Military Model III

[Lord God Jinnai]

Advancing rapidly should cost something.

[Mark_Everson]

Hi LGJ:

Advancing rapidly does of course cost something. You are probably forgetting that there is a terrain-based cost associated with any movement (that stuff is in the previous post I referred to). The total tick points for movement will involve the terrain cost (usually something like four ticks or larger depending on the terrain) added to that for movement speed and caution. So advancing rapidly over clear terrain with no roads would cost 4+ 0 = 4 ticks whereas advancing cautiously over the same terrain would cost 5 ticks.

Hope this clears up the problem...

[Lord God Jinnai]

Is there a differance here between rapid advancemnt and reckless advancement. Rapid being atleast worrying about the possibilty of ambushes, etc. but still moving quite fast and reckless being no heed at all for extreme circumstances (like you know you're army is needed at the capital yesterday cuz its about to fall and you would normally be too far away to help.

[Richard Bruns]

I have some extra time, so I have looked over the military model again. I will continue to try to fit in the province based combat option in a way that will please everyone:

Could we create a situation where battles are based on provinces but the player has the option to micromanage the battle within the province?

For example, suppose I, the Romans, want to take Gaul, a province that is essentially a collection of native villiages. The economic turns are currently set to 5 years.

I don't want to mess with units in the map. I want to send a big ol' task force into the place and let them take the province by themselves. I to press the "End Turn" button, let five years of campaigning go by, and get the province on the next turn.

But many people want to micromanage the battle. So can we let them "zoom in" on that province battle and let them manage the tactical side of the battle? A battlemap would be drawn based on the province map and surrounding areas, perhaps zoomed in 4x so each world square created four battle squares.

This would create a mini-wargame within the overall Clash game. The player would direct the campaign in detail, taking settlements one by one and manipulating the core units individually. The wargame would run in one month turns, so the player would have 60 turns to conquer the province. After these 60 battle turns, the game applies the results of the battle to the rest of the game.

This approach was used in Lords of the Realm, and it worked well. The player could choose to micromanage the battle, or just let a general do the job and auto-calculate the battle. If a player wanted to get into the military, they could do so in detail. But if they just wanted to manage the kingdom, they didn't have to worry about the fighting.

Another advantage of this is that it could solve the problem of the boundary between economic and military turns. If the battles were handled last, after all of the economic decisions were made, the economic scale could run at the same time scale and provide reinforcements at the proper time.

Here is the timing I propose for the game turns. All phases are simultaneous, and players cannot enter a phase until all players have finished the previous phase.

1) Battle Announcement Phase. All conflicts are planned and all existing troops are given orders. Players are notified of enemy actions directed against them.

2) Economic Planning Phase. Here, all civ management is taken care of. This is the primary game phase for non-wargamers. Plans are laid out for the time until the next turn. Provinces can be ordered to build military or militia units and send them to the battle sites determined in Phase 1.

3) Battle Resolution Phase. After all players exit Phase 2, all of the battles commence. If battles are calculated automatically, players watch as the AI caculates the conflicts in all of the provinces. If the battle is manual, the battlemaps are drawn.

If only adjacent provinces are being fought over, the game generates one big, zoomed in battlemap of teh combat area. If battles are occuring all over, units are moved on the main game map. Players then fight on the map as the economic model runs at the same time rate. So if you ordered the construction of city walls during Phase 2 and it takes three years to complete them, the enemy tactical commander would have 36 turns to take the city before the walls were finished.

Also, reinforcements can arrive on the battlefields as the economic model makes them. If it takes a year to train a new force and you order continuous construction, reinforcements will appear on turns 13, 25, 37, and 49. The fifth force would appear at the beginning of the next economic turn, so it would be available on turn 1 of the next battle.

So if you were the Guals, you would have time to react. You would be notified of the invasion in Phase 1 and would order economic defensive measures during Phase 2. During Phase 3, your economic buildup results in fortresses and new units appearing on the battlefield. Almost five years of production can be used in your defense, and it all occurs at the proper time.

I think this system or something like it could solve the problems of micromanagement and interface between the two time periods. Players could choose the level of management they want, and all of the game events would happen on the same timeline. Another benefit is that modern battles could be more detailed if the player wanted them to be. If the economic turns were one year and the military turns were one week, the same game dynamic would apply and the battles could be more detailed and realistic.

