Clash Scales

[axi]

Concerning the issues of Mass/Space/Time scales in Clash

Clash is meant to be a quite realistic Turn Based Strategy game, actually a low-resolution simulation of human history. It is also meant to be higly flexible, since experience has proved that in such games, customizability has led to longevity. Of course the player should not be confined to play each and every time in the earth globe and for the standard 6000 years of human history. There are 4 points to be addressed, concerning the scale of the game:

1) Clash should be able to simulate events in a global, continental, national and regional scale in either the real or a fantastic world.

2) It should be able to cover any timescale, from the 37 millenia of a fantastic world like Tolkien's Middle Earth, to the 6000 years of human history, to the 6 years of WW2, to the single day of the Battle of Waterloo.

3) The player should have a choice about the resolution of his game in both time and space, by choosing the total number of turns of the game and also the number of map tiles.

4) Because of the distortions the above choices will create, quantitative figures and growth rates in all the models have to be reshaped, in order to restore the proper density. This mechanism is indispensable, since it helps restore realism, while leaving the actual game machine intact.

The variables concerning scale, that should be decided by the player in the begining of a standard game, or by the creator of a scenario, should be as follows:

A) Space elements

1) Size of tile. 1 tile = AxA km.
Default (D): 100x100 km = 10000 sq.km area.
For smaller or bigger squares, multiply area with respectively a negative or a positive power of 2.
Proposed sizes for normal games:
1/16D___________625 sq.km______25x25 km (minimum)
1/8D___________1250 sq.km______35x35 km
1/4D___________2500 sq.km______50x50 km
1/2D___________5000 sq.km______71x71 km
D_____________10000 sq.km____100x100 km
2D____________20000 sq.km____141x141 km
4D____________40000 sq.km____200x200 km
8D____________80000 sq.km____283x283 km
16D__________160000 sq.km____400x400 km (maximum)
Proposed limits for scenarios: anything, as long as the creator endeavours to balance the models himself.

All economic figures concerning the tile are directly proportionate to the size of the tile (such as population, number of resource sites, production, trade income). All variables are computed with the default values and if they are not displayed abstractly (with heads, arrows, shields, beakers, etc), they are multiplied accordingly. As a result, the nominal sizes of the TFs are similarily adjusted (larger world - larger armies).

2) Map resolution. X tiles length, Y tiles width, number of tiles N=2XY (because of the diamond grid).
Default: 280x120 so N=67200 tiles. The default resolution derives from the default tilesize and the assumption of a spherical earth with an equator of 40000km and meridian of 20000km and with the exception of the polar areas. The length of a diamond tile (which is equal with it's width, although the isometric view makes it look larger) is approx 141,4km (100sqrt2).
Proposed resolutions: The figures are chosen in order to facilitate map and savegame transformations through the appropriate utility.
Ultra Gross__70x 30______4200 tiles
Gross_______105x 45______9450 tiles
Thick_______140x 60_____16800 tiles
Coarse______210x 90_____37800 tiles
Normal______280x120_____67200 tiles
Slender_____420x180____151200 tiles
Thin________560x240____268800 tiles
Fine________840x360____604800 tiles
Custom resolutions: Anything between 30x30 and 1000x1000, if you are patient enough to wait.

3) Size of the world Derives from the above two, assuming that the diagonal of a tile is 1,414A.
Default: As per above, 39598km x 16971km
Some combinations of the above figures, showing how the same world can be played in different resolutions:
Ultra Gross+1/16D=2475x1061km
Ultra Gross+1/4D=Thick+1/16D=4950x2122km
Ultra Gross+D=Thick+1/4D=Normal+1/16D=9900x4244km
Ultra Gross+16D=Thick+4D=Normal+D=Slender+1/4D=Fine+1/16D=39598x16971km
. . .
Fine+16D=475176x203647km

The above three (actually two) variables will be set through a single window, where selections will be made among size and resolution, while the deriving world size will also be displayed. In the same window also belongs the decision for the shape of the world, which can be:
1. flat, appropriate for sub-global maps
2. cylindrical, appropriate for global maps
3. toroidal, appropriate for cool sci-fi worlds
4. weird, maybe cylindrical, with north and south edges teleporting to the appropriate tiles in the 3 missing directions (simulating a passage over/under the polar cap), for better realism in modern conditions.
Other options such as a geodesic sphere based on a cube, or a fractional jump grid that could be considered, are problematic. The above options have the advantage of not messing with the map grid itself but only with the edges.

