Page 1 of 1

The enhancement of Strategic Elements in Globulation

Posted: Fri Jan 12, 2007 8:16 pm
by Xylix
It is perhaps unfortunate that globulation, as of now, lacks deep and compelling strategic elements. However, this is not only easily addressable, this can be addressed without adding much, or anything, to basic game play.

But first, before we even discuss measures of increasing the depth of strategy present in globulation it is necessary to understand what strategy is... not in the sense of definition, but rather in the sense of derivation.

What is Strategy?

It is natural to presume that strategy is complexity... however, this is far from the truth. The greatest and most difficult strategy game currently played by humans is anything but complex. “Go” or “Igo” if you rather, involves only one type of piece and a rather simple rule set. However, the difficulty and depth of strategy available in the game is such that computers, which can already beat the greatest chess masters, cannot even face off against high ranking amateurs in the world of “go”.

So, if strategy isn't complexity then what is it?

Well, strategy derives from one thing, and one thing only: Choice. Strategy results from the player having multiple choices, but even this is a simplification. Real strategy is more than memorizing choices, but rather in the making of reactive choices. There is no one “perfect” opening in chess or go, though there are “better” openings. More importantly, all moves very quickly begin to be relational. One player moves in response to another. From this I can derive the following conditions for choices to enhance strategy:

* There must be no one definitively right choice, though there may be better choices
* Choices must be relational to the choices made previously by all players

Unfortunately, globulation does poorly in meeting these two standards. Globulation has choices yes: Which buildings to construct, what units to build... However, for the most part there is only one “right” choice. The building of a second, third, and fourth level training facility is not a “choice”, but an absolute necessity. A swimming pool is required, the race track has such a profound impact on logistical reach and economic production that it cannot be ignored.

Not to say that there isn't choice in here. The order of building, upgrading, and even placement is also choice. However, these represent only small periods of game play. Worse, the order of upgrading buildings, the position of building placement and even construction is, for the most part, a memorized strategy...

... or in other words, no strategy at all.

Equally, globulation tends to have poor relational choices. Like most RTS games the player has no idea what the other player is doing. Now, I'm fine with that, but the real problem is, and especially when on smaller maps, that player to player contact is generally short lived. When a group of warriors comes bursting into the other's base the fighting is usually fast and definitive. Victory by one side or the other is generally decided shortly after the initial attack or two. This leaves little time for construction and relational choices to take effect. This limits real strategy.

Unfortunately, all relational choices start when, and only when, the players start attacking each other. Given the quickness of victory (especially with lv4 Warriors, see the wiki warrior rebalancing on my wishlist), and more importantly the utter crippling effect resulting from the loss of certain buildings there is little that can be done in relation to the actions of the other player. Instead, all actions are taken based on “guesses” of what the other player will do and when. Strategy is reduced to little more than “canned” methods.

Now, globulation is far from being the only RTS that is like this. However, there is no reason that some thoughtful rebalancing and some consideration cannot drastically improve both of these elements.

Adding Choice

In order to improve the strategy of globulation the first thing to do is to maximize the two elements mentioned earlier. That is, to not only add choices, but to help ensure that these choices have the chance to become relational. In this I can state some basic goals:

*Every basic strategy should have at least two “good” choices or solutions
*Game play should be adjusted so that it takes multiple waves of assaults to achieve real victory, and in such, taking enough time so as to allow the player being attacked to develop an effective relational counter strategy.

As a greater idea, I think it would be an excellent idea for players to have a wider ability to choose and develop from the beginning of the game more uniquely. In this, the goal is that the number of “right” or “good” choices in the beginning of the game is not only sufficient that each players start differently, but also so that these differences continue to develop throughout the game and effect end game play. Hence, the actions early in the game cause late game strategy to differ... preferably in a significant manner.

Choice Enhancement Suggestions

Here are some suggestion to improve basic choice:

*Upgrades offered to units by the Barracks, Racetrack, Swimming pool should be sufficiently small as to never be required... or at least, never be completely required (lv2 Warriors should be able to “fight” lv4 warriors).

*The upgrading of buildings should not be the “obvious” correct task. The order of upgrades should not be obvious either. A player should be able to be successful using all level 1 buildings, via using his resources on other important tasks.