Can we work out a way to make this an option? I really think it would help the game.

[Lord God Jinnai]

Rich:
I am of the opinion that it wouldn't work for multiplayer games and we should try to make the single/multi-player differances as little as possible.

And even though i probably won't do too much detail with the military model i don't like the idea of campaigns to conquer entire provinces only instead of strategic points of land which is much more common or conquering more land and giving some back in exchange for other settlement.

[Richard Bruns]

Lords of the Realm had multiplayer. If you were attacking a neutral territory the battle was calculated automatically but the players controlled the battles when they fought each other. That would be a good standard.

I don't think that the battles would take that long. Fifteen seconds per battle turn would be a good limit, and it would mean that the battle would take 15 minutes at most. That's not too much time considering that waiting for everyone else to move units around in one civ2 turn can take about that long.

If someone was left out of the battle phase, we could give then the chance to look over their civ closely and spend extra time managing it. You have to do something like this every once in a while anyway, and I figure that the normal turn time limit would prevent a good study of your civ's condition. This would mean that players who don't fight as much will have more time to manage their civ, which is an accurate and fair result of the system.

You don't have to conquer the whole province. You can conquer whatever portion you want, and when you exit the battle screen your gains are transferred to the normal game map. The map would also have some areas outside of the province in question. So the province you order an attack on is the main focus, but you can go outside of that province a little or just take a part of it. Whenever the campaign is finished, it adjusts province borders or splits the province as nesessary.

But if you don't want to micromanage, your generals try to take the whole province. If they fail, they either retreat or the province gets split.
[This message has been edited by Richard Bruns (edited September 12, 2000).]

[Lord God Jinnai]

See that's the problem....i might not want to micromanage anything except what land is taken FE. I many not want to waste my forces on the entire province which i might know would spell distruction for my civ, but want specific areas, but don't want to get too meshed up in the details. This also doesn't allow for mass conquesters like those the mongols did where you might override several provinces by perhaps cutting paths through them first.

[Richard Bruns]

You don't have to take the whole province. You can go in, take the area you want, tell the troops to fortify, and then call for a truce or automate the ramainder of the battle. If you hold the area, it is added to your nearest province or it becomes a new province.

When the battle screen opens, it would include part of the province you are attacking from. All of your troops would be lined up in your civ's territory near the border, waiting for orders. You could leave 90% of them fortified and have the rest be a small strike force.

You can try to blitz through provinces. Just select all of the territory you think you can take, and the battlemap would include it. If the chunk of territory was large, the battlemap would be exactly like the main map and it would be like the normal system with smaller turns.

If you don't automate the battle, you are free to do exactly what you would do in the normal game. You just have the option of doing it in more detail, and the mil and econn timescales match up properly.

[Mark_Everson]

Richard:

In trying to sell your way of handling military matters at the province level you raise a bunch of objections to the current system. I believe all the objections you raise are easily taken care of by player-selected alternatives within the existing system! In fact, I have already made this speech before. But I will go down the list one more time...

But first, could I encourage you if you want to continue talking about this to start a new thread about this potential alternative model? It is such a radical departure from what at least I am trying to do with the military system that I think it confuses rather than helps the discussion by being here interspersed among discussion of the "default" system.

Now, on to your points and my responses.

You say your new system will "solve the problems of micromanagement". Because the player will be able to tell the AI "take this province using these troops". If you don't want to micromanage the military system there is absolutely no requirement to do so in the default system. So what exactly are you solving??? We have very carefully worked on the existing system, and will work on the AI, so that Any area the player isn't interested in can be turned over to the AI! So you can do (when the AI is sufficiently advanced) exactly what you claim is a great benefit of your military-provincial system in the Existing System. And in fact in a much more versatile way, because you can tell your AI general to take any arbitrary sized piece of land rather than a specific province or group of provinces. So, other than the fact that you personally really want Clash to be a lot more like Risk in the military area, I just don't get it.