B) Time elements

According to what is said in the recently deceased military thread (may it rest in peace!), there is a need for a distinction between military and civil timescales, for the sake of realism. The predominating model seems to be 1 Civil Turn (year) = 12 Military Turns (months) = 120 ticks (3 days periods), which most of us find representing enough of at least the modern military reality. The complications of such a model though, in addition to some other major issues, need to be discussed extensively and decided upon soon enough. These are mainly the following:

• Is time in Clash going to be linear or exponential?
  Perhaps this decision is already taken, but in this point everything has to be reexamined.
  If exponential, it is easier for the "action" to be distributed fairly among the ages (civ2, though exponential, failed in this, because of the micromanagement). On the other hand, civil turns are unequal, so either military turns are unequal too (with ridiculous results in the antiquity), or there are more of them in earlier ages, which is equally hard to digest. Also, there are some ridiculous leaps when the time scale readjusts (Take the woeful example of CtP, using a factor of 10 for time AND population!).
  If linear, two problems emerge: the length of the game and the over/under-representation of early and late periods respectively (Civ2 on the other hand is accused by quite a lot of people for cruising through the pre-modern era too fast - which is the opposite effect). I believe though, that because of the micromanagment reducing mechanisms of Clash, the player would feel quite free to press the 10T or even the 100T button more often, assured that there will be a halt every time something worth mentioning happens (and since a Clash world, unlike civ, is NOT EMPTY in the beginning, but filled with cultures to be conquered and assimilated, there will be plenty to do, even in 3000BC).
  My personal opinion about this is that it should be the player's choice and below I propose a method for achieving this.
  
• Which is best to be kept constant, the military/civil turns ratio or the duration of the military turn?
  I don't know much about the facts from the real world, but from what is already said, a month is a valid operational time for campaigns in all ages. What makes ancient campaigns longer is in fact the unit technology that results in lower movement, but this is already taken into account in a war theatre scale. So a month is good for a DEFAULT setting.
  The civil turn, on the other hand, is much more arbitrary in reality, because (IMO) the only real facts to justify a one year turn are the harvest and the state budget, both major economic events (the one in antiquity, the other nowdays) that happen once per annum. The need for flexibility in the game length also dictates that the civil turn will vary from game to game, while inside the game, that will depend from the choice of linear or exponential time. That leads us to the conclusion that the turns ratio will probably not be constant.
  
• Will the military turns be always on?
  If they are always used, this will inevitably lead to enormously long games, even if we keep a standard ratio of 12:1 or less, while most of those turns will be without operational interest. Another aspect is the waste of processing power, because, even if the player does nothing in a whole string of 12MT, each AI civ will need to move it's units around 12 times before the player gets his next CT (Of course mouvement will be simultaneous and all the AI civ will make their plans while the player makes his, but when the player does not participate, the PC will inevitably stall).
  IMO, MTs must be turned on only when a war is happening, and only for the civs that participate. One would claim that there will be hardly ever a period without war, so there is no point in this, because it will not do much to reduce the total number of turns. But then, not all military actions are campaigns. Raids from hostile cultures, as long as they are normal, they are totally abstract, so they need no military ops. If they are in force, it's just like an invasion, the player gets kicked into campaign mode. This occurs from the first MT if the invasion is anticipated, or from the MT of the invasion, if he was caught off-guard. If he is in the offensive, he turns into campaign mode himself (Unearthing the tomahawk of war!) and starts moving his TFs towards the campaign target, even before war is actually declared. Attrition warfare, with raids, sieges and piracy (respectively air-raids, trenches and convoy hunts for modern times), can be fought over in CTs. In peace time, no TF mouvement is necessary, since the TFs can be initially deployed or redeployed anywhere (in national territory or in an allied country following a treaty) within a CT, which represents not only mouvement time, but also establishing a base and reestablishing supply lines. If the war ends before a whole CT has passed (not by a treaty, since diplomacy works in CTs, but through a truce or ceasefire) the player will be able to demobilize and then warp to the end of the turn.