*Many critical buildings: Swimming pool, Racetrack, Barracks, and School, should be deemphasized or made redundant. It may be worth considering reducing the size, and/or resources consumed to build these buildings... and halving the number of units that can be trained in them at one time (especially for the higher level ones!). The point being to cause bases to have many more of these buildings, and thus render the loss of these buildings less crippling, and less decisive. (A percentile loss, rather than a complete loss)

All of these suggestion achieve both ends of the choice spectrum. They add both a greater number of “right” choices, and a greater ability for relational choice. A quick run down is as follows:

If building upgrades are less important, then a player's choice to not upgrade is less definitively “wrong”. Same with unit upgrades, and even with the ability to better control the number of training facilities. Equally because the destruction of specific buildings is less decisive, due to being less important or more redundant, the player is less likely to be crippled in an opening strike (or the first couple). The greater survivability increases the ability of the attacked player to create relational strategies, that address how they are being attacked.

Buffing it more with: Advantage

While adding choice does help, shoring up the relational aspect of the game is going to take a little more than just that. For this we need to talk about advantage. But first, I'll clarify what I mean when I use this word.

Advantage is the statistical benefit a player gains from various things. For instance, having level 4 warriors instead of level 1 warriors gives a huge advantage. This is because one level 4 warrior can destroy many level 1 warriors. Another example is if one player has a hospital close to the battle field, and the other has one very far away. In this a player gains a statistical benefit because his warriors spend less time walking to and from the hospital, and thus, proportionally, spend more time in battle.

Now, and this is important, we can even be more specific. The level 1 warrior vs the level 4 warrior situation can be called a “global” advantage. This is because it is true no matter where in the map it occurs.

Equally, the hospital scenario is a “local” advantage. This advantage peters out and decreases the further from the hospital the unit is. Further, it can even reverse, as the battle moves closer or further from a certain player's base.

In the most basic sense in globulation the following buildings already serve to garner certain advantages: Barracks, Racetracks, Schools, and Swimming Pools give global advantages, and Hospitals, Inns, and Towers give local advantages.

If you think about it, it is obvious which of the two advantages is more important to improving strategy: the local advantage. This is because it gives areas, or zones of strength. Further, it inherently allows theses zones to be strengthened more as needed... while not, at the same time, necessarily putting the player closer to victory. It also allows players to control the power of these zones either through the strengthening of their own, or the weakening the local advantage of their enemies. Finally, and most importantly, the collapse of a local zone of advantage has no effect on zones of advantage elsewhere... whereas the destruction of a source of global advantage negatively impacts the player everywhere.

The importance of the difference between the two cannot be understated. When one player destroys another players “global advantage” they set themselves on a road toward inevitable victory. The destruction of “global advantage” creates a snowball effect, where one player keeps getting stronger, and stronger, and stronger in relation to the other. Once one player has started destroying the “global” advantage of another player, it become increasingly impossible or difficult to stop them. This is more true, the more important and decisive that “global” advantage is.

However, when one player destroys another player's “Local advantage”, the story is different. The destruction of a local advantage does not, inherently lead to improved chances of victory elsewhere. There is no “snow ball” effect, because the defeated player is no weaker... except in that specific region. However, it does give the opposing player a chance to achieve his own local advantage in that area... this can lead to a “creeping line of force” style of play.

Advantage and Strategy

As I mentioned earlier, the correct manipulation of advantage, especially local advantage, can be used to enhance the strategic elements of globulation. The core idea here is to change the notion of the requirements of victory. In this, the equalization of advantage can be considered a strategic necessity to achieve victory. The idea being that the time dedicated, and the motions involved in, gaining, maintaining, or destroying advantage will be inherently more visible to the opposing player (Because they must either occur near the opposing base, or involve attacks on the/a opposing base). Hence, because of the presence of awareness and time, be relational.

To put it simply, enhancing the importance of local advantage forces the construction of “Forward Bases” so as to solidify local strength... and visa versa the attempt to destroy these bases so as to reduce the other party's local strength.