You really don't like the difference in time scales between military and other parts of the game. I agree with you that it can potentially produce some bizarre effects. But if you want to play 6000 years worth of game at a monthly scale... well, you can do the math. It's too many turns! Your approach does not solve this at all IMO. It guarantees that if the player wants to continuously run military action that they will have to play many tens of thousands of turns. That is madness! And I guarantee it would kill Clash! (At least the whole-history game, your approach would of course work OK for a short scenario, but so will the default system with turns set to one month.) Now what you can do in the default system if you like, is to go ahead and set the timescale for the turns be one month. However, if a player were to do this they would of course need to trust the AI to do most of the work for a whole-history game, or they would never be able to finish even a single game. The existing system already allows the timescale to be set to one month for the turn length. Again, your system does not solve this issue in any way for the person who wants to play out all the battles. It simply makes the game impossible if you want to play all 6000 years of history.

As far as I can tell these were your two major points in this whole thing. I'm sorry if I have been a bit snippy in this post, but this idea seems to come up in a new incarnation every six weeks. I understand you don't want to move around a lot of military units, and the default system means you will Never have to if you use AI generals! And if you want you can play a 72,000 turn game. But personally I think relaxing realism a little bit to allow the game to actually be played by people who want to do different things than you is vastly more important than a rigid adherence to perfect real world time scales.

I just don't see the point in having two different systems when the second system, at least to my perspective, does not add one single thing to the game. YMMV.

[Richard Bruns]

In my proposed system, a battle turn takes much less time than an economic turn. The player is only moving a few units and the game is only calculating a few things, so the battle turns should take about 15 seconds. This means that five years of real-time campaigning will only take about 15 minutes.

Also, these battles will not occur every turn. If we make a game that allows players to constantly wage war on large scales, then we have failed to provide a good model of civilization.

Even the especially warlike Romans were only engaged in wars for about 400 years of their 1000 year history. This means that micromanaging every single one of the accurate-time battles for an agressive civ under my system would take about 20 hours. Given that a single civ2 turn can easily take 20 minutes in the modern era, this is not that much time.

The current system puts military action into every economic turn. Someone who does not trust troops to the AI would have more micromanaging to do, as the troops are modeled by unit rather than by province. They would have to manually move troops around within the civ and set up border guards and fortifications. If we assume that military movement takes three minutes per turn on average like it does in civ2, that means 10 hours is devoted to managing the military.

So the time difference is not that great. In addition to this, the current model only allows the Romans to undertake 200 months, or less than 17 years, of campaigning. This is IMO unacceptable. The Roman empire could not possibly have been built in 17 years of military action. I have not seen anything indicating how the military model would work around that. I think that this could easily lead to bizarre and game-breaking behavior.

And the new method is not a radical departure. The battle screens would use exactly the same military model that we have now. The only major differences are that moving units within the civ and setting up defenses are handled abstractly to save time and effort, and the player has an option to skip boring battles and concentrate on the more interesting ones. I think that there will be a few battles that even the most military minded player will want to skip.

By the way, what does "YMMV" mean?

[Frank80878]

Hey guys, I've just been reviewing the past month that I havent been checking the forums and I just wanted to discuss some things. I apologize if any of these topics are settled.

First I was wondering if anyone was wondering if the subject of soldiers morale was being discussed. Normally a soldier is content if he has a meal, a bed, and ammo for his rifle. But this is not always the case. One of my favorite examples of courage above and beynd the call of duty is the 101st airborne at Bastogne during the battle of the Buldge. Normally a division that is entirely surrounded, freezing cold, low on supplies and ammo would become33 discouraged to say the least. But these troopers wouldnt surrender, and eventually won, because they knew they were the best and because of the natural pride of being part of an elite unit. Anyway my point is that if there is a system for morale does the level of training have any influence on the formula.

Second I was reading over a discussion about rebuilding a decimated unit. Obviously building an entirely new unit should require more materials and men than just reconstucting an old one. BUt what I have to add is that the people who do survive would have an major influence on the new recruits. A battle-hardened NCO will make sure that he teaches his troops the skills that they need to know to survive in a combat zone since he's been there. So maybe a reconstructed unit should have a slight favorable modifier over a newly created one. And yes the REMF's most likely will survive so I cant understand how a unit will be entirely wiped out unless the uni is completely routed very quickly without giving the higher-ups time too retreat back to a safe zone.

Just some suggestions, its ok if you have to correct me if I missed some discussions I wont mind.

[Mark_Everson]

Richard:

YMMV = your mileage may vary, indicating the other person may have opposing viewpoints...

