Announcement

**TempLeland** · February 14, 2001, 22:26

Design Doc

Version: 0.2 (proposal)
Created: December 30, 2000
Last changed: February 15, 2001
Responsible: ???
Contributors: The Joker, Leland, Amjayee, ElmoTheElk

Contents

Introduction
Game Philosophy
Role Of The Player
Turn System
Map
Units And Movement
Regions
Population
Economy
Combat
Infrastructure And Improvements
Cities
The Rise And Fall Of Great Powers
Social Model And Nationalities
Religions
Politics
Diplomacy
Diseases
Technology
Disasters
Scenarios

Introduction

The purpose of this document is to serve as an overview of Guns, Germs and Steel, a real-time historical strategy game. It is not intended as an exhaustive description of all game features, but as the game development progresses there will be links to appropriate models which take care of detailed design. This document is also a starting point for new participants in the project, and it should be generally accepted among the active designers and contributors.

Game Philosophy

Our game philosophy is to create something more than a mere civ clone. GGS should be a historical imulation as well as being a turn based strategy game.

The game will be a multiplayer game as well as a single player one. We are starting with the multiplayer part, while developing the game, since single player will require a great AI, and such can not be created until the game is further developed. As of now we are not sure how many players a multiplayer game can support, but we are hoping for as many as possible, while still keeping the complexity of the game algorithms intact. An ultimate goal could be to have 100s of players play the same game simultaneously for months. In such a game players would come and go, civilizations would rise and fall etc. It would be virtual history being made!

The game is going to be as realistic as possible. Our philosophy here is, that reality is both immersive and intuitive to the player and that ruling a country through history would be fun. So the game tries to copy that, though alternative histories and what-if scenarios should also be possible. This will mean that the game will have previously unseen levels of realism. Of cause we can not be realistic in all aspects. There are areas where gameplay will have to be more important than realism. In those areas we will then have to consider the two carefully.

Role of the player

An always difficult thing in empire building games spanning 100s or 1000s of years is the role of the player, since no person can live that long. This is especially an issue in GGS, since the player is not in complete control of his civ.

So we decided that the player would be the ”government of the civ”, whoever that is. This would mean that the player would have to work under very different government types, and would have to adjust his policy to fit this. At the same time the player will only have the same level as control as the government would have. This would obviously vary from government to government. If the player runs a military dictatorship he would have more control and elbowroom than a leader of a democratic government, but at the same time there would be penalties to the dictatorship. For example the player would have to suppress revolutionary tendencies at all times.

Turn system

The game is going to be turn-based. We are not, however, going to use the old fashioned turn system from Civ2. In stead we are using a preplanned turn system. This means, that the turns will roughly be divided into two phases: the planning phase and the execution phase. The planning phase is where the player is active. He gives orders to his units, he manages his provinces and everything else. The point is, that all players will do this simultaneously and they may negotiate with each other as well as sign treaties and other contracts. Only nothing really takes place in this phase. The world would be ”on pause”, so player can manage his things in peace. Units would therefore not be moved at all in this phase. In stead you would choose, where they should go, who they should attack etc., and there would be symbols describing the orders that you have given – a line through the hexes the unit would move to, a line and an attack symbol at the unit it should attack, and so forth.

When the player has done all his planning he would hit the ”end turn” button. When all players have done this the execution of the turn will begin, or alternatively a time limit could be set to stop individual players from stalling the game, especially in multiplayer games.

Here all units will move simultaneously. Fast units would move faster, slow units slower. The player would have very limited control in this phase, as the world would just glide forward in a realtime mode. This means that the player would give some fairly general orders to his units in the planning phase – what to do if they encounter an enemy units, whether his units should chase the enemy if it flees etc.

This preplanned turn system not only means more equal conditions for all players – for instance there
would be no more huge advantages from starting a war – it also makes it possible to have many more players in the game, compared to old fashioned turns, since all players do their moves simultaneously and not one after another. The AIs could plan their moves simultaneously with the player, which would mean less waiting for your turn time. The same thing in huge multiplayer games.

Map

We are using our own map. It will have hexagonal shaped tiles, since they provide more realistic movement than squares. The hex width would be 20-50 km. On an earth sized map this would give about 1.4 million to 340,000 hexes. To the player individual tiles would be of little concern since all the decisions are made on higher levels.

Different aspect of your civilization and people will be represented as layers on the map, and each layer can be turned on and off. For example, terrain will be one layer, region borders another and population density third.

Units and movement

One of our goals for the game is to reduce micromanagement. One of the most annoying things in Civ2 was to move sometimes hundreds of units one by one, one tile at a time. The fact that GGS is going to use maps that are way larger than anything seen in Civ2, and that we will try to make unit movement much larger and realistic means that the tile by tile movement is just not an option. In stead units are combined into armies, which would consist of several units. To make the player use the armies large groups of units would receive bonuses when fighting against small ones (again realistic). The armies would mean that the player would probably never have to move than 10 armies per turn, even in wartime. In peace periods it would be much less.

We are also making the whole movement of units mouse-based (like the rest of the game). All movement should be done via a simple left click/right click system, as seen in realtime strategy games.

The game will also have two types of movement ranges for units – deployment range and operating range. The deployment range is the area within which the unit could be moved to within one turn. The movement could include an attack on an enemy unit or something else, but would take the whole turn. The operating range will be the area in which the unit could just move around freely. If the army has sufficient scouting the whole acting range could be visible, and enemy units entering the range could be intercepted. So the unit could be set to guard the whole area within the operating range. Furthermore the unit could move around within the operating range freely. There would be a limit of course, but it would be high, and would often not be reached.