Unfortunately, there is little that can be done to address local strength factor of these buildings beyond adjusting rate and throughput. As it is, local strength can be measured, as applying to hospitals, as:

Hospital Strength Multiplier = (Time_Fighting ) / (2*Time_Walk + Time_Heal + Time_Fighting)

Hence, if it takes 20 seconds of battle before a unit requires healing, and 5 seconds to walk to the hospital, and 10 seconds to get healed the strength would be: (20 / (5*2+10+20) = 50% (or 50% of warriors engaged in battle would be fighting, 50% would be recovering). If another player had a 15 second walk time his local strength would be: (20 / 60) = 33% (or 33% of warriors engaged in battle would be fighting, 67% would be recovering): The local strength advantage would be: 0.5 / 0.33 = 150%, to the player with the closer hospitals.

There is, of course, a local strength advantage for inns as well and it follows roughly the same formula. However, it should be clear that a 150% advantage isn't much... the advantage between a Level 3 and Level 4 Warriors is greater than that (220%). But again, this only represents inns.

So the question is: what can be done? Well, several things. The first would be to reduce the time of healing and/or fighting, and thus increase the importance of walking distance. The second would be to increase walking time, by either slowing down walking, or using larger maps. Clearly, the prior is probably more favorable (halving heal time could easily be accompanied by simply halving the number of units that may be in a hospital at one time, halving fight time can be achieved by many modifications).

Doing this would create: 10 / 25 = 40% vs 10 / 45 = 22% = 180% advantage, for the player with the closer hospitals.

The other option would be to reduce the importance of the global advantage, by reducing the bonus from training warriors to higher levels. This, in result, increases the comparable potency of the local advantage, by eliminating ways of circumventing this advantage.

It is also worth noting that the race track currently creates a very large impact on the effective reach and drop off rate of local advantage. It is probably well worth considering reducing the growth resulting from the use of the racetrack, so that the effect is more standardized.

The strength of local advantage is, again, a very important consideration. The less of an impact it has on the game, the less players are going to pay attention to it. The less players pay attention to it, the less likely they are going to involve themselves in strategic assaults on the enemies sources of local advantage, and thus, the less likely the players will gain a meaningful amount of time to engage in relational choices.

Towers and Local Advantage

Towers also give a local advantage, however they are different than inns in hospitals in two very important ways. The first is that they are an end in of themselves, hence, towers are effective with very little unit back up. The second is that they represent a “hard” limit of reach. You are either within their reach, and thus the effect of the tower, or you are not. Because of this, working with towers is simpler.

On the opposite side, towers currently give very little meaningful effect on advantage. The reason is that their “range” is so terribly short. In fact, currently towers are useful mostly to control “choke points”... which do not exist very well in globulation in the first place. In fact, in order to use towers as a form of “local” advantage many, many towers need to be stacked or chained so that they may influence a sufficiently large area. An effect that is not only too expensive, but waters down the strength of the towers until they are meaningless.

Perhaps more importantly though, unless guard areas are placed with great care, it is likely that an enemy force can practically ignore anything less than a thick bramble of towers.

Now, towers are already in a situation of precarious balance. In order for them to be meaningful to “advantage” the reach of their presence needs to be large enough to create an “area of control”. Yet, they cannot be made so powerful as to overly tilt the battlefield, especially when concentrated. As such, an intelligent consideration of power and reach is desirable so as to allow towers (or multiple of them) to create zones of control that are more than mere pinpoints in the map.

I'd consider increasing range of the base towers up to as much as 9 or 10, and decreasing their damage proportionately as a penalty. A reach of 7 would also be quite workable. When doing this I'd suggest not allowing upgrades to increase reach further, but to rather concentrating the emphasis on how effectively, and how permanently the area is given an advantage.

General Summation of Ideas on Advantage

The general idea of advantage, is again, to force a war of local advantage before the achievement of victory. The reason for this is that this war, does not inherently weaken the ability of the “attacked” player to respond, while inherently making the “attacked” player aware of the attack, and thus capable of making a counter strategy... thus filling, or enhancing, the “relational” aspect.

The ideas covered here are:

*Reduce the importance of global advantage (unit upgrades) – this is also mentioned as good for increasing the number of “good” choices.
*Consider tweaking Inns and Hospitals in ways so as to maximize or increase local advantages.
*Consider reducing racetrack boost so that it doesn't so severely effect Inn and Hospital advantage.
*Consider increasing tower ranges so that they are more fit for governing area “control” and advantage.

