First step to a successful game? Fleshing out the design, and then having it hammered by other developers. So here is the design to a game I called: Shattered Lands. Shattered Lands would be a real time strategy game based on strategy instead of micro-management. The game would begin by generating a overview map (think LOTR style...large map), and would randomly decorate it with several "cities." Each of these cities would be neutral, and two cities would be labeled as player capitals (the cities would be the two cities which were the furthest away). The goal of the game is to kill the other persons "commander." The commander is a special unit (the only experience based unit in the game) who embodies the player themself. So the game works like this: To conquer a city, you must breach its walls and slay all enemy forces (this might need work). If cities are not owned by an enemy player, they are neutral, but still hostile (and therefore, will be AI controlled and will fight back.) To march upon a city, you must have one city "connected" to it. Cities are connected if they are within 20 miles of one another (Think ut2k4 onslaught). Upon conquering a city, you gain access to all its buildings and its people, who generate stone, wood, and gold daily based on their location. Mountain cities will generate more stone, etcetera. There is no micromanaging here: income is calculated daily based on assets minus peasant expendentures (each peasant requires 1 wood, 1 stone, etc). Each city also would have anywhere between 1 and 10 plots of land, upon which you can build several types of buildings. These are: Farm -- Generate food for your people House -- Create more peasants Barracks -- Train peasants into soliders. Archers Range -- Train peasants into archers. Smithery -- Research war technology Lumber Mill -- Create lumber Quarry -- Mine rocks Mine -- Mine gold Wall -- These are special, and can be built anywhere Arhcer's Tower -- A defensive structure Capital cities are a bit different: These are the cities that the player starts with, and they each have a castle, and many more plots of land than a regular city. Unlike many RTS games, you would not need to send peasants to do tasks. Peasants would simply exist as a number, and peasants would be assigned to buildings. For example, you could assign 50 of the 150 peasants in a town to mine in a gold mine if you are running low on gold. You must fire them, however, before you train them into soldiers, or something like that. Soldiers would be trained at either the barracks or the archery range, and would be created individually. Each building has a maximum number of peasants available to work there. Peasants also, before coming soldiers, are assigned a house, so once all the property in the town has been filled, new peasants will not become available, unless you make old ones into soldiers. Soldiers and archers are the only units you can actually interact with. To fight a battle, you must march your solders across terrain into another city. This is where you can start thinking star craft, or something like that. The overview map is where you would see your soldiers marching -- and their speed is based on their terrain. Once you arrive at a town, however, it loads like a starcraft map. At any time, you can return to the overview map and click on another city, to "warp" there. This allows several battles to be occuring all over the map at once. City maps basically have the city situated in the center of the map, with random terrain elements all around (trees, cliffs, etc). When an army arrives, they are placed randomly along the perimeter of the map, so as to disallow the player who owns that city to place mass units at an "entry" point. So what happens, Mr. Visage, if two armies cross paths while marching to war? Well, neither player would know about the opposing army until it was too late, and a map would be immediately generated in real time, and the two armies would square off. That point would immediately exist on the overhead map so to allow each commander to send reinforcements if they would like. For now, there are only two types of units: Footsoldiers and archers. This is to help with the simplicity of the game (though, perhaps later down the line other types will be added.) Each has its obvious pros and cons: archers are weaker, but can fire from a distance. One of the features I am pondering adding is allowing certain cities to be controlled by "lords", which would be AI entities which would control the city. So you could set the lord to make the city defensive based, or have him pump offensive units, or have him make it entirely economically based. The goal is to make this game 1v1 network based, so thats how you can imagine it. Probably no versus AI play. So that is the design so far. Comments? Please, be harsh. If it doesn't sound fun...tell me. I am designing this to be a simple 2d tile based game with very primitive graphics (no fancy schmancy animated buildings or anything -- the only animated thing would be the soliders). Thanks for reading!

Sounds genius, i would defitnitly want to play it considering im always been looking for a RTS that influanced more on strategy (I completely enjoyed Warhammer 40k:dawn of war)...Anyways let me cut to the chase as i read through your post i noticed something that could become a major problem, have 2 types of units doesn't seem like a bad idea, but if you really want simplicity you should try and create a check and balance system to make the game fair (I suggest looking at the advance wars series on the gba, the gameplay, check and balance system, and whole thing was perfect). The thing i saw as a prevalent problem came when the words "randomly generated terrain" and "archers" were mentioned, i can see now having a ton of archers played o a cliff or somthing firing down on footsoilders who cant reach them, thus ending there march. What do you think about this "problem", and excuse my spelling.

I dont quite know the level of terrain I will be using (whether I will allow for height differences, as you pointed out), but why would you consider using the terrain as an advantage a problem? Wouldn't that truly be the test of strategy? Having 20 archers fend off 100 foot soldiers while reinforcements arrive? If the archers were able to make it onto the cliff, then the footsoldiers should be able to, right? Playing the terrain would be part of the game, I suppose. Though, you do make a good point: should the terrain be balanced, or should I allow it to be completely random, and let the cards fall as they may for the players...

The randomly generated terrain was probably going to use the perlin method, so it would be far more sloping than abrupt (though, I would probably allow for some abruptness to occur).

Anyway, I really liked the advanced wars system, and if I included another unit, would make the game more rock-paper-scissors like. Soldiers beat archers, archers beat cavalry, cavalry beats soldiers, or something like that. Probably something much more realistic than that...

