Agreed. And not only that, usually wars would only be fought like 3-6 months every year at most (no one would fight in winter, nor in harvest,...). But it's really rare for strategy games to take that detail into account.

Just to clarify, the original post addressed the problem of players being rewarded heavily for always building the strongest units they are capable of building, and never weaker units.  So if you were unable to continue to build death knights, you would move to the next most powerful unit you are capable of building.  The net effect of this behavior is that you typically end up with armies that are non-mixed or "monolithic." 

In most strategy games, there is little incentive to give ships / soldiers less effective equipment, even at a lower cost.  In Galactic Civs and Sword of the Stars, for instance, it was almost always more cost effective to swell the ranks of your combat fleet with not only the best technology, but also the largest size of ship classes. 

Well yeah... This is almost always a problem, and my point was merely that limiting army sizes severely exacerbates the problem.

The solution, IMO, is just to balance that initial training and production costs, as well as upkeep costs, so that there is a reason to build units besides the best they possibly can. Although honestly I think shying away from restricting army size will mostly solve this problem on its own. Total War is a good example - towards the beginning of the game, before you're anywhere near capable of fielding a maxed out army, you are generally better off by mixing troops of varying strengths and types. If you only train the most powerful troops available to you, you will be shooting yourself in the foot.

Sadly Total War's limits to army size are still relatively low, and so the problem does arise by mid-late game. 

Two points:

1- Frogboy(? - or someone) mentioned that troops don't just appear out of thin air -- if you want a large army you need a large population.

2- I'd like to see recruits have differing potentials -- anyone could become  a 'peasant' troop, many could become 'normal' troops, some could become 'skilled' troops, but few could become 'exceptional' troops. Training would bring out these inherent skill levels, not increase someone from one skill category up into the next -- ie no matter how much I train or what resources I have access to I'll never be a Michael Jordan or Muhammad Ali.

Combining these 2 ideas, I'd like to think that large armies of exceptional troops would not be possible, as they require exceptional people who, being exceptional, are in relatively short supply.  No amount of training time, resources, technology, etc. could enable this.

Thus we'd have armies comprised of a lot of average folks (levies/peasants/etc.), a number of skilled troops (swordsmen/crossbowmen/etc.), and a relatively small number of exceptional units (knights/etc.).  What type of exceptional units would vary depending upon resources, technologies, etc. (ie -- knights or samuri or ninjas or ???), but their numbers would be limited.

We could decide to not field large numbers of average troops of course, and just go with exceptional troops, but could not decide to field large numbers of exceptional troops by 'training' regulars to become exceptional.

20 Unit stacks in one army. It's been like since Medieval 1. If you know how to Hex edit and what to Hex Edit you can greatly increase the unit size in the Total War games. The whole reason they limit it like that is that most peoples rigs start lagging when you have more then 20 stacks on each side of a conflict. Limiting it to only 20 means 40 unit stacks on screen.

That's how they make the intro movie and demonstration movies for the games where you see MASSIVE amounts of units and you think "I can't wait to make a army that big" but then when you get the game you're like :Wth....how come I can't make a army as big as I saw in the intro?". When they make those videos they're doing so on what most users would consider a "Super Computer" which is why you'll see 40,000 men with no lag.

Medieval 2 is probably the one in the series that best illustrates this. Start up M2 and watch in the intro video, same for Rome. You'll see at least 40,000 men on screen in a siege in one battle but you can't make a army anywhere near that large.

How about buildings that would enhance the quality of the peasants that you use to make your armies? Build a training ground in a city and all swordsmen built there will have an increase in their initial stats. An archery range and your archers would start out a little better. If there is a cost to these buildings (taking your much needed space and resources) then you would have to consider city specialization.

 

That's a good solution. Another is to break resources balance: 1 full plated knight can cost like 100 peasants, but will die fighting 20. The problem there is to make the knight interesting to build even knowing the fact that he is not cost effective in simple engagements (could be done with maintenance cost, population lost, giving special bonuses to commanded units,...).

If players are rewarded in Elemental for having non-Monolithic armies, I definately don't want it to be for reasons like "leather armor and plate armor both serve different, but equally legitimate roles."  In other words, certain improvements in equipment and soldiery should simply be more expensive and better (sure, a soldier in leather armor might be faster, but still not as valuable per capita as a plate wearing soldier.)

The strategy in that regard ultimately would come in off the battlefield while you are deciding how to manage or exploit the resources that you are given.  For instance, I wouldn't be building a large army of lightly armed soldiers because lightly armored soldiers offer some kind of strategic advantage necessarily, but because I pursued a strategy where I sought an abundance of food and population.

HoMM actually is one of the few games that didn't invariably run into exactly that problem. Have you ever played as a necromancer in, say, HoMM 3? And raised an army of thousands of skeletons? Black dragons better beware, because those skeletons can wreak havoc on them. Thousands of skeletons will do major damage to even Black Dragons... This is not the only scenario in which low-level units are still effective. For example, they make excellent meat shields (using up retaliations).

And medium-range units can be even more damage in large numbers. One of the reasons for this is that HoMM doesn't really limit the number of creatures in a stack, just the number of stacks in an army. This allows you to make huge armies of weak creatures, under the right circumstances, that can still be effective against stronger units. In HoMM you will hardly ever end up with an army consisting of 7 stacks of Black Dragons (unless you also have armageddon...), because you'd be better off combining those dragons into fewer stacks and adding in full stacks of other, weaker units. 

First -- what's wrong with that?  It adds choices which is good game design.  Second, it doesn't force anything as you can freely choose to only use a certain unit(s).  It may put you at some disadvantage to do so as you'd be deliberately restricting yourself, but tactics/strategies may even the field.

Let's assume exceptional units take special buildings and/or special training and/or special resources and/or longer time.  Will you go with no troops while you're recruiting/training your exceptionals, all the while defenseless (troopwise)?  That would put you at great risk to an early 'rush'.

You wouldn't be able to mass produce exceptionals -- remember limits on troop numbers. 

Let's say your population could support 1000 troops total, and 10% have exceptional potential.  The choice wouldn't be 1000 exceptionals or 1000 normal troops, the best you could do is 900 normal troops (starting relatively soon) and 100 exceptionals (starting relatively slowly).  This suggestion would prevent ending up with 1000 exceptionals (or any number > 100).

Differing skill potentials does not preclude/limit strategies for customizing armies, it does the opposite -- adds choices which is good game design.

Let's asusme there'll be some sort of rock/paper/scissors aspect (pikemen advantage vs cavalry, or firemage advantage vs cold), this would add another, similar, dimension -- large numbers of less-skilled vs. small numbers of skilled, multiplying options/choices. 

It's also 'realistic' as not everyone has the same potential skills, and training can only bring one up to one's potential.

Nick, your idea for exceptional individuals is an interesting idea, but perhaps there is another way to go about it than just having a "pool" of them waiting to be tapped.  What if, instead, those exceptional individuals would be your experienced units that emerge from many battles and many victories?  Sure, you can throw any old unit an ax of uuberness and a suite a full plate mail, but perhaps they can't get anywhere close to tapping into its full potential unless they have reached a certain level of experience.

To stream line the process of requipping elite units, there might be a feature in which you highlight an army and then give the order for X number of soldiers over level 10 to return home to re-equip with the gear that they are more fluent at using.  

There is such a thing as a monolithic strategy--- in this case, building the highest calibur units every chance you get.  You know what we're talking about.  Juggling around definitions doesn't prove a point.