And in general, attempt to think about how all new units and modifications effect of local advantage, and always attempt to maximize it's importance..




Related Idea: Explorers and Advantage

There has been a lot of discussion over what to do with explorers. The current decision is to remove the ability to attack buildings, and only allow the attacking of units. Personally, I would prefer a more strategic use of explorers than as a unit killer. As it is, using the concept of advantage, it seems to me a role is already open for them.

The role I suggest is not one as a unit killer, but rather as a vehicle for the destruction of sources of local advantage, most notably: Inns, and Hospitals. I'd also add in Level 1 towers (but not 2 and 3, thus playing up the “permancy” aspect of higher level towers). I'd also add workers, but in this case mostly to reduce the annoyance of hunting down every last enemy worker.

The idea of this is as follows:

Explorers make an excellent “hook” strike, because nothing can block the direction of their progress except their death. This allows them to effectively strike enemy forward bases, without being stalled by existing troops. Further, it renders the use of explorers very strategic because they are only useful for indirectly damaging the enemy. What is more, no amount of destroyed hospitals or inns will every instantly spell the doom of the opposing player, though the action may be quite annoying.

I add in the destruction of Level 1 towers to prevent explorers from being too casually guarded against (especially if ranges are enhanced!). Where as I am perfectly fine with Level 2+ towers being impervious to explorers, as this represents a solidification of defenses. Heck, it'd make for a great point of upgrading in the first place. (Again, upgrading as a choice, or to address a specific necessity).

Now, achieving this is actually pretty easy. All that needs to be done is to set explorer damage low, remove the splash, they don't need to be able to slaughter construction sites. I'd also suggest increasing explorer range because as of now, their AI makes them incapable of hitting crap without the splash damage. An increase attack speed to may also be worth considering, so that they still do enough damage. The final step is to merely allow their attacks to be zeroed by unit/building armor.

After this, setting the armor of inns, hospitals, and even level 1 towers at 0 will achieve this end. If necessary a further tweak can be made to workers, so that explorers are not especially effective against them.

Re: The enhancement of Strategic Elements in Globulation

Posted: Fri Jan 12, 2007 11:34 pm
by Elvish Pillager
Xylix wrote:* Choices must be relational to the choices made previously by all players
I like the city-building aspect of glob2, as opposed to your idea of "strategy". Currently, it's somewhat of "Who can build the better beast?" rather than anything about making and countering strategic moves.

Posted: Sat Jan 13, 2007 1:57 am
by Xylix
I like the city-building aspect of glob2, as opposed to your idea of "strategy".
I'd defer, that it isn't "my" idea of strategy, but rather the fundemental from which strategy derives. The existence of strategy is objective, not subjective. Games are either possessing greater or lesser strategic elements. What one "feels" about it is quite irrelevant, this isn't a war of one persons view of "strategy" vs anothers.

Now, I'm fine with you disagreeing with my "model" of from what strategy arises, presenting an opposing model, or even challenging the implications there of... though, I'll ignore you if you don't present evidence or logic for your reasoning.

...

In any case, my model of strategy isn't in the slightest, in conflict with city building. I didn't say city building shouldn't exist, merely that to insure it isn't a matter of memorizing: build A, B, C then D, actions must be relational.
Currently, it's somewhat of "Who can build the better beast?" rather than anything about making and countering strategic moves.
No, it is more like: "Who already has memorized the best way to build?"

I'd perfer that building, constructing, and war all be strategic..., and hence relational, and not reduceble to a single correct method. Hence, that building the best "beast" so to say, requires more than just memorization, but a deeper understanding of the game, and the implication of actions of other players.

In any case, if there is no "countering" of any form, there isn't any real strategy involved. It's only a test of how good the memorized method of each player is..., at best. At worst, it reduces the game to Rock-Paper-Scissors, with fancy graphics.

In any case, most of what I pointed out already exists in the game to a lesser extent. A slight fidding with certain numbers will just bring it to the fore.

...

As a final point, seeing as this is a Real Time Strategy game, methinks that strategy should be emphasized. Besides, it'll make the game not only more enjoyable, but greatly enhance the replay value over time.

However, I do agree that this doesn't mean dismantelling other gameplay aspects, especially primary aspects (limited number of unit types, minimal micromangement). I also agree that the high through put "endless army" is good part of the fun too. This is just an attempt to use what already exists to maximize strategic aspects of the game. It shouldn't derail anything in specific.