Thanks for the response!

i don't mean to say terrain differences aren't an advantage, i mean it would prove the ultimate test of strategy to use it to the best advantage(think the avalanch scene in Mulan). But i would suck if it could be taken advantage of, using archers placed in an inaccesable area firing on all your oncoming troops and theres nothing you can do. I mean you bring up the point that if the archers got up there then so should the footmen, does that imply that the map will be symetical?, maybe only one players forces were able to make it that location because they entered the city from the north. And a rock paper scissors would work perfect.

All the perimeters of maps would have to be flat, so that all areas are accessible, no matter where you enter on the map. Perhaps I could even make it so that the player could choose where to enter the map (4 cardinal directions) or something like that, so he can surround a city and attack like that. Who knows. Tossing ideas out like this is exactly what I am looking for.

Maps would not be symmetrical, because I want to allow certain strategies to emerge. Cities will be layed out fairly equally on the major map, but other than that, the terrain will be pretty random, so that the true strategist would have the advantage (if I end up making terrain play that big a role...which I think it should, but this may be too complex for me to handle).

Any thoughts on a third unit? Obviously, an air unit would complete the rock, paper, scissors, but that introduces a whole new level of complexity that I know I wont be able to deal with. Thoughts?

Thanks again

Yea that does seem like a great idea, then you can surround a city and watch your enemy squirm as they realize hope is lost. A third unit, well before i can give good suggestions i would need some information on the other units so then i could balance out my ideas. but as for thrown out ideas...supply lines, dynamic terrain(sandstorms, avalanches, etc), uh i might some other stuff but it'll take a sec.

You have some good ideas there. You might want to look at STRONGHOLD or the SETTLERS series for partial implementations of what you're saying here.

I have some comments on your army placements:
1) You could send scouts ahead of an army, or around it in a circle pattern like the tip of a lance, that would help spot potential threats to your armies.
2) Expanding on nr 1, once your scout finds an army and it is undetected, you could be able to make a surprise attack or lay low and wait for the army to go away until you can move forward safely.
3) About random placement .. it would be nice, but will negate many times any tactical advantage an army has because of sheer luck. Instead, you could generate the map according to the large world map data and place the armies relative to the smallest square/rectangle of the map that contains both armies. This way an army that's coming from the west will still be on the west of the map , while an army coming from the eastern mountains will still be at the east side, and most likely on mountains.
4) The same would be true for attacking a city. You would give an unfair advantage to the city if the enemy army has to be scattered across the map because of random placement, instead of being grouped at the entry point of the map (entry point decided by the direction that they came from. You could look at the TOTAL WAR series' city sieges for this.

Perhaps it might be a good idea to give armies the ability to "camp" in different areas. What I realize is that forcing an army to stay on a path would be just as or more difficult than allowing them to wander anywhere -- and wandering anywhere coudl give different tactical advantages. The city link would still be there, but this could allow for several armies roaming the map, and attacking cities from different ways. This would allow armies and squads to camp in strategic locations. This would also fix the "attack from" city problem I was talking about before.

And instead of having gold automatically given to the player, the resources generated in a city can only be used in that city, and must be "transported" to use in other cities, which would allow some cool tradeline stuff.

I have been playing around with some weather ideas. I know for certain that terrain will change marching times, etc. I have been thinking about throwing in "snow" and "rain" storms, which would be generated randomly on the map and would move around. Anything caught in those would slow down dramatically.

Thanks for the replies guys

A couple of thoughts I had while reading through this:

You mentioned a couple of times that there was no micromanagement, but there is. You've raised the micromanagement level a bit so that you aren't controlling individual peasants, but it will still exist in the game you've outlined. This isn't necessarily a bad thing, but still...

My other concern was that this is a real-time strategy game, but you can "warp" from one battle to another? Play two or more battles at the same time, in real time? I shudder to think... ;)

In response to TaoLung's post, I'd like to offer some vague and useless advice regarding a possible implementation of your battle system. This is targetted at the "StarCraft-like" part of the game.

I'd like to see an RTS game where the only available viewpoint is from your commander character. All the orders would be given via chain-of-command, with some kind of signals to convey instructions from the player to troops in the field, or messages sent by officers on horses. "Blue Chevron Archers, move to support Volunteer Light Cavalry" or "General Retreat" would take the form of bugle calls, flags, or word-of mouth. You'd stand on some high point, using a spyglass and a tabletop map to keep track of the battle's progress. If your position was threatened, you'd have to relocate. That would be sweet.

You could even have a Starcraft interface, but on the model of the battle. You click on a unit's little miniature figure, set a few waypoints, and issue an attack command, and then some guy jumps on a horse, tears across the field to that unit, and then tells them what you decided. Then he'd come back (hopefully) and not only report on his assignment, but update the unit positions and their status on the tabletop display.

It would make for cool multiplayer co-op, too. Heralds running back and forth trying to cement a plan before the situation changes too much would be intense. You'd have to put some effort into preventing players from using other voicechat methods, but you'd never be able to kill cheating entirely.

Edit:

I'd also like to weigh in on what Visage said. For a neat iconic system resembling his suggestion, take a look at Ogre Battle for the SNES. I'm sure you can find some kind of emulator to try it. Not Tactics Ogre, which is totally different, but Ogre Battle. There's also Ogre Battle 64 on the N64, but it's tougher to find. Try the SNES version. It has neat troop movement rules and engagements are largely automated.