I do like the idea of logistics affecting the size of armies.

I also like the idea of being able to go beyond the nominal limits of logistics (Napoleon discovered this the hard way).

So for instance, you could have have "Expert Logistics" provide a nominal army unit size of say 8. Every level above that costs N gold per turn where N is how many beyond 8 you have.  So if you have enough gold, you can have a bigger army but it'll cost you dearly.

Agreed on both counts.

Also Good, agreed.

But then smaller kingdoms still wouldn't be able to keep up with bigger ones. Because smaller kingdom will usually mean smaller economy, and therefore less gold and less research. I'm not saying the larger kingdoms shouldn't have an advantage, but this 'double win' could easily unbalance the game.

Also, it promotes de-stacking when there are no enemies around, and only re-stacking when you plan to attack or fear you will be attacked. This adds a lot of busy work IMO.

Very good points, which is why I suggested that armies that surpass their logistical limit suffer combat penalties.  That way, there won't be any army juggling to game the system.

This really is the same reason why I'm against spells that are more potent against larger stacks.  In theory, it would seem to work.  But in the end, it will just end up being a massive quagmire of backward player activity.  A competent player would be forced to constantly break up their army, remake their army, and check and analyze an opponent's casting capacity every single turn to see if their opponent is capable of casting stack busting spells. 

As I and others have suggested before, the best way to discourage players from putting all of their units itno stacks of doom is to provide disincentives to do so.  Make rivals analyze how many units are in an opponent's territory after their opponent has marched off to war and have that rival attack if that number is small (thus, forcing the opponent to leave units behind to defend.)  Empower other strategic objectives which force good players to split up their forces.  This is the true way to control super stacks. 

But this goes on the assumption that the upkeep cost of an army is a linear progression, which doesn't sound like what Frogboy intends with his idea. If it is strictly linear such that a kingdom 2x the size can afford an army 2x the size, I agree with you - but that's not the impression I got from the admittedly brief pitch of the idea. The way I picture it has stronger scaling economic penalties beyond the "soft" army cap.

For example, both A and B would be able to get 8 units normally, but if A wants more it has to start paying x% of its total current income per unit - which is a much greater number than a normal upkeep of 1 unit.

I agree, no unit caps, but also no caps lock either please.

Having all your troops in one spot should be exploitable. Moving around large armies should be harder than moving around single units. That way you should be able to harass their cities and use mobility to your advantage.

Of course, good players would exploit this by disbanding armies before they moved and reforming them before attack, which is why I think there should be a movement penalty based on troops around them rather than only the ones in the same army. Of course it's not perfect either, but I think it might be a good place to start.

Well, a unit cap won't necessarily limit the size of the armies you are able to field, but their flexibility.  A limit on the number of units means you can have 8 units of 6 men each, or 8 units of 100 men each.  Investing in the ability to employ your military less rigidly could be very worthwhile with flanking, area of effect weapons and the like.  Now overall numbers, would be a different story, and that would have to be an escalating monetary cap based on the total number of men your nation has under arms.

I'm absolutely for a stack size limit, because in any game that doesn't impose some sort of limit the game swings towards putting all eggs in one basket approach, which leads to rather boring game mechanics: build up phase, train phase, stacks of doom clash - one player wins the other player dies. Basically HOMM style of gaming. So very static, not very dynamic kind of play.

So there should be some kind of limit. Preferably a modable limit. I would appreciate some kind of special game mechanic attached to it. But please, please not something like a bunch of techs like Logistics+1, Logistics+2, etc. this is such a bland and boring game design. I already hated such unimaginative tech design in GalCiv2. On top of that: I think there should be never ever a tech that grants you a bonus outright, like +1 logistics, +1 move to all etc. (also very often used in GalCiv2...). Good designed techs introducing new game mechanics. They maximal grant boni via proxies like units, buildings, wonders, what ever. I think it is better to have a tech tree that is lean and slim but with quality techs, than just add some 100+ techs that just add not much to game dynamics. So guys, a bit less of GalCiv2 or Space Empires 5 kind of tech tree, more something of Master of Orion 2 (great tech tree concept).

Now to the game mechanics of how I would tackle the logistics issue. Let's keep it to just one tech. As soon you (re-)discover the technology of "logistics" you are able to commit production of your cities on logistics per turn. The higher the commitment to more free slots you receive. With each slot you received you can expand one army with one additional unit. Those extra slots are not permanent - you have to keep the level of production funneld towards logisctics. If you reduce your commitment so you lose your free slots. Should you raise your production commitment then you raise these slots. Please note that it's really up to the player whether he spends the production of his cities on that or not and that it becomes increasingly harder to get more slots the more you allready received. This adds a wonderful dynamic to the game. You don't just receive some flat bonus by a tech - No! - you constantly have to deceide what has priority: more stuff (buildings, units) or bigger unit stacks? Those hard choices make games intressting.      

I can't wait to mod a race of Apemen to murder them all out of spite because of you.  Joking, but larger armies shouldn't be harder to move around, they may be slower because a army is only as fast as its slowest unit.  The game should flow easy and certainly not punish a player because they are smart enough to create a beast size army.  Don't forget about dragons and spells that will most likely wipe out hundreds of men as your guys level up.

This was a good system for determining army caps, one of the problems though was that the AI didn't follow it. So during wars you'd see one province minors building up armies of 100,000+ when they couldn't reach whomever it was that they were fighting, and even in peacetime they fielded much larger armies than you could in a single province minor.

Personally I think that here, champions should be counted seperately from units (even though units might be small).  Since your champions are going to be your leaders once your army sizes get larger, I think having two separate caps, one increased through research in adventuring techs (for your champions) and one researched through warfare techs (for your units) would be the way to go.

I agree

in addition with this separation, I think most/all Champions should have a higher morale/ stay at high morale longer ... (like my 100 point morale system, where 3 numbers provide a barrier between High, Normal, and Low Morale, and Panic) - where units can have high morale and still have "less" morale than Champions, not visible by strength until the unit's morale drops to normal sooner than the Champion's morale.

So anyways, with this 100 point system you could have "default moral" for a unit to equal the morale they enter a battle with w/o any modifications. This default morale could be used as a limiter for who can go into certain dungeons.

Kind of like "you must be this tall to ride the dungeon." And units below this level can still be pushed into the dungeon (if they can fit) at a 50% penalty to morale (cause they are scared/ unprepared/ etc). Meaning that a unit not "qualified" for a dungeon will never have high morale ... more likely low morale. A hard cap is fine, but would probably prove too limiting. The point is that it would (should) be far more cost effective for a Champion-only user to fight his way through the dungeons (if there is such a thing).