I'll admit though, it'd make good AIs much harder to design.

Posted: Sat Jan 13, 2007 2:15 am
by Donkyhotay
Yes allowing explorers to attack buildings could easily be done, it would actually be a reversion though. Originally explorers could hit both units and buildings and were overpowering. They were so powerful the game was little more then rushing to level 3 schools and then swarming with explorers. Since then, they have intentionally been designed to be effective at killing warriors but unable to damage buildings. Currently it is set where explorers are good at taking out warriors (magic goes right through warrior armor), towers are good at taking out explorers (explorers are incapable of damaging buildings), warriors are good at taking towers (tower attack will kill off a couple warriors by themselves but will fall to a mob). This is to encourage creating a balance between the 3 "attackers" (warriors, explorers, towers). Care needs to be made when making changes to not create more unbalances. Personally I think allowing explorers to train at level 2 schools would be better. This way you don't get them right away for rushes but early enough that it is possible to build up armies of both units before the game ends. I notice you didn't mention anything about fruit when talking about strategy which is a definite factor. There has been alot of talk about how to tweak fruit to increase the strategic consideration of the game.

Posted: Sat Jan 13, 2007 3:03 am
by Xylix
I'd point out that changes on explorers is an addendum idea anyway. However...
Yes allowing explorers to attack buildings could easily be done, it would actually be a reversion though. Originally explorers could hit both units and buildings and were overpowering.
I know.

In any case, the ability to hit buildings and units was taken away very recently (so recently that it's still true in my version, which is labeled 8.21 alpha!). Further, I'd point out that the danger of these units was for the most part well balanced merely by the intelligent decision to allow towers to shoot them down.
Since then, they have intentionally been designed to be effective at killing warriors but unable to damage buildings. Currently it is set where explorers are good at taking out warriors (magic goes right through warrior armor), towers are good at taking out explorers (explorers are incapable of damaging buildings), warriors are good at taking towers (tower attack will kill off a couple warriors by themselves but will fall to a mob).
I'm also more than aware of this. However, it is clear to me that the implications of this were not thought out. Lets look at it this way:

*Bases have towers.
*Warriors spend most of their time in bases
*Explorers die very quickly to towers.
*Warriors go to Hospitals, and thus injuries that occure near a hospital are meaningless

Hence, the only real use of the "explorer" ground attack is as a home base defense against warriors...

... which, uh, was the original point of the towers...

Essentially meaning, this action is virtually the same as taking away explorer attacks and making towers effective against warriors again.

You'll pardon me for finding this silly.

Plus, I've never been overly fond of the fact that actually using this ground attack ability requires an annoying level of micromanagement. I have better things to do than put down a throughly tweaked flag over every little thing I want destroyed.
This is to encourage creating a balance between the 3 "attackers" (warriors, explorers, towers).
Which is exactly the wrong mode of thinking. Instead, each unit should be viewed as having a specific purpose... and measures should be taken to ensure that they can acomplish these purposes.

In this case, my vision would give the ground attack the express purpose of destroying: Inns, Hopitals, Level 1 Towers (to prevent way too easy sheilding of buildings), and stray, lonely, unsported workers.

This serves two purposes:

1) Giving them a clear strategic late game use, that warriors often cannot fufill (because other warriors are stopping them from accomplishing the task).

2) Allowing them to still serve the purpose of mopping up that last few stray irritants, in the last 3 minutes of game play.
Care needs to be made when making changes to not create more unbalances.
I doubt that making explorers that are only effective against 4 units will be in any form game breaking. Much less when one of those 4 units, the tower, is only at vulnerable at level 1 and is the explorer's mortal enemy.

Especially if my suggestion of increasing tower range was also included...

In any case, this is easily balanced. I'll agree though, it can't be done without taking some time to think things through first! Players shouldn't be running 1 explorer into a base to acheive this... and it should be practically impossible to pull off on a late game base without 100s of explorers to sacrifice.
Personally I think allowing explorers to train at level 2 schools would be better.
I quite like that suggestion. And, the explorers should be balanced as needed so that this isn't an issue.
I notice you didn't mention anything about fruit when talking about strategy which is a definite factor. There has been alot of talk about how to tweak fruit to increase the strategic consideration of the game.
That's because I'm still not sure what exactly what should be done with it. Really, I need to play with it alot more and do some real fiddling before I feel justified saying anything on the matter. (I usually ignore the fruit)

As it is, fruit is kind of off on it's own right now.