Regions

The regions in GGS will pretty much take the cities’ place in GGS as the basic economical and political entity. Regions is where resources are pooled, politics are done and things are built. It is also where revolts or riots will happen. The reason we introduce regions is partly to eliminate the unrealistic 21 tile resource system of Civ. In reality lots of cities can lie within a small area. In modern times cities lie right next to each other. To manage all the cities would not be much fun. That would mean managing 100s of cities and therefore way, way too much micromanagement. This is why we are using regions in stead. In modern times there would usually be lots of cities in each region. This drastically reduces the amount of entities that the player will have to manage. Even the largest civilization would never have more than 30-40 regions. Most larger civs will have only 5-20 regions.

The regions will be completely defined by the player, and be changed as the player sees fit. The main limit here would be, that a region could never be spanned on two landmasses divided by over 1 (or 2) hexes of water. There would also, however, be a max distance (distance would be in movement points, so hexes with difficult terrain like mountains would mean more “distance” than easy terrain ones like grassland. Furthermore transportation infrastructure would reduce the distance) a hex in the region could have to the region capital. Beyond this there would be a at first small, but increasing penalty the further away the hex is. This max distance would depend on the technology level. At the very beginning of a game a region may not be bigger than just 1, or a few hexes. In the end of the game a region could be several hundred hexes.

Each province would also need to have a certain percentage of its population as administrators. This percentage would go up the more provinces the civ has, but tech levels would reduce the amount needed. The capital region would need a larger percentage than the other provinces. If the percentage is lower there would be penalties.

Each region could have its own relationship with the mother civ. Many regions would be normal, integrated parts of the civ with the normal rights and obligations. But some could be more or less independent, could have its own military and control itself, and perhaps just have to pay some tax to the mother civ. There would be several degrees of this. Another version could be a colony, where the mother civ would have more direct control, and in a democracy would be given more free hands in regional matters there. On the other hand the people in such a region would not be very fond of this, and therefore a rather large military force would have to be garrisoned in the region just to avoid revolts. A third form would be occupied territory. This would mean that the “region” (not that it would be a real region) would give nearly no recourses to the player, and that the regions economy would be severely hurt. On top of this the people in occupied territory would most likely revolt at any chance they would get. So a very, very large military force would have to be present. Occupied territory would be for land that you have just conquered, and that you therefore have to give your military totally free hands to fight down any unrest. Usually occupied territory would only stay that way for a few turns.

Population

The goal of the game in this area is to make it as much like reality as possible. Therefore the game will have a real population, in stead of the “heads” from Civ. We are also going to use age groups, mortality, birth rates and other advanced stuff, to make it all work as the real world does. The mortality will be determined by the technology level, food per capita, and by the separately handled disease model. Birth rates will be determined by education levels, cultural things and more.

Population is modelled independently from regions, but regions are the primary way by which the player can control and interact with his people. Merely setting region borders is not sufficient to affect population properties, but imposing different taxation on different sides of the border or limiting migration will cause the population to differentiate. Thus, the population is not geographically uniform: people living in different parts of regions may have different properties. Just like in the real world, migration as well as spreading of ideas and wealth will tend to homogenize the population if there is no outside pressure.

Economy

Here, as with population, we are aiming for realism. This is why the whole economy runs on its own via an advanced system, where people demand goods depending on their income, and each good has a price calculated from supply and demand. The entire economy is to a large extend based on real economic theory. This also means that the player can have pretty much the amount of involvement he wants. If we doesn’t like to mingle around with it he can just set a tax rate to have it match his expanses. If he likes economy he can do pretty much as much as he wants, including a progressional tax rate, sales tax or tariffs. The players role in the economy will be that of the government in reality. The player can choose any economic system he wants, all with their unique effects.

Combat

The combat system will be both realistic and entertaining. To encourage people to use armies (which will be reducing micromanagement) large groups of units will get bonuses over small ones. Armies will, however, require consideration. Using different types of units will be required, on top of handling scouting, supply routes and more. And at the same time guerilla warfare has to be implemented as well.

Infrastructure and improvements

Improvements in GGS are built either on regional or national level. Unlike in Civ, the improvements will usually be on a very abstract and general level: the player orders to create higher education instead of universites and industry instead of a single factory. Some improvements may show on map, but the position where they will eventually be founded is not up to the player. As in real life, the citizens of the nation or the region will be the ones to build the improvements and it is possible for them to take the initiative in some cases. In fact, under capitalist economies most improvements would rise spontaneously and the player could only affect them indirectly through taxation and regulations. Another distinguishing feature in GGS improvements is that they are quantitative rather than simply present or not present. If twice as much resources are allocated to improve the scientific facilities in one region than in another, the region should conceivably produce twice as much scientific results, all other things being equal.

Infrastructure, like roads and railroads, should probably be handled in a similar way. Having to build roads on every single hex is not at all fun. So in stead this would be handled by allocating money for infrastructure on your budget (in a region or nationally), which will get it built automatically. Besides this you could order specific improvements built on specific hexes. So you could drag a road between your region capitals, or from a region capital to the sea, for easier, cheaper and faster unit movement and to increase interprovincial trade. Fortifications will also have to be player built, and would mostly be placed where there are cities, to protect them, or in lines at the border of a powerful and aggressive enemy.

Micromanaging improvements and infrastructure should be discouraged. After being built their upkeeping is automatic and depends on the general wealth, class structure or tech level of the people responsible for it. The purpose for having spontaneous improvements is also meant to relieve the player from building the same basic stuff to every region. For example, in dawn of industrial revolution factories would automatically start popping throughout the country, and after heavy wars people would start rebuilding their damaged infrastructure on their own.