The model that I propose for time elements is in essence the same with the space one, only more complicated, since non-linear elements kick in:

1) Base CT duration. The maximum duration that a CT can have throughout the whole game. For the length of Civil Turns, the following options will be available:
B=0.25y (=3months=1season)
B=0.5y (=6months=2seasons)
B=1y
B=2y
B=4y
B=8y
B=16y
B=32y
B=64y
B=128y

2) Linearity. Q How many of the above 10 ranges will be used in a telescoping timescale. Eight options will be available: Q=1~8 ; 4 main and 4 secondary.
Linear game______________1 range
Telescoped game__________3 ranges
Graduated game___________5 ranges
Exponential game_________7 ranges

3) Total number of turns. T There is a primary and a secondary range, both based on a power of 2, to conform with the other variables. Options are:
Ephemeral game________128T____(192T)
Very short game_______256T____(384T)
Short game____________512T____(768T)
Normal game__________1024T___(1536T)
Long game____________2048T___(3072T)
Very long game_______4096T___(6144T)
Eternal game_________8192T__(12288T)

4) Total length of game. Derives from the above, using the following method:

From linearity we can derive the analogy of turns spent in each range, following the principle that the amounts of years spent in each range are equal, with the ecxeption of the longest range which bears a 1:1 analogy to the previous one, making it so that S, the total number of turns deriving for each turn of the longest range is always S=2**(Q-1) and not 2**Q-1 as it would have been otherwise. Then D, the equivalent of S in years can be computed: D=B*(Q+1)/2. The total length of the game is then L=T*D/S. D/S, in years/turn is as per the following matrix:

B\Q |_____1 ______2 _____3 _____4 ______5 ______6 ______7 _______8
0.25|0.25/1 ____NON ___NON ___NON ____NON ____NON ____NON _____NON
_0.5|_0.5/1 _0.75/2 ___NON ___NON ____NON ____NON ____NON _____NON
___1|___1/1 __1.5/2 ___2/4 ___NON ____NON ____NON ____NON _____NON
___2|___2/1 ____3/2 ___4/4 ___5/8 ____NON ____NON ____NON _____NON
___4|___4/1 ____6/2 ___8/4 __10/8 __12/16 ____NON ____NON _____NON
___8|___8/1 ___12/2 __16/4 __20/8 __24/16 __28/32 ____NON _____NON
__16|__16/1 ___24/2 __32/4 __40/8 __48/16 __56/32 __64/64 _____NON
__32|__32/1 ___48/2 __64/4 __80/8 __96/16 _112/32 _128/64 _144/128
__64|__64/1 ___96/2 _128/4 _160/8 _192/16 _224/32 _256/64 _288/128
_128|_128/1 __192/2 _256/4 _320/8 _384/16 _448/32 _512/64 _576/128

Without even extending this any further, we can achieve a range from 32 years, which is the average duration of a great leader's carreer, up to (brace yourselves!) 1.572.864 years, which is the average duration of a whole species!

For a default civ-style setting, starting in the early BCs (doesn't have to be exactly 4000BC, since the map is NOT EMPTY in the beginning) and ending into the close future, we could choose one of the following:

___T___Q____B_______L
1536___1____4____6144
1024___2____4____6144
1536___3____8____6144
1024___5___32____6144
1536___7___64____6144
for various linearities and
_256___5__128____6144
_512___5___64____6144
1024___5___32____6144
2048___5___16____6144
4096___5____8____6144
8192___5____4____6144
for various turn numbers

All time elements will be set through a single window, so that toggling with the ranges provided will give easily a good result. If actual dates are implemented, they will be set with the use of L and a standard ending date. Of course all this will be different in a scenario, where the author will be responsible to set these parameters and will also have the freedom to add more turns in the beginning or the end of the sequence, to set the CT/MT ratio (even to 1:1) and to edit the ending date (and consequently the starting one too).