Posted: Sun Jan 21, 2007 5:20 am
by Ferk
I just registered to say that I really love Xylix ideas.
I think that globulation could have much more strategic potential. Currently it is not so different than other "strategy/fast-action" games such as Warcraft that lack of real tactics other than being the fastest improving buildings and training units.

Increasing the power of local advantages could make the game even better, This also will make the role of the map terrain much more crucial..

Posted: Mon Jan 22, 2007 12:27 pm
by Elvish Pillager
Xylix wrote:I'd defer, that it isn't "my" idea of strategy, but rather the fundemental from which strategy derives. The existence of strategy is objective, not subjective. Games are either possessing greater or lesser strategic elements. What one "feels" about it is quite irrelevant, this isn't a war of one persons view of "strategy" vs anothers.

Now, I'm fine with you disagreeing with my "model" of from what strategy arises, presenting an opposing model, or even challenging the implications there of... though, I'll ignore you if you don't present evidence or logic for your reasoning.
Perhaps you take the word "strategy" to mean a slightly different thing than I do.

Anyhow, I don't know whether you'd consider what I value in the game "strategy" or not, but I'll take a stab at presenting the model.
Xylix wrote:No, it is more like: "Who already has memorized the best way to build?"
It isn't.

You need to build buildings where it is effecient for globs to get to them. You need to build them where they don't get in the way of glob traffic. You need to build them where they're not too hard to defend. You need to build them where they don't block off resources. Where they don't discourage future expansion. When you have enough workers to spare. Where they won't take too much of your workers' time to build... and many more factors.

And because you have limited area to build in, and because you have limited resources (and beating them back is sometimes a relevant goal, too), and because you have limited workers and limited time, and even because the enemy attacks you sometimes, you have to constantly compromise between these factors. There's so much to be learned about when you can or can't afford to build in certain ways that simple memorization isn't nearly enough to make you win.

Even if you are skilled enough to come up with an optimal strategy on a certain map (in which case there will still be various different strategies, debatable as to their optimalness, most likely), then you will have to go through the same process every time you play a new map. Glob2 isn't always exactly the same game.

Granted, the game requires a good deal of memorization as well, but so does every game.
Xylix wrote:I'd perfer that building, constructing, and war all be strategic...,
I wouldn't, and here's why:

Glob2 is a real time game. Because of this, the player only has a certain amount of time to think about each decision. Strategy requires that there be meaningful decisions to make, and thus, if war as well as city-building was strategic, then war would require you to make decisions, and thus to spend time making those decisions.

When I play Glob2, I can spend all or nearly all of the available time working out how to build my city as effeciently as possible. Thus, I spend all or nearly all of the time strategizing, as I would put it.

Increasing the amount of strategy per game element doesn't necessarily increase the amount of strategy per player time.

If anything, it discourages strategy and encourages memorization, because there is less time to work out a move or response during the game, and more need to have an already-worked-out idea on how to respond to a given move.
Xylix wrote:As a final point, seeing as this is a Real Time Strategy game, methinks that strategy should be emphasized.
You're getting caught up in the name. "Real Time Strategy" is the name of a genre of computer games, and like a genre of anything, an individual game can deviate in any given way from the overall scope of the genre.

- - - - -

As a final note, I'd like to say that I'd love to see a fork of this game using your ideas on strategy. I might even like it better than this one. It's just that we shouldn't lose what Glob2 currently is while doing that.

Posted: Mon Jan 22, 2007 6:10 pm
by Xylix
Perhaps you take the word "strategy" to mean a slightly different thing than I do.
Perhaps I don't, however since I presented a partially operant definition of strategy, I cannot consider yours valid unless your definition also addresses why certain games are more strategic than others.

Not to say yours isn't just as good. However, I can't discuss something that has yet to be shared.
Anyhow, I don't know whether you'd consider what I value in the game "strategy" or not, but I'll take a stab at presenting the model.
I'd love to hear it. I have found that merely understanding and discussing fundementals, and how the game takes advantage of them often improves the game itself.