Cities

Although cities wont have the economical and political importance it had in Civ, they are still being distinguished from the rural hexes. They have to be different, since they are where industrial production and trade takes place. Their population will also have to be handled specially, since cities until the 20th century actually had a constantly higher mortality than birth rate. So the city needed constant immigration to survive. And if the city stopped being the political and economical center people would stop migrating there, as shown in history. This should be modeled in the game as well. Cities would develop on themselves. The player could order a city built on a hex, but if people didn’t want to move to it it would die out. But on region capitals there would almost always be a city. If it wasn’t a city when the hex was selected capital the migration would almost always make sure it would become one.

The Rise And Fall Of The Great Powers

To make the game fun from beginning to end and to keep the historical accuracy high we want to implement the at Apolyton well known rise and fall idea. The concept is, that where in Civ a civ was large and powerful it would just stay that way. But this is in fact neither realistic nor very fun. History has shown that a civ once strong will always be surpassed by other civs at some point. And a game that is more or less decided by 1000 AD is much less fun than one that will stay thrilling from beginning to end. So in GGS civs will rise and fall. This also makes huge, long multiplayer games possible, where players can come and go, and the ones that was in the game first would not automatically end up being the most powerful ones.

The rise and fall idea is in fact a difficult one to implement. And we are not yet sure exactly how to do it. The key is to make game algorithms that makes an empire weaker, the larger it gets, just as there is algorithms that makes an empire stronger, the larger it gets. But these has to work in different areas of the game, and have different scopes. So when an empire grows it could get stronger in some areas and weaker in others.

The areas that civs will weaken in with increased size are probably going to be by needing more administrators, making this an increasing expanse, and thus hurting the empires economy, to require more troops around the civ to stay strong, thereby also hurting the economy, suffering from more corruption, making it less efficient and less flexible, and making it a victim of more nationalist strife, which should be huge in a very large civ. If the civ has based its size on mere conquest these weakening factors should be even more devastating.

So generally a civ won't fall bcause it is large, but because of several natural effects it becomes hard to upkeep such large civs.

Social Model And Nationalities

An important aspect of the game will be the relationship with the people. Some of this will be through the social model, which would handle different nationalities. The concept here is, that people are not genetic. Different nationalities exist, and this should be modeled in the game. So if you just conquer other civs without thought you would have a large amount of hostile foreigners in your civ, who would revolt at any chance they get, unless you treat them very, very nice, or you have an enormous military force to keep them down. In the long run an empire solely based on conquest would never last.

Another thing would be, that even a part of your own people could develop their own nationality. This would especially be so in a far away colony. In a longer run this too could create huge problems, and could very well force the player to give up the colony.

Religions

Besides the nationalities people would also have different religious affiliations. Having different religions in a region or in a civ may cause problems. The people would often not get along very well if they have fundamentally different beliefs, though some religions are more tolerant in this respect than others. On top of this religions may be governed globally, by a central religious center, that would control the religion class of that religion in all the countries in the world. This would give a religion with a lot of believers enormous power, and could let it virtually control a lot of civs. If you make political decisions that go against the religion's agenda you upset its believers. So if 90% of your population worship one religion you better not upset that religion. This could some times let the religion dictate your policy more or less.

Of course the player would also have some effect on religions. He could support or forbid certain religions, which would have some effect on their popularity. Or he could make a grand scale inquisition, by killing believers of a religion.

Religions would rise, fall and evolve like civs. It's possible for a religion to split into competing sects and for the agendas of religions change drastically due time. There would be stats for each religion, determining its ability to convince people to believe in it. This means that religions would start by a guy preaching in just one region, and afterwards it might die out immediately, or it could spread to surrounding regions via trade, migration etc. The player would not know the stats, so he could only try to figure them out from the religions success or lack thereof. A religion would also have cultural effects, that could in fact effect people’s preferences and other things.

Politics

You will not just rule your civ because you feel like it. Ruling means that you need the support of someone and you will not just do as you please. In stead you will be subjective to your people. To model this we have a system with social classes - workers, capitalists, nobility, intelligentsia, military etc. - that combat for power of the civ. Your task is to get the support from some, by being nice to them and doing what they want you to. Each class would have an agenda. The militaries would be many military units, the intelligentsia would be research, and humane policies, the capitalists would be free trade and as little a public sector as possible. So support those who fit your policy, and they will support you, and suppress the rest. Whether you run a military controlled police state or a friendly democracy is up to you - and the social classes. Cause you will not be in total control of what you choose. There will be revolutions, where you have to pick a side, or the class you base your power on may demand that you go to war with an enemy, even if you don’t want to! Refuse and you send your civ into a destructive civil war, accept and you will be getting a war that you may not be interested in. Even foreign powers could play the game in your civ. They could support a class suppressed by you, give it money and equipment to make it stronger and possibly revolt against you. Either just to annoy you and weaken you and let them conquer the leftovers, or they could have an agreement with the class, that if it gets to power they will have some influence on your civs policy.

The political thing should indeed be one of the most important and most fun aspects of GGS, and would be something quite unique to our game.

Diplomacy

Interaction with other players and countries is a crucial factor in successfully managing a civilization. A nation with little or no outside contacts will inevitably stagnate to death, and even the strongests civs cannot ignore rest of the world. In many civlike games the diplomatic agreements have been one of the most interesting and rewarding elements, and GGS will be no exception: on the contrary, since multiplayer game is our first priority there will be no compromises on the complexity and diversity of diplomatic means.

