You complain about killer stacks and then suggest that killer stacking should be the norm and not the exception? The reason we have killer stacks at all is that there is a cap on army size. Its possible to overwhelm elite troops by drowning them in grunts. After a point its possible to deploy so many units that the skills and capabilities of your enemy are meaningless. Once the cap is there this is no longer possible.

Bleh too many uses of "possible", anyone got a better word

But in any case it seems we probably won't have this problem with elemental. It sounds like we won't have many elite "units", or if we do they would be hella expensive. Instead we would have elite soldiers or squads. Like the dragon for instance, you will only ever have one if at all.

16 capship fleet in SoaSE is my stack of choice. IMO, if there are extremely powerful units they should be one per army. Give them all huge egos so they refuse to work with one another. Or something.

To deal with the "huge omfg 10000000 troop" armies, have logistics effects based on size. Army speed for example. 10,000 people can't move nearly as quickly or efficiently as 10 people can. Also logistical things such as commanding units who need to be payed more. So if you make a huge unkillable army, it will be so slow you can't go raiding enemies or defending your borders effectively as well as having a huge upkeep cost due to the army beaurocracy.

The thing is though, economic damage doesn't matter when they already have a stack of doom.

One option to killer stacks is some sort of local supply rating, perhaps with buildings/improvements/spells you can use to help that out.  A big army takes resources to maintain in the field. You can port all those with you or forage from the surroundings.

I can see "porting supplies" in the form of caravans of supplies having to be purchased or some sort of production penalty to nearby towns if you exceed their supply. Which in turn means your inherent supply is better near your towns than in the midst of an enemy.

It would also place emphasis on quality versus quantity. Say it's cheaper to build a powerful stack of cheap units, but the supply issue can be crippling if it gets away from the premium supply zones. Where as that really expensive to create stack of elites doesn't have a supply problem, but takes longer to build and costs more.

I wouldn't want to delve too far into an entire supply economy, but I can certainly see the idea of supply caravan units supplying an army. Either prebuild enough for your army's trip or wagon train them there and risk your supply line being intercepted and killed. That would certainly make fielding a killer stack a more significant strategic effort.

It can if the troops desert/rebel due to lack of pay.

Ok I'll be the voice of dissent... I like the killer stack.  Let's say I spent the time and money to research the "storm giant" spell, and I have the mana to cast it 5 times, so why not let me use them all in the same place.  Let the AI be smart enough to make their own killer stack, or better yet have it start producing counter units.  Sounds like fun to me 

Stacking units is not just a means to build up an army but also a way of conveniently moving a number of units at the same time.

If game mechanics are such that damaging units in a stack is more effective (because of splash damage) than damaging unstacked units, the obvious work-around is to spread units out. This, of course, means you have to move more units each turn, making it a lot less convenient than moving stacks of units. 

Call me lazy... but I don't really want to be forced to play in an inconvenient way to gain a tactical advantage

Well in the world map it's turn based and if you finish your turn with the units spread out they're more vunerable to being picked off by small enemy parties hiding in stealth via the trees or magic.  So you may avoid most damage from a map spell such as those seen in AoW:SM, but it does leave you vunerable to surprise attacks.  For AoW:SM the ideal strategy becomes multiple groups where any one group is not a loss and it avoids most damage from map spells.  AoW:SM was a big success for fantasy TBS so I recommend the same method be used for Elemental.  If you haven't played AoW:SM I recommend playing the game before commenting to better understand since my few sentences doesn't cover the depth of strategy involved.

I don't like map spells though. If you cast "fireball" on my army stack. Am I to believe to that you just cast a fireball big enough to cover a square kilometer or more? If you can throw those around once a day then tactical battles would be over in seconds. Spell effects should not cover the entire army unless you had enough energy to make your spell big enough at a tactical level.

This isn't so bad in AoW because each unit is a single unit, one spearmen is just one spearmen. So a group of 9 units doesn't cover much ground. But when you scale that up to MoM and Elemental size it just doesn't make sense any more.

Great feedback.  NTJedi hit the nail on the head, at least where I was coming from.  The AI, was very inefficient at developing killer stacks which meant A human player could win as long as they survived for a while.  I'm a firm believer in having as many options for players as possible, it is important that each option has positives and negatives.  Its also critically important to have killer AI:)

I'm not entirely sure how you can deal with killer stacks without making combat seriously downplayed.  Like the point of a killer stack is that... its a stack that kills everything other than another killer stack in combat, and without such a stack then what is combat then?

Multi-turn combats (like that go for more than 1 game turn) might do it though.    It would require some sort of turn limit be placed on combats to mark when a battle is pushed to the next game turn, and I know some people are apposed to that idea.

I guess you are talking about AI more than the stack itself, but still.  I mean it is what it is.

Cost is an answer, but I don't think it is a complete answer.  It still is only a race, however, with the currency being essence it does create interesting tradeoffs.  That's what I'm talking about.  Creating tactical/strategic tradeoffs that players will have to make.  This is much more interesting than a race.  It also has to do with the criticality of losing a production space.  If only offense matters, it makes the Killer stack option much more attractive.  However, if your killer stack becomes weaker if you lose some important positions on the board, then it makes life much more interesting.  I could see a lot of these types of design elements built in to make player decisons more important and varied.