I am astounded by the unevenness of your presentation. The military turns in your system are supposed to take only 15 seconds, whereas those in the standard one somehow magically take three minutes, more than ten times longer! If you are going to pick numbers out of the air like that to make your case there is really no point in talking about this.... Perhaps this huge asymmetry is because you think micromanaging the economy is going to take a huge amount of time. If the economy is done on a monthly basis, really very little of interest to the player will happen much more frequently than on a yearly basis IMO. So that effect can't be responsible for this enormous difference. I could go on with paragraphs of details, of what you might or might not be thinking about this, but I only have a little time.

On a different subject, I did not mean to imply that every single turn will be taken up with fighting in either system. But as you yourself point out for the case of the Romans, fighting can occur in a very substantial fraction of all turns. So in a "worst-case" you could spend something approaching 50% of the time in war turns.

The way you want to do it, even using your own somewhat biased numbers, gives 20 hours of play micromanaging TFs for just the thousand years span of the Roman Empire. I couldn't have made my case better! So if you prorate that for the entire span of the game you have a 120 hour game. This is ridiculous! As I said before it will kill the game for anyone who is interested in military issues...

As I said before, someone who wants to have monthly turns on economic scale to match the military scale will simply need to trust the AI. And similarly, using your approach, I think it would be very difficult to play the military game for other than an occasional crucial battle. Otherwise both approaches would also take this enormous amount of time which I think is untenable for all but the most fanatically committed players.

Anyway, we are both firmly convinced that we are right. We have very different views of the nature of the strategic military action in this game (as you say there is no difference in tactical battle resolution). You can go ahead and keep throwing out reasons why your way is better, but I'm not going to respond any more unless I see something that really convinces me. I expect we will only get the final answer on this from playtesting, and perhaps not even then.... I of course have no objection to you trying to code this yourself if you care to, but it's going to be a lot of work.

I think that when gameplay collides with realism game play absolutely must win. I read your system as taking the opposite viewpoint. I just don't know what else to say about this. We could go on forever, and as far as I can judge right now it will not help the game progress one bit.


Frank:

Morale has only been discussed in a very general way. I agreed with you that level of training should certainly have an influence in the formula. We don't have a formula yet... Also, given equal levels of training, I think we should see things like well-paid and respected professional armies having the highest morale. After that, conscripted armies in wars with popular support would probably be next. Then starting morale would go all the way down to completely green enslaved troops in an unpopular war for an unpopular regime.

On rebuilding units, I don't really have much knowledge in that area, so anything that people familiar with the guts of military units can agree on would be fine by me.

Feel free to go ahead and suggest details for either of these. Although for the first you might need to look into the social and government models.

[Richard Bruns]

I did not pull the numbers out of thin air. They come from looking at existing games. In civ2, you had to manually move every unit from your central cities to the war front. Every little army had to be manually guided from the center of your civ to the war area. Then, every little unit fought an individual battle. If you were producing units in large enough quantities to fight a big war, this meant that you would spend a couple minutes every turn moving all of those individual units all over the map.

The 15 seconds in the battle scanario comes from experience with wargames like MAX. If you are only concentrating on a single area with a few units moving simultaneously, the game will use a lot less time. It is not magic, it simply comes from a change in scale and responsibility. I never said that an entire economic turn for the whole civ would take 15 seconds. I said that moving the troops in one theater of war would take 15 seconds.

I think that the province system would do a good job of automating the movement of troops within the civ and the building of border defenses. Each of these actions would currently take a couple minutes per turn when individual units are moved, but the province based modeling should cut that time down quite a bit.

The battle turns focus only on moving the troops in the area that is actually being fought over. Considering all of the border guards elsewhere, this would only occupy a small fraction of the units. So instead of managing every unit every turn, you manage a few of the units for more turns.

Realism is not what I am concerned about in the one-month rule. My concern is that it will be practically impossible to build empires. I have read that the military model is to have accurate movement, casualties, and results. This is a good goal. But under the current system, only 17 years of militay action happen in 1000 years. This means that the conquest of Italy alone would take more time than the entire rise and fall of the Roman Empire. The functionality of the player's military has effectively been divided by 60.

So the player currently has two options:

1) Accept a timescale that cripples the military and makes wars happen in bizarre ways.