Concerning now the time - mass interaction, or, in other words, how are the time settings going to matter to the game:
All the interaction will derive from the length in years of the CT, which will provide us with a multiplier for all increase/decrease rates that are used (space affects quantity and time affects rates). Movement points, birthrate, disease, research, life span of characters, everything will be adjusted so that playing 1024 years of game will be the same when played in 8 turns as well as in 1024 turns.
For games that do not have the standard length (6144years?), the player will be jumpstarted in the appropriate era, so that, in combination with a provision for premature endgames, a player will have the choice to play only the one or two historical eras that he enjoys best.
There may be a few problems here though, the chief one deriving from the compound interest law. As we all know, a monetary sum, albeit having the same interest rate throughout a year, gives bigger returns in the end when capitalization of interest occurs more often. The total return is following a suite where lim = e = 2,712. Exactly the same thing would happen with any of the rates in Clash, so that the same amount of time, when played in a lot of turns (say 32 or more), would be more than 2 times more productive than it would be when played in a single turn. If this is left unsolved, it can ruin all our efforts. One thing we can do about it is to tweak appropriately the rates for each resolution, so that all returns over the largest time strech concerned get evened out around a value of our choice. This should not be very time consuming.

C) Comments

The greatest advantage of having this kind of standardized mechanics is that you can achieve the appropriate nominal game size/length, having in the same time a variety of choices concerning the number of tiles/turns and the linearity of a timeline. By this we promote customizability while we preserve compatibility, since games, even if they are different, they will be easily comparable and transformable to any suitable format, via the transformation utility.

Multiplayer issues: Multiplayer always has big troubles with time, which is either too short (lag in RT games) or too long (waiting for your turn in TB games). Clash is aspiring to partially solve this, by being an inherently simultaneous game, so that nobody's time ever goes really wasted, since all the players and the AI are simultaneously plotting and revising their plans for the turn to come. Nevertheless, if we decide to use the concept of CT vs MT, problems occur since the warring parties are playing a serious amount of extra, intermediary turns, while the others are waiting, unable to participate. Some people fear that this is going to ruin the multiplayer aspect of the game. Although I am not a multiplayer gamer myself, I have the feeling that conditions will not be so harsh, and here's why:
a) The MTs will not last very long. Unlike the hundrends of units of civ, Clash will have a small number of TFs, while there will hardly ever be any non-combatant units, such as caravans, engineers of spies. A MT has no reason to take more than one minute.
b) A big problem would only occur only in the very early turns, where a CT would have scores of MTs in it. But most armed conflicts are usually over in a few stages and if they aren't, they degenerate in some form of attrition warfare. Real armies are seldom engaged in a mobile war for a very long time. Anyway, if there is a problem, the players will be free to choose a more linear game or even tune down the MT/CT ratio on their own accord.
c) Since Clash is simultaneous, the non-fighting players will be given more time to fine-tune their civs and micromanage their favorite aspects. In the same time, the fighting players should feel the pressure of having to cope with a three-fold load: to wage the war in the MTs, to manage the state affairs in the ongoing CT (some of them are not in campaign mode from the beginning, but were attacked, so they were interrupted from this task) and worst of all, to shift their economy towards the war effort. Since mindless and repetitive actions will be minimized, all the pressure will fall on swift decision-making and not on a clickfest.

I would be glad to see my processings even partially implemented in the final game, although I recognise that it is one of the last elements to be included. I have the feeling though that now is the time to settle such fundamental things once and for all, because otherwise our whole view of the game would be inconsistent. I for one have tried to cover all possible aspects, from a wholistic point of view and I hope that the following discussion will share the same philosophy. The main goal of this post is not to impose a cerain approach, but to ignite a very much needed conversation on the subject.