The diplomacy system will integrate chatting system for human players, a way to create binding contract between two or more parties and tools to manage these contracts. Chat is included to allow flexible negotiations between parties, and to give the players something to do while waiting for the turn to end. Contracts in this context refer to bilateral or multilateral agreements, which are basicly automated command scripts the execution of which is observed by all participants. Contracts may be broken, but the player is given a warning before he attempts that and the other contractors are notified of the violation. Furthermore, contracts are something which the people of all participating civs are aware of. The public image of untrustworthy civs will be tarnished for a long time, which makes future diplomatic relationships difficult to maintain. To ease the player's choices, the game will include many ready-made contracts such as treaties, trade pacts and financial loaning arrangements.

To manage contracts efficiently there will be many more options than in conventional civ-like games. The player may try to break them in secrecy, but if this is exposed (either by accident or outside intervention) out the citizens may revolt and reduce your authority. The possibilities and consequences of foul play depend on the government type: in dictatorship it is easier to keep the domestic population at bay, but the foreign nations will be more wary than they would of a democratic nation.

Diseases

Since diseases have played such an immense role in human history they too will be implemented. A disease would start out at a single hex, and would spread to nearby hexes or via trade routes. The rate at which it spreads, its mortality, its cureability and its incubation time will be determined by some stats, unique to the disease. This will give each disease its own life and characteristics. Diseases could be cured, depending on the technology level and on the diseases cureability level. When a disease has been on a hex for some time the people on it would achieve immunity towards it, which would mean that it would simply kill a very small percentage of the population per turn. If the disease then came in contact with people not previously exposed to it, however, it would become an epidemic once more.

(continued in second post)
[This message has been edited by TempLeland (edited February 16, 2001).]

**TempLeland** · February 14, 2001, 22:29

Technology

The human history is also history of technology. Throughout times, people have searched for better ways of doing things, and when that knowledge has slowly cumulated, we have reached current high level of knowledge and technology. When it comes to modeling technological advancement with a computer program, we encounter problems. The old civ2 system of technology tree has a great gameplay, but quite weak realism: to list some problems, the tech tree simplifies tech too much, classifying it into large, arbitrary chunks; also research is restricted to only one field at time, which is not realistic. Also, tech development is not so straighforward that it can be modeled as a tree of prerequisites, and there are also the differences between different fields of knowledge, the great difference between theoretical and practical work, and broad patterns of research work and single great inventions and innovations.

So, in the proposed game model we would have Science, which means theoretical knowledge, Technology, which means practical knowledge of applying the information in everyday use, and great innovations, termed Milestones. Science includes fields like Mathematics, Physics, Astronomy, etc. Technology includes fields like Construction, Shipbuilding, Metallurgy etc. Milestones include inventions like Wheel, Damasc Steel, and Automobile.

Milestones would be one-time packages, but scientific and technological fields would have many levels of knowledge or skill. The higher the level, the better the people excel at that field. This is to prevent the ridiculous idea of researching a complete package of mathematics in one blow.

Every level of science and technology and every milestone is concidered much like the advances of civ2. But, the development system would be different. Instead of creating a tree of the technology, we would have a network of influence between the different techs, and each advance could have quite large range of different prerequisites, of which not all would be required to get the advance. For example, in addition to possessing the knowledge of existing techs, it could be possible that an advance requires certain social conditions, certain amount of practical work, or some other event of game. Also it might be possible that in some conditions certain advancement could be gained randomly at certain probability.

The idea would be to create a rough environment for technological advancement, where technologies come roughly in correct conditions. In different kinds of worlds and in different games the progress would be different. If we do not use the time-line of our world, but instead always start counting the years from zero (like in smac) we can create an own history of technology in our worlds, that goes sufficiently realistically.

One main aspect of technology system is of course the spread of technology. One of the main ways for making technological progress should be to be in contact with other countries. This is what has happened in our world, and that's what is one of the main reasons for the success of Europeans; lots of countries competing with each other and in close contact with each other, so technology has spread rapidly. Tech would spread by trade, but also by espionage .

Disasters

Human history is not a walk in the park. Dangers lurk around every corner and unpredictable events may toss a thriving civilization into oblivion in a matter of few centuries. In Civ the winner of the game was usually settled well before modern times, and that made the end game rather boring. GGS will try to avoid this by thowing a multitude of natural, social and political disasters into play.

Natural disasters include crop failures, earthquakes, floods, disease outbreaks and toxic waste, among others. They depend mostly on terrain and the extraction of resources thereof. Social disasters are events which trigger social unrest, revolutions and riots being good examples. Political disasters are results of risky governing. Committing atrocities and getting caught is n instance of a political disaster. All disasters have two characteristics in common: they can happen only if certain conditions are met and they are more or less random. A region with volcanic activity will have a volcano sooner or later, and having bad living conditions for workers is a recipe for riots. As in real history, mere initial conditions do not determine whether a disaster will occur or not: there is always a probability that you get away with annihilating a social group that's been a thorn in your side, or that your crops don't fail in the middle of an important military campaign.

The balance between randomness and predictability should be set so that the player cannot just lie back and see his civ grow on its own, but on the other hand there should be enough room for long term strategic decisions. In multiplayer game, disasters can also be used to give new players an edge so that they won't immediately be wiped out by their elders.

Scenarios

Every turn in the game will be one year. This means that a full game, from prehistoric times to science fiction, will take a very long time. And via the Internet this could mean having huge, online games lasting for months and months, where players come and go, civs rise and fall. History is made. But not all players will want to play this long. Therefore it should be possible to play shorter games about certain historical periods. Imagine jumping into the world of 1938. War is virtually inevitable. But how will it go? It will be up to you. All eras and events could be played - the age of the European expansion, the American civil war, the Roman empire. The possibilities are endless, making GGS several games in one. When this is said it also have to at one point become possible to play a full game, from prehistory to science fiction, in a shorter amount of time. We are not yet sure how to do this, but the game should be as flexible as possible. So hopefully it will be possible to play a full game, taking only general decisions, in just some hours. Or some days. Or months. Depending on the player's preferences.