It's the same for pretty much any "artistic" work, because all art is anything but magical, and may be reduced to component parts.
You need to build buildings where it is effecient for globs to get to them. You need to build them where they don't get in the way of glob traffic. You need to build them where they're not too hard to defend. You need to build them where they don't block off resources. Where they don't discourage future expansion. When you have enough workers to spare. Where they won't take too much of your workers' time to build... and many more factors.
But Elvish Pillager, for any one map:

*Buildings always take up the same space
*Resources are always the same
*Build areas are always the same.
*The amount of workers available at X time is always the same.

etc...

Play the same map 50 times, or even 100 times. You'll find that after a while you construct the same way every time. Every map has a memorized strategy.

Now, changing maps shakes things up, but where one pattern exists, a higher pattern also exists. Play Globulation 200-500 times, with lots of maps, and you'll almost certainly finding yourself doing the same thing every time you play on any particular map.

In the end, a memorized strategy is a memorized strategy no matter how complex that memorization is. It is inheriantly obvious because once memorized, the execution remains the same. The only way to avoid a memorized strategy is if things change. Unfortunately, the only source of chaos is the other players. Hence, they are the only thing that can be invoked to result in a fundementally "non-memorized strategy."

Of course, there may be other ways to do this.
Glob2 is a real time game. Because of this, the player only has a certain amount of time to think about each decision. Strategy requires that there be meaningful decisions to make, and thus, if war as well as city-building was strategic, then war would require you to make decisions, and thus to spend time making those decisions.
Now, that I agree with. Time is the one resource that is split fairly no matter how many players there are. To this I have several answers:

1) Reduce micromanagement: Building is a memorized strategy, at least until other players get involved. Take advantage of this, and create effective systems and "general AIs" that can reduce the burden on players for this I suggestion:

*Automatic upgrading system
*Automatic rebuilding system
*Delayed construction (Construct X, but only after you constructed Y)
*More generalized AI effects, so that micromanagement of buildings is decreased.
*Ability to set building features: E.g. the production distribution of the swarm, while the building is being constructed, or performing normal activity.

etc... Things that reduce the amount of time the player is spending doing, but also which don't particularly constrict choice. All this frees up this desperate amount of time.

2) Balance with time as a consideration: Currently, nothing is seems to be balanced with time as an all important consideration. However, everything should be balanced so that the time everything takes is sufficient that players are liable to get overwhelmed. In this, things like: The time it takes to kill things, the time it takes to build things, the time it takes to upgrade things... all should be adjusted for maximum time.

I've already addressed balancing with the time for warriors to kill eachother as part of the "Warrior Rebalance" idea I put forward long before this.


In essence, the idea is that freed up time elsewhere is time that may be spent elsewhere. The places it should be freed up from are from the things that can become tedious. Rebuilding, and re-upgrading the same inns every few minutes is meaningless tedium. So is having to come back to an inn you KNOW you want to be upgraded to level 3, and hitting the upgrade button twice. Everything that can compact or limit this nursemaiding of buildings is a good thing.

The same goes for war as well.


However, I do agree that time must, must, must be considered by the developers.

In any case, I'd also argue that distribution of time can also be considedered strategic. Personally, however, I'm annoyed if my time is stretched to its limit. That's a large reason why I like where Globulation is going, because reducing micromanagement gives the player back time.

In any case, this shouldn't threaten time all too much. Mostly because many of the changes I suggested would consequently increase playtime... something that has to be evaluated in and of itself.
As a final note, I'd like to say that I'd love to see a fork of this game using your ideas on strategy. I might even like it better than this one. It's just that we shouldn't lose what Glob2 currently is while doing that.
I personally would never consider making such changes, except as an initial test. As always, it is best to keep the old paths open in case the new paths don't work the way one expects. Everything can look good on paper, but not everythig always translates so well to reality.

I have played with some of the changes to a limited extent on my computer. Personally, I found no difficulty juggling until the end game (at which point, constructing forward bases, and attacking at the same time can be challenging). But even then, that was more because I was over reaching than because I needed to use that time.

But, I haven't yet, and cannot yet, simulate the full thing. Not without delving into actual source code. Something I'd rather avoid doing.