I also dug up some (really old) related threads:

• SIS: ones and zeros and decimal points, oh my!, by Don Weaver (from the old forum!)
  
• RT vs Turn Based, by F_Smith
  
• Telescoping Timescales - Like in Civ, by Mark_Everson
  
• Timescales, Mark, timescales.., by Jools
  
• World Shape, by Kanzid Stonebreath

As I have now reached the end of this monumental post, I would like to thank anyone who had the patience to read it.


------------------
"In a time of universal deceit, telling the truth is a revolutionary act."
George Orwell

[This message has been edited by axi (edited January 28, 2000).]

[Mark_Everson]

Axi:

Wow, that's quite a post!

I agree with you on the spatial scales. My only quibble would be, why does everything have to be by factors of two? Isn't it simply enough to let a player or scenario designer set the length scale arbitrarily?

I agree with your basic analysis of the problems of having multiple military turns per civil turn. But I don't agree with you that that is what the plan is! It is simply that the two types of turns should always proceed together. But that they can have different timescales. So, 1 military turn Always happens at the same rate in the game as 1 economic turn. This means that you can have a military turn being one month, while the civil turn represents one year of stuff you could build... I know it's strange but bear with me. I've presented this, which I thought was the best way to do it numerous times, but you may have missed it. And as far as I know Harli said that, although he didn't necessarily understand it completely, that he was willing to try it. (Harli, please speak up if I'm wrong...) Here is an excerpt from the dead military thread that I hope will make my position more clear... If this little bit is not enough for you, you can back up in that thread to read more, and also read the military discussion thread, in which this is talked about at length.

posted January 18, 2000 08:58 from the dead military thread page 4

I think the goal right now is to have a single unified turn for economic, diplomatic, and military actions. A military turn will always cover one month of activity (at least that's the proposal and seems to work in terms of the other constraints needed by the system). At the start of the game the mismatch between economic and military scales might be as large as 100. By contemporary times I'm sure we can reduce the mismatch to a factor of 4. Whether a factor of two, or even 1: 1 can be achieved will not be known until sometime in the future.

End quote...

I'm now working on how exactly to handle this weird scaling at the interface between the economic and military models. I think it can work just fine, with just a few "huh?" places in the system which would be most blatant in ancient times.

On the rest of your stuff, again I agree with your general thrust. However, I don't think we necessarily need to stick with the rigorous factors of two thing in either time or space. I do think that we can let the player be the final arbiter as to how much game time is spent in ancient and medieval world vs. how much in the modern...

Anyway, your careful analysis and presentation will hopefully help us to reach a final agreement on all these issues. Once we agree on the official line, would you be willing to modify your excellent post so that we can present it as The definitive answer on the scaling question by placing it on the web site?

[Richard Bruns]

axi:
Thank you for your comments and evaluation. Clash needs to be customizable if it is to survive. I have tried to do this at a small level by making a tech system that can be customized or even rearranged rather easily. Your global analysis should be a big help if you are able to keep up with all of the game models at once and comment on how they fit together.

[Lord God Jinnai]

One thing that may help (or hinder) determine the timescaling question is that u must consider that civilizations began at aprox 5000 BCE. It might be possible to get by with 4000 BCE, but not any later (except for scenerios) because then u start the game off with many civs having major advantages over others militarily, culturally, scientifically, economically. Still I think we should aim for 5000 BCE as that is when the first cities were being built and agriculture was becoming part of the norm over nomadic traditions.

[Kanzid Stonebreath]

Axi wrote
----------
or a fractional jump grid that could be considered,
----------
Could Someone please tell me what a ractional jump grid is as I have never heard of them before.

Other than that little question I have to say that the proposal looks very good especialy that if you doubble the turn length you doubble the amount of things that happen.

------------------
What does this box do I wonder?