**amjayee** · February 15, 2001, 05:07

I have been a little inactive lately, due to some schoolwork I have had to finish. but now that has been don, and I can contribute more to this work. I think I will finish the region and population models soon. But first I will comment some posts.

I really liked this design doc. We have come through a difficult road, trying to create a unique game system on the internet with many kinds of people involved, not always able to contribute as much as they would have wanted to. But I think we have succeeded well, I think. This game system sounded really good when I read it, I don't think anything needs to be changed very radically. I especially liked the notion made in the disasters section; for disasters to occur, some conditions are required, usually known to player to be risky, and then some randomness. So with careful consideration, the player can avoid most disasters or at least minimize the risk of terrible catastrophes. These kinds of things make for a great gameplay even with lots of realism.

So, I vote yes for publishing the document - it is not entirely finished, and we perhaps need to clarify some things more, but in my opinion nothing is terribly wrong in it. You have done good job with it, Leland! And it's fine for me if you want to be responsible for the document.

**ElmoTheElk** · February 15, 2001, 07:01

I agree with this doc obviously, because I made the changes I wanted already.
I didn't let the disasters section out bacause I don't like it. I just forgot...

As Leland already said it's already on the web site, but I definatly vote to make this version 1.0.

**korn469** · February 15, 2001, 15:14

Game Philosophy

i agree

Role of the player

i agree

turn system

i disagree...i think that preplanned turns would quickly fail

1. when a player had a large number of units turn planning would be inefficent and the preplanned action phase would have to be relatively short so that chains of actions could actually work...this would increase the length of the game becuase more turns would be needed, also this would mean relatively long periods of time waiting on other players, and doing nothing

2. units would be heavily reliant on the AI, even in the pre alpha stages of the game, basically the player would have to give a lot of discression to his AI commanders

3. it would give the player a feeling that the AI wassing the show

here is what i propose:

a continuous turn system like in fallout:tactics with the following modifications

1. turns generate slower (this is an empire building game and not a squad based tactical combat game)

2. units can have order queues this would allow for a player to coordinante actions...you give your units the orders which do not go into effect until you you tell them to start and suddenly you can launch simultaneous attacks and such...

map

i agree

units and movement

not sure about this one

regions

i agree

population

i agree, except that we should only include traits thats have a direct impact on the game in the list of population traits

economy

i agree

combat

i agree

infrastructure and improvements

i agree

cities

i agree

rise and fall

i agree completely

social model and nationality

i agree

religion

i agree...we could possibly add some EU ideas in here

poitics

i agree

diplomacy

i agree

diseases

i agree

technology

i agree for the most part...i can't really say before i see exactly how we will implement it

disaster

i agree

scenarios

i agree

as with amjayee i haven't been able to contribute as much as i have wanted to, but this game has really came a long way...it's almost to the stage where everyone can see the same game in their imagination

well besides some a few changes and a couple tweaks i think that this is a fitting place to start the design at

korn469

**chrispie** · February 16, 2001, 16:16

I read this doc today in work (I got bored ) and I just had a quick glance at it now..it all looks fine to me. I'm happy with most of it, though the idea of player defined regions will need a lot of thinking about if we want them to be just definable for all the obvious reasons.

However, is there more on combat? That section seemed very short and didn't descibe anything about how combat will take place or the stats that will be involved.

Also, I'm not at all sure about the idea of all units moving at the same time in 'execution' mode, I think this could put too big a strain on a system and the net connection to have possibly hundreds of units moving all at once.

As for infastructure, I -really- don't want to see roads and railroads built for you, I want the players to place the roads (it's really not that bad!), and to have armies gain bonuses for moving along them. Throughout history, governments have placed infastructure like roads, mines, etc. I guess minor roads could 'grow' but to be honest, I'm not sure if they're needed graphically at all, I want to see the game keep much of it's 'green' look throughout and not be spoilt with roads on every tile.

"The key is to make game algorithms that makes an empire weaker, the larger it gets" Sorry to say that I must disagree on this, my view is that we need algorithms to make a larger empire more difficult to maintain, but not necessarly weaker. But more difficult I mean it needs more administration for it's regions - but most importantly, better systems for managing places that are remote and far from the capital. Maintaining an empire is all about keeping the people in the knowledge of who is in charge, and to whom they owe their aliegence, which is possible even with an empire the size of asia.

There, I think I'm done

**TempLeland** · February 16, 2001, 22:58

Thanks for your compliments, but don't forget that Joker wrote most of this, I just put it together and posted it. I hope he doesn't mind me hijacking the doc. In any case, I am happy to see that there these topics can still spur conversation. I may not be an expert on all of the issues raised, but at least I can try give my thoughts.

Korn:

quote:

1. when a player had a large number of units turn planning would be inefficent and the preplanned action phase would have to be relatively short so that chains of actions could actually work...this would increase the length of the game becuase more turns would be needed, also this would mean relatively long periods of time waiting on other players, and doing nothing

The way I see it, the preplanned commands would be kept on a fairly general level quite intentionally. The same applies to the number of armies the player controls. So, there would be no "large number of units" and there will be no "chains of action", if that means having players act and react to each others moves.

As for players waiting, I believe it has to be solved with a time limit. In open games there may be a mixed group of tourists and hard-core players involved, so the game cannot allow one player to stall the whole system. But if there is a time limit many people will probably make the best use of it, thus rendering all turns more or less equally long.

quote:

2. units would be heavily reliant on the AI, even in the pre alpha stages of the game, basically the player would have to give a lot of discression to his AI commanders

3. it would give the player a feeling that the AI wassing the show

During early development, I think that the AI will be equivalent to tossing a coin. I think that combat really needs to be designed so that the emphasis is on strategic, not tactical level. But that maybe just because I myself dislike combat, so new suggestions are of course welcome.