2) Handle an entire economic turn every month.

I'm trying to compromise and find a middle ground between those two options. I'm trying to find a way to create a very fast, very limited "military only" turn that involves no economic calculations yet allows the military to be effective. I don't understand why I keep running into a stone wall.

I know that my idea has some flaws, but I don't see any other effort to deal with this. I know there are other alternatives that I haven't thought of. Could the player set the military turns to five years so the military moves faster? Could there be another way of dealing with the problem I'm trying to solve?

My goal is not to tell exactly how something is to be done. I never mean to hand down a decision, even if it sounds like that sometimes. My goal is to toss out new ideas, generate discussion, and work toward a common solution. I have to do my best to defend the idea so people can see what I am thinking, but that does not mean I am unwilling to change.

[Richard Bruns]

This was originally posted by Mark Everson on May 11, 1999 in this thread.

quote: I'm pretty much done sketching out the military system. There are still a huge number of details to resolve, but its getting there. I hope to have my proposal up by thursday or friday. Anyway, one thing that has been brought up by various people is the possiblity of a tactical mini-game for fighting out army vs. army conflicts. In the spirit of Clash this would only be used when the player actively wants it. To that end I am trying to design the strategic combat system so that it would be possible to add on a tactical system at a later date. I'd like to hear who would use the tactical one if we put it in there, and, if you'd use it, what kind of features you'd like to see in it. Right now I envision for a tactical system: simultaneous movement some fog of war (initial positioning also done partly blind) combat occurs for units in the same mini-square ranged weapons that fire x squares infantry and more mobile troops air units that can range the battlefield at will (unless fighting each other) simple morale system Any ideas about these proposals or anything else, fire away! -Mark

It seems to me that my recent proposal happens to be very similar to what Mark was talking about. In addition to solving the problem with the uneven timescales, it is similar to something that many people have expressed approval of. If the option of a detailed combat system was good then, why is it bad now?

By the way, I wasn't actively searching for something like this. I like to look over the old threads occasionally to see if there are any good ideas that have been forgotten. I happened to stumble across this one.
[This message has been edited by Richard Bruns (edited September 20, 2000).]

[Mark_Everson]

Richard:

How's this for a late reply!

I was talking about an optional system for fighting out battles in detail in a single square. That proposal did not in any way affect the basic game mechanics of TFs moving and fighting square-by-square. Your proposal is very different (IMO) in that it changes the high-level mechanics of the game. Its not simply an add-on, it would require different movement rules and AI among other things. Until we have the core game done, I think its not profitable to discuss options that are very time-consuming to code and test (both your "provincial" one, and the optional combat model).

The reason I think the optional battle system is bad now is for several reasons, one due to a quote I read from Sid M. He said something like "One good game is better than two great games"(!) Meaning, don't distract the player with things that aren't core to game, because they can 'lose the thread' of the overall game and so lose the just-one-more-turn factor.


All: (an attempt to bring some life to the mil thread!)

In another forum I commented on armies vs units being the main military actors in the game. I believe those thoughts might be of some use here. I've edited the comments so they'll read better here.

1. Historically more powerful armies win wars (duh!) but only by concentrating military power at the critical points. The only reason this doesn't work in Civ is because arbitrary rules, like only one unit in the stack defends, make 'real' armies untenable at least in the open. In a game without these rule distortions the more powerful force will generally win if it is concentrated. A more powerful force will lose if it is dispersed while its opponent concentrates, since each small army can be defeated in turn by the large army with very small losses. Of course, guerilla war is an exception, because the dispersed forces refuse set-piece battles.

2. Armies are pretty much the hands-down winners vs individual units running around. That's because of reduced micromanagement, and more historical gameplay. But there is one caution. You need a vaguely reasonable supply system or the 'reality' I speak of in item 1 will imbalance the game. First, supply rules will limit army size since only so much food can Get to a single square. Second, without supply limits or other restrictions on task force (TFs) movements the first really great army can sweep the board. What prevented this in history was geographic boundaries, supply limitations, and human concerns, like Alexander's troops forcing him to turn back from India.

Also, if the AI is concentrating into armies, a player preferring moving around lots of individual units (for example, small TFs) will be forced to do so by the inevitable results of losing all their single units in turn. At least if the military system is vaguely realistic.