[axi]

Some quick answers:

Mark:

quote: I agree with you on the spatial scales. My only quibble would be, why does everything have to be by factors of two? Isn't it simply enough to let a player or scenario designer set the length scale arbitrarily?

This kind of standardisation is the only answer to the compatibility vs customizability problem in any industrial design and that is why ISO exists, so that equipment designed from different engineers would be compatible. If everyone chose arbitrarily design parameters, then worldwide industry would be reduced to chaos. Following this line of thought, if every player were allowed to screw with the numbers, different Clash games wouldn't be comparable or transformable. With the system I propose, a single savegame could be easily transformed to the format each player prefers. A game could even start as a very brief one, and later be transformed by anybody into something much more detailed. If someone would chose to use non-standard parameters he is free to do so, but he must be aware that he is on his own. As for the factor of 2, it allows for the smoothest increments possible (remember again the tenfold transitions in CtP!)

quote: It is simply that the two types of turns should always proceed together. But that they can have different timescales.

Then you are proposing a constant MT/CT ratio?

LGJ: When a payer is jumpstarted in 2000BC or 0AD or 1000AD or whatever, he is already some way down the tech tree and his land is developped in some degree. Wether he should have a handicap or an advantage against his rivals is a game difficulty issue. We could always have the AI play by itself (in the crudest possible setting of course) and hand over the pre-worked civ to the player at the starting point.

Kanzid Stonebreath: The fractional jump is a term that I made up in order to briefly describe the grid on the left, as opposed to the normal square grid on the right. All tiles are equivalent and movement is possiblebetween any two tiles with some common borders. It is a silly idea actually.

[______________________]...[][][][][][][]
[__________][__________]...[][][][][][][]
[______][______][______]...[][][][][][][]
[____][____][____][____]...[][][][][][][]
[__][__][__][__][__][__]...[][][][][][][]
[_][_][_][_][_][_][_][_]...[][][][][][][]
[][][][][][][][][][][][]...[][][][][][][]

All: I would be glad to help with the discussion here, but you should be patient with me, since it's the beginning of the examination period here in NTUA, so I don't have a lot of spare time. The previous post actually took me about ten days to write, in the midst of sleepless nights and exhausted deadlines.


------------------
"In a time of universal deceit, telling the truth is a revolutionary act."
George Orwell

[Harlikwin]

Ok, scale stuff as it pertains to the military stuff. I am willing to try Marks idea, it doesn't seem too bat too me... But with this in mind:

The main concern of the military scale issue is to present realistic movment rates and number of battles between opposing TF's. That said in terms of the ancient game we can go with marks X:1 system and just screw around with the movement rates of the TF's This should be able to "fix" any huge discrepancies in the ancient ages. As for the batttles, well for the most part in the ancient ages, one army one the other one pretty much got totaly destroyed, most engagemnts were pretty decisive one way or another, so we do not have to worry about too many battles. Also my aim in the battle system is to keep the number of battles limited to the important (read large) ones.

Btw: Correct me if I'm wrong on your sytem mark...

-Harli

[Mark_Everson]

Axi:

I am not proposing a constant ratio in the way that you meant it. What I propose is that one military turn always equals one C. T. However the actual timescales in the two kinds of turns can be very different. So for a particular example we can have that an economic turn equals one year, and the military turn equals the standard one month. So the building of economic infrastructure over one turn will be an amount that is correct for what happens over one year in the real world (at least as best we can do with our simple models). During that same turn, military actions that would be expected to take one month in the real world will occur. The only real difficulty is where the two models interact. Specifically the speed of building a military unit's hardware, and training the unit, the rate of building of fortifications, and several other things. These are the problem points that I mentioned above. I believe that they can be solved without getting too bizarre results compared to what happens in the real world.

I believe this abstraction is basically what Civ uses. It has its flaws, but it really works very well in general. If you want to pursue this idea further, you need to look in the references I cited above.

Harli:

I think we're in agreement although one can never be sure until the details are filled in!

[This message has been edited by Mark_Everson (edited January 29, 2000).]