So, basicly I'm proposing that the combat decisions would be much more general than in Civ, and this would also apply to the combat algorithms. So, and AI wouldn't have to simulate a battle of hundreds of units but would rather work on a statistical and higher level of abstraction.

quote:

here is what i propose:

a continuous turn system like in fallout:tactics with the following modifications

1. turns generate slower (this is an empire building game and not a squad based tactical combat game)

2. units can have order queues this would allow for a player to coordinante actions...you give your units the orders which do not go into effect until you you tell them to start and suddenly you can launch simultaneous attacks and such...

I haven't played Fallout myself, so I'm eager to know the details of this system. Nevertheless GGS should incorporate command queues (or maybe even command trees), which will span several turns and contain conditional commands which the AI can use to resolve some conflicts. In design doc there is nothing about queues yet, maybe there should be.

quote:

units and movement

not sure about this one

Care to elaborate?

I reread it myself, and I find it acceptable except maybe for the number of armies (should it be higher?), a more detailed explanation between units and armies (a distinction which I haven't understood entirely) and the limit on operational movement which was mentioned in the last sentence.

quote:

population

i agree, except that we should only include traits thats have a direct impact on the game in the list of population traits

The problem is, I don't think we know what those traits are yet. So I guess we'll have to settle for the current population chapter for now, and maybe update the model later if necessary. Or should we simply not mention any population properties in this document?

quote:

religion

i agree...we could possibly add some EU ideas in here

EU? European Union? What does that have to do with religions? Anyway, I think I'll start working on the religion model once Pop and Government are looking better, so I'm most interested to hear any suggestions.

quote:

technology

i agree for the most part...i can't really say before i see exactly how we will implement it

Me neither. But the section looks general enough to allow plenty of room for design changes and improvements, so I think it suffices at this point.

Chrispie:

quote:

I'm happy with most of it, though the idea of player defined regions will need a lot of thinking about if we want them to be just definable for all the obvious reasons.

Because regions will roughly take the role cities have in Civ clones, defining regions is equivalent to choosing a city location. But regions are more dynamic, they change and evolve as time progresses. I am not a strong advocate of having the regions fully definable by the player, but he should have considerable power nevertheless.

As an alternative to fully definable regions would be that the player has a largish selection of small regions of which he groups the actual governmental regions. Is this better? Should the design doc take a stand on this issue?

quote:

However, is there more on combat? That section seemed very short and didn't descibe anything about how combat will take place or the stats that will be involved.

I have no idea. I recall asking a similar question in the old design doc discussion thread, and I would like to see the combat designed with a little bit more detail. Especially I want to distinction between units and armies clarified, does anyone have any thought on that?

quote:

Also, I'm not at all sure about the idea of all units moving at the same time in 'execution' mode, I think this could put too big a strain on a system and the net connection to have possibly hundreds of units moving all at once.

I think the strain on the network will not be insurmountable. At least not per player. There maybe some problems arising from all the players sending their commands at once, this might flood the server. Maybe the sending can be distributed over time a little bit?

But number of units is not a problem, because it will not be that high and the commands will be on a general level. That is, "move to point X" instead of "move left, left, left, up, left, right, fortify". As far as network connection is concerned, I worry much more about all the data which is used to update the client side representation of the world. Some heavy compression schemes must be devised.

quote:

As for infastructure, I -really- don't want to see roads and railroads built for you, I want the players to place the roads (it's really not that bad!), and to have armies gain bonuses for moving along them. Throughout history, governments have placed infastructure like roads, mines, etc. I guess minor roads could 'grow' but to be honest, I'm not sure if they're needed graphically at all, I want to see the game keep much of it's 'green' look throughout and not be spoilt with roads on every tile.

Well, there is a mention that player can decide where to build roads by dragging two places together. The idea of partially automatizing road and infrastructure building is that not everything is decided by the government. People may want to build roads of their own, and especially after wars the rebuilding of infrastructure should be automatic rather than up to the player.

quote:

"The key is to make game algorithms that makes an empire weaker, the larger it gets" Sorry to say that I must disagree on this, my view is that we need algorithms to make a larger empire more difficult to maintain, but not necessarly weaker. But more difficult I mean it needs more administration for it's regions - but most importantly, better systems for managing places that are remote and far from the capital. Maintaining an empire is all about keeping the people in the knowledge of who is in charge, and to whom they owe their aliegence, which is possible even with an empire the size of asia.

A good point. The Fall and rise section is maybe too vague on this, though it does mention maintenance.

* * *

To summarize: we need to think over regions, turn system, units, armies, movement and combat. Also it should be checked if population and rise and fall need to be defined more unambigously or explicitly. Otherwise, I see no reason for change.

Leland

**The Joker** · February 18, 2001, 15:19

I really like that design doc. You (Leland) are obviously more advanced in the use of the British language than I am, so the smaller changes you made are all welcomed. Great job!

And since I lack time I don't mind at all you hijacking my doc.

Korn:

I agree with Leland on the preplanned turns issue. I think it is a fantastic way to not only increase realism and make the game "new" and fresh, but also to allow more players in the game. If we use oldfashioned turns we are already limiting ourselves to 10 players or less in a game. Something very, very bad. Preplanned turns can support as many players as we want to put in it.

Chris:

Again the action part of the turns will mostly be on one server (which could be a player pc), and then the stats will just be sent to all players afterwards. It really does not have to be very advanced nor hard to do.

