and daggers and short swords being tier 2)

tier 1 requires nothing ... tier 2 requires appropriate resources ... tier 3 requires appropriate resources + city size.

tier 4 requires appropriate resources, city size + Specialty building.

lets say only 1 specialty building for all melee weapons, and only 1 specialty building for all ranged weapons.

Re city development & specialization & planning:  how about some one-of-this-group-per-city buildings instead of or in addition to the 1-per-city and 1-per-empire buildings ?

One group would be consist of buildings with some % bonus to  gildar / tech research / arcane / food / recruitment speed / mining / shard harvesting, etc.

There might be a 2nd group with fixed-amount bonuses like +10 tech research, +10 gildar, etc.

Any city could have one from group 1 and one from group 2.  If I tried I could come up with a third group.

For best results, using tech as an example,  find a lost library, build a city near it, build as many studies/libraries there as you can, pick the tech buildings for both group 1 & group 2, and pick tech as the specialty when leveling-up.  Also, of course shovel in a school and university if those remain, but I suspect at least 1 would be re-positioned as a one-of-this-group tech building.

--------------------------------------------------------------------

A whole different approach:  focus on the level-up bonuses.  Add options for farming bonus and "nothing" in addition to the tech/arcane/gildar options, and proclaim that ONLY ONE city per empire can get each bonus.  Ordinary cities get 'nothing'.  That's for level 1 -- for other levels you are locked in to what you picked for level 1.  Then they might need to re-balance by making the bonuses bigger, or making some building like schools help more.

I don't like this a much, but it might be easier to implement.

-------------------------------------------------------------

Yet another approach:  in addition to the old 5-square limit on how close suburbs of different cities can get, forbid founding a new city with its center less than maybe 20 or 25 squares from the center of any existing city.  That largely kills city spam and indirectly makes planning / managing the relatively few cities you have more important.

I think they could actually do all 3 of these if they felt really ambitious.  Right.

----------------------------------------------------------

Meanwhile, re Brad's original idea on population costs,  from his 1.29 beta log post  "LOTS of code to handle the increasing costs in terms of people to run your improvements. If anyone thinks that getting highly talented people remains equally easy as you grow, ask google, microsot or apple about that. It's not."

... he wasn't perhaps projecting a bit, possibly traumatized for life by what he has had to pay to attract talent to the frozen northern wastes ???? 

It might have been easy at first, hiring locals who had been to programming school but didn't want to move to silicon valley or Seattle, but once he has to start paying people to move a place they don't really want to live -- yikes !      (southern Kaliforniac here, letting my climate prejudices show)

BTW  "microsot"   ?        

So you are not freaked out by DPS in turnbased game...

Aha, so that was the objective behind the whole hit point reduction thing. Okay, fair enough, I understand. The issue is that the rest of the variables remained unchanged... destroying the military aspect of the game.

The issue is the following: the mid-game weapons are in the 20-damage range. Per soldier in a unit. If a soldier has 3 hit points, we reach extreme levels of silliness: armies with 80 power, and a dozen hit point.

The hit points weren't too high. They were too low. 12 HP per soldier isn't enough when some weapons can easily reach 40. 20 or 25 would actually be a sweet spot, with armor and training being there to prevent late-game 1-hit-kills. With 20/25 HP per soldiers, the armor and weapon system becomes easy to balance, and you can finally have decent fights between groups.

But then, how can one balance heroes so that they can fight groups on an equal footing ? Well, you have to increase their defense a bit... and their attack power a lot. That means giving heroes a lot more HP than soldiers (twice ? three times ?) and making their attacks a lot more effective. 

Here's one of many ways to do the latter: introduce a "Strikes" stat, that tells you how many blows the hero will deal in ONE attack - but never twice on the same soldier in a group. Make it start at 2 or 3, and make it grow with level or Dexterity/Strength. 

This allow heroes to deal damage to several soldiers in a group at once, effectively multiplying his attack power. Since the health of soldiers isn't tracked individually, what this means is that each attack is actually made of min(groupsize,strikes) "subattacks", with only one counterattack, and using only one action point.