[Richard Bruns]

Three things:

1) I would like a game that starts in 6000 BC. That date corresponds to the founding of Jericho, the first real city. I think that is a good start for a human civilization game.

2) I have an idea for treating military units in the early game. Assuming that the average turn early on is 10 to 20 years, why don't we treat early military units the same way we treat airplanes or cruise missiles? A soldier will only be fit for fighting for at most 20 years anyway, and after that they get too old and exhausted. So people can build attack units with the realization that they will last one turn. They send it out on a campaign path in the same fashion that they would order an airstrike.
For defense, players build garrison units that automatically replenish themselves every turn. If their territory or a nearby one is attacked, they 'scramble' like fighters to meet the threat. If they win, they return to their base.
When the armies meet, the current military model takes over, allowing the player to choose tactics and commant troops like they normally would.
This system has several advantages:
1. Less troop micromanagement, and thus quicker miltiplayer games
2. Realism. The scales for miltary and civilian operations will not have to be different. Also, you will not have some military unit exploring the map for thousands of years. Map exploration would be done with units that return to base at the end of the turn. This would be compensated for by giving the units more movement. Alexander's armies explored and comquered all of Persia in less than 10 years. A Clash player could duplicate this by building a lot of armies and sending them swarming into enemy territory. They fight and return to their homeland in one turn. The territory is then open for your people to colonize.

I am not suggesting that this be the only method of troop movement. I am suggesting that it be an option for the early game. It should require very little excess coding; just import the code for fighter and bomber behavior. I know that this idea in its current form is not complete enough to be used, but I think that it can be used as the basis for something suitable. I would like to know what others think about this.

3) I am currently working on a detailed tech tree post. It will be posted under a new topic called, "Demo 5 Tech Tree" before Monday.

[Mark_Everson]

Richard:

I'm really looking forward to seeing the Tech Model!

On your 2. I don't see how you could do it that way. You would be able to fire attacks all over the map, with no real reflection of the military realities of it. You say it's more realistic. The only way in which it is realistic IMO is that the times to do things are historically correct. Every other aspect of realism I can think of with respect to military campaigns is completely blown out by this approach.

All:

I just put up information on how I think the military and economic scales should interact in the Military Model III thread.

[This message has been edited by Mark_Everson (edited January 29, 2000).]

[Richard Bruns]

You couldn't attack anywhere you normally couldn't. The limit of the attack would be the defensive frontiers of the enemy or your normal movement. You could move exactly the same distance you normally could in that time. Besides, historically governments have been able to send attacks "all over the map" in periods of less than 20 years. Alexander and Caesar conquered huge chunks of territory in less than 20 years.

This system simply streamlines the process for players who don't want to micromanage troops in a time scale that is different than the normal one. I am not suggesting that you abandon the current model. I would like it to be an option for players who want to focus on managing the civ.

[Krenske]

Axi
On your point below -
"Since Clash is simultaneous, the non-fighting players will be given more time to fine-tune their civs and micromanage their favorite aspects. In the same time, the fighting players should feel the pressure of having to cope with a three-fold load: to wage the war in the MTs, to manage the state affairs in the ongoing CT (some of them are not in campaign mode from the beginning, but were attacked, so they were interrupted from this task) and worst of all, to shift their economy towards the war effort. Since mindless and repetitive actions will be minimized, all the pressure will fall on swift decision-making and not on a clickfest."
I find this as the obvious solution.
The problem with it is that this will mean timed turns else the two warring sides just keep moving along with their war and slow everyone else down even after fine tuning etc.
Another option would be to have in multiplayer a limit on the tactical depth that the player can go to. For example in Single player the players can drop all the way to TF commander and give general orders per engagement on the battlefield. In multiplayer it would be best to just limit players to the basic chalkboard ordering as originally proposed.

Multiplayer players would then be able to target TF's on the enemy etc but not control each battle.

[Lord God Jinnai]

This was discussed before on the multiplayer forum. Check it out first.