Combat:
Yes, there should propably be more about it. But since I am anything but a combat freak could someone else please write it? Personally I would be satisfied with a simple combat model, although I know most people wouldn't. So yes, we should write more about it, only I shouldn't.

Infrastructure:
I agree. As Amjayee said, to avoid making the game textbased there should be some drawing things to the map. And roads are excellent for this. So is some other things, like mines. But we should limit it. And many things (like education) should be handled solely at a province level with budget balancing.

Rise and fall:
Actually I think a large empire should be weaker than a small one in some areas. Yes, harder to maintain. But isn't that the same as weaker? It would be weaker to civil unrest, weaker in efficiency areas and so forth. I don't think we disagree on anything really, so it's not a big thing. I just wanted to point it out.

------------------
"If you are to hurt someone you better do it so good that you don't have to fear revenge."
- Machiavelli

GGS Website

**korn469** · February 19, 2001, 00:18

Joker, Leland, et al

i still do not think that preplanned turns are going to work for a multiplayer game because the orders are going to have to be so vague that basically it will either be the players watching the ai run the show or it will turn after turn of wasted and inefficent action

SMAC like simultaneous turns or fallout tactics like continous turns are going to be what a multiplayer game needs...not preplanned turns

to plan a turn is going to take time...time when everyone is setting there (yes planning their turns but still setting there) and some people are going to need longer to plan their turns, if these players aren't getting enough time they are going to quickly give up the game...diplomacy is going to take time...putting an effective plan into action is going to take time...so i see each turn having a 4-10 minute waiting period when the players are planning their turn

after this when the plans are being carried out i think most of the time it is going to just be chaos unless the AI is REALLY clever

the civ2 chain gang turn system won't work if the multiplayer games are going to be really gigantic but i don't think that preplanned turns will either...it is going to take a realtime hybrid to excel here

but that is my opinion

korn469

**amjayee** · February 19, 2001, 07:26

Turn system: I'm aware of the problems in choosing the turn system we are going to use. Civ2 system (traditional pure turn-based) will not work due to long waiting times; we couldn't have very many people in one game, and even with 8 players the game would be slow. SMAC simultaneous turns also has problems; it will allow too much advantage for the warmongers, and would be in times a little chaotic. Pre-planned turns has the problems korn mentioned - but I think they would not be so severe. Unit number will not be as high as earlier, due to armies, and because war will be more like strategical commanding, the limited level of control for player is not a very large problem. And besides, it doesn't need to be a problem at all; we would make it a game feature. I think strategical commanding is the most fun part of combat, and would be quite enough for the scope of the game. Also this would make it more clear, that combat will not be the main thing in this game, unlike in civ and smac. About the Fallout : Tactics system; is it similar to XCOM - Apocalypse system? if it is, the problem would be, that we would anyway need a planning phase for managing the empire. We could of course make a "real-time" system for moving the units similar to that f:t and xcom, but making it work without massive chaos would require allowing the players to pause the game, which would turn annoying, or make the units move really slowly and disallowing pausing. Either way has its problems.

There was some analysis of the possible systems. True real-time is out - this is not C&C, we need also to manage the empire. And also the scope would be much larger; you could be waging a World War all around the world, and handling that in real-time is impossible, or at least chaotic. So what I think is, that we need a planning phase, where the game is in pause mode and could have a time limit. In this phase, at least empire management could be handled in peace. Also units would be given orders. Then we would need another phase, which would resemble real-time, as proposed earlier... to allow the player more control, we could change the original system in the way that the player could change the orders of the units, but the game would not be paused. Pausing just wouldn't work in MM, perhaps in single player it could be allowed at least on lower skill levels...

All in all, I'm in favor of pre-planned turns; taking into consideration the game system we are going to use, I think it would work the best. Unit numbers and combat wouldn't cause problems, I'm sure. The system just needs to be designed carefully, so the amount of data doesn't create too much load for the network.

About AI, making the combat system work needs lots of it, but that kind of simple decision-making and orders-following AI is easy to do. Making AI for multiplayer (AI assisting the players) will not be tough. Making the AI to replace the human players will be tougher.

Roads etc:

I think there would be smaller roads and then larger, better roads. Perhaps only the largest roads (like the Roman roads) would be shown on map, and in addition to that there could be a "transportation level" either for the tiles or for the whole region, which would increase the movement?
[This message has been edited by amjayee (edited February 19, 2001).]

**TempLeland** · February 20, 2001, 16:46

I spent some time going through the old combat-related thread and frankly, I had the impression that some of those ideas are not very up to date. The combat stuff is based on a Civ-like game, whereas GGS and Civ are beginning to have as little in common as Space Invaders and Pacman. Also, I do not entirely agree with amjayee's turn order suggestion:

quote:

Originally posted by amjayee on 02-19-2001 06:26 AM
So what I think is, that we need a planning phase, where the game is in pause mode and could have a time limit. In this phase, at least empire management could be handled in peace. Also units would be given orders. Then we would need another phase, which would resemble real-time, as proposed earlier... to allow the player more control, we could change the original system in the way that the player could change the orders of the units, but the game would not be paused. Pausing just wouldn't work in MM, perhaps in single player it could be allowed at least on lower skill levels...

Two-phased turns are not a good idea for the following reasons:

1. It will be boring. If a player doesn't have any armies or units in combat situations, he really doesn't have anything else to do but to watch the units glide very slowly. I for one would find that frustrating, because I'd like to see the results of the turn rather than my commands carried out in slo mo.

2. It will be confusing. I think it is much simpler to have only two alternating phases (plan next turn, see the results) instead of three (plan, real-time, see the results).