I've bumped it up so u can check.
[This message has been edited by Lord God Jinnai (edited January 31, 2000).]

[axi]

***bump***

I brought this to the surface because in two occasions people were asking about things included here and I had to redirect them.

I also brought it up because it doesn't seem to me that we have decided on it yet and I don't like to present it as a guideline while it is not approved by the team.

Mark_Everson: The 1:X CT/MT concept is it still valid? Will it be used in he upcoming demo 5. Will there ever be a demo realisation on the 1:12 concept, so that they can be compared?


------------------
"In a time of universal deceit, telling the truth is a revolutionary act."
George Orwell

[Mark_Everson]

Axi:

Well, I have already given you my opinion on most of these things. I will try to recapitulate my opinions here. I think that for many of the things you are commenting on, it is really too early in the design to lock in. We really don't know whether the "standard" tile size will work yet. So IMO it is premature to lock in a whole bunch of things based around it. I agree that everything in the game needs to be done with scalability in mind as much as possible. So here are my thoughts. I don't have much time, so things are put very briefly.
A)
1) Size of tile. 1 tile = AxA km....
I agree that resources should be scaled to the area of the tile. Everything else follows from the density of resources, so scaling should be no problem. The military model poses some difficulty, since some of our assumptions on whether a battle can be joined by an outside unit or not depend on the 100 kilometer square size. As I said before I disagree that we need a rigorous hierarchy of square size is that increase by factors of two. I think your idea for scaling maps so that people with different desires for maps size can play the same game is valuable.

2) Map resolution. X tiles length...
I like your calculation for the "Normal" world size. Beyond that I feel the other restrictions are unnecessary.

3) Size of the world ...
similar to you, my comments on this follow from 1 and 2. Agree we should allow for various parts of the map to be stitched together at different boundaries, giving things like cylindrical maps...

B)
we have already discussed at length military and civil turns in the rest of this thread. Unfortunately much of your initial analysis was IMO based on a misapprehension of where the military model (with respect to turns and timescales) was going. I am not going to respond to that part, because at least in my opinion most of it is irrelevant to where we are now. My basic feeling is that the economic timescale will change exponentially over the game from perhaps something like five years per turn at the beginning, to three months per turn at the end. However, the player should be able to adjust this at their discretion. IMO the military turn length should be kept constant at one month for the standard game.

On the rest of section B I don't have that much to say at this point. There is a lot of really detailed stuff, and I don't think I can either agree or disagree until we see how some more of the basics of the system work in demo 5. Your issue about getting compounded growth correct is very good. I think we can correct for this discrepancy with a sufficient fudge factor without much trouble.

C)
on your issues with multiplying, most of the issues go away if there are nonmultiple military turns per civilian turn.


"Mark_Everson: The 1:X CT/MT concept is it still valid?"
I'm sorry, I don't know exactly what you mean by this. As I have said repeatedly, IMO there should only be One Turn. Both military and economic things happen within it. The only discrepancy is that the timescales of things that occur for the two turns may be "out of sync" with the real world.

"Will it be used in he upcoming demo 5."
We haven't decided yet, but my guess is that what we should do is have the economic timescale = 1 year, and the military timescale equal one month (as always). This will give us a good test as to whether our concepts for handling different timescales will work. If it will work for a 12: 1 timescale ratio, then I'm fairly confident it will work all the way from 60: 1 through 1: 1

"Will there ever be a demo realisation on the 1:12 concept, so that they can be compared?"
Well, for now a test would be meaningless. Until the AI is of fairly high quality we will learn nothing from this IMO. As I mentioned in the military thread, I think the 120 ticks per turn, with economic turn of one year, is just a specific case of what the player will eventually be able to do in Clash anyway. Specifically the player can accomplish this concept by setting the economic turn timescale to the same as the military turn timescale (one month) and then when playing always push the "ten turn" button. So, as I remarked to Krenske before, I think the approach he suggested will be easily doable in the final game. Some players, including me, will probably play this way if the AI is good enough. However, others will want to handle every single turn by hand.