3. The game does not have that much tactics. The way I see it, there is far less stuff to do with units/armies than there is on Civ2 or SMAC. There are no diplomatic units, no caravans, no settlers or formers. At least they won't act as traditional units.

4. Conflict resolution should be as automatic as possible. In PBeM games, for instance, there won't be humans around to oversee the "gliding real-time" mode. So, either in this case a lousy AI would have to be used, or the combat mechanics would have to be simple enough for an (elementary) AI to handle well enough. The former is frustrating to humans, the latter makes the whole real-time thingy boring.

I am totally advocating preplanned turns, but not any real-time extensions to it with humans doing the conflict resolution. I believe that players should just analyze the events of previous turn, do the diplomacy, give their orders (or more precicely, refine the old orders) and end the turn by sending new commands to the server. Upon receiving the turns from all players (or when the time limit expires) the server should make the moves that do not conflict with each other, and in case there is a conflict appropriate and simple algorithms will be used to resolve the situation. I don't want AI "wassing the show" any more than korn469 does, but I don't think it is necessary if the orders are kept on strategic and not tactical level.

I think wars will be more about production, diplomacy and economics than tactics. A typical war might take only a few turns, but the preparation and aftermath (that is, assimilating conquered areas, making treaties, rebuilding, etc.) will consume a lot more time. But this is just my opinion, and I am kind of biased because combat is not my favourite pastime.

And about turn duration: 5-10 minutes seems like the minimum, or so I've assumed most of the time. I wouldn't be surprised if 30 minute to 24 hour turns would become commonplace, though that will make the games incredibly long.

Oh, does anyone has any idea what is the exact relationship between units and armies? It seems that units have been ditched as individual entities, is that so? Are they just properties of armies similarly to weapons and armor being properties of units in SMAC? Or should units be defined in broader terms so that tile improvements (forts, mines, etc.), foreign embassies or government agencies are also units (just an idle thought... shoot it down! )?

Leland

**chrispie** · February 20, 2001, 19:02

Although I agree with Leland on the idea of just seeing the results of unit movements, I still feel we need at least some of the game to be about unit movement tactics. If we take that out too much, we turn it into an economic simulator (hears yawn all around ).

Just to throw in something somewhat unreleated but a bit connected, I had an idea that 2 ppl can run the same empire - one of them runs the economy and the other the military. I think that'd make for a brilliant side game, basing it on the co-operation of those 2 ppl. Just an idea.

**TempLeland** · February 20, 2001, 20:10

Ok, how about the ability to give each army a preferred tactic out of a predefined set? Maybe these tactics could be a branch of military technologies (albeit a small one). So, you could tell your army to do protect the civilians while attacking, or to minimize own casualties or dig trenches and stay put. Essentially these "tactics" would be orders given to the armies, so the game would still not be very tactical, but it would allow more variance to combat.

Nevertheless, the way I envision strategic combat is that in addition to economy (which is a bore, I agree) the player has to choose when and where to attack. Should he take over the enemy colonies first, or head straight for the mainland? Is it better to attack by sea or by land? Is it better to decrease morale by bombing civilians, or should only ilitary targets be chosen? As for defense, one has to notice potential weak spots, be they geographical or political, and guess in advance where the enemy will cross the border. I don't think this will give quite as much things to do as moving individual units would, but I also don't see it as plain economy and slide bars. The player would get to draw arrows on the map, at least.

Two people playing the same civ is not an impossible idea, if all the player does is translated to commands. It's not difficult to implement on server side (because all that's needed is the capability to accept commands from more than one sources, and to send the appropriate updates to more than one places), a smart UI may be more difficult. So, it's not inconceivable to have such mode sooner or later, it's entirely different matter if people would be interested. (Especially if there is not much tactical combat.)

The diplomacy screens should incorporate some sort of common strategic map in which allies could plan their moves collaboratively. So even if the game wouldn't support distinguishing military and economy, two players could make an agreement to leave the military control totally to another, in exchange for economic security or something. The possibilities are endless!

Leland

**The Joker** · February 24, 2001, 01:23

How great that I find myself agreeing with everything Leland is saying.

Korn, could you describe your proposed system, please? I have not played any of those games, and I am not sure how it would work.

Generally, like I said, I agree with Leland. I think two phases as we initially agreed upon would be best. And I think his ideas on different army tactics sounds great.

Furthermore I agree with him that war should be other than just moving units. Finding enemy weak spots, including his sources for important raw materials (intelligence would help you find out where those were) and attacking them should be truly important. Red Alert style stacking a lot of tanks and marching towards the enemy capital should be pathetically ineffective against a well managed military with good defence at important places, a good production and political stability.

We should include as many things in the concept of war as possible. Diplomacy (including the secret part of negotiating with your enemy's enemies or dissatisfied social classes in his civ), politics, economics, strategy etc should all be a part of it. Moving units would be a small part of it.

And just to define armies and units:

Army:
Everything that moves on it's own. An army can be of 1 unit or 100 units. When you move your newly built unit from the region where it was constructed to the army it is to join it is itself an army when underway. There will be advantages to making armies consist of many units in stead of few (under most circumstances). An army needs supplies. This will propably be handled automatically when it is within your own territory, but outside of it you would need a special unit to handle it.

Unit:
A military (or other) entity. Units are the building blocks of armies. And although a unit/several units can be taken out of armies whenever you want there would, as I said, be advantages to keeping them stacked in armies. A unit can operate completely on it's own, and is built on it's own. So you build units but move armies.

------------------
"If you are to hurt someone you better do it so good that you don't have to fear revenge."
- Machiavelli

GGS Website

Announcement

Design Doc 0.2 discussion and vote

Design Doc 0.2 discussion and vote

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment