logistic tech and immersion

The reason unit "merging" is not supported is very simple, a stack is simply a unit with an extra number. So a stack of peasant basically has the stat of a peasant and a number, say 12. All other stats are derived and never kept. If you support "merging" you have to keep track of the individual unit stats separately. So a stack of 12 units would compose of all the stats for 12 units, even if they were identical. This, in turn, leads to performance problems, since the game has to keep track of many more units/stats than it needs to (up to 12x more). You would also have to implement some form of targeting priority for individual units (to determine which of the 12 will die first). In the end, it's just much simpler to go without all this trouble, even if it breaks the immersion for some.

i suppose you are right, it would be easy to just add it all up and average it though. 

What if one out of 8 guys has an item with a clicky buff?  Do you get 1/8th of the effect or does it only work 1 out of 8 times?

What if one has a bow and 7 have swords?

What weapons/armor do you display for the unit?  The squad is multiple identical units and with the fractions there probably won't be a combination of equipment items that results in the exact average stats.

The only way that could work, is if you sent unit's "back into the barracks" in a city, effectively disbanding the unit.  The barracks would only save the combined XP value of the so disbanded troops.
An 8 guys unit with 20 XP puts 160 XP into the barracks' XP pool.  (or maybe 70 % of it... whatever)
Then, when you train a new 12 figure unit in that town,  it gets 13 XP to begin with.  (160 / 12)

That is easy enough to manage.  Only one integer to save per city.
You wouldn't completely lose the old and experienced units - only their gear which really is too much of a headache to deal with.
The old gear is worn and used.  Use it or lose it.

Different / improved military buildings would give new units a larger withdrawal amount, so they might get a max of 10 XP from the barracks but up to 20 starting XP from a fighter's guild. (provided there is enough XP accumulated there)

Abstract systems are not necessarily worse.  Your experienced soldiers return to the barracks to train the newbie soldiers!
In fact, this could be expanded so the most experienced unit that you disband there becomes the virtual drill instructer and all new units automatically start with 25 % (or whatever) of the instructor's XP.
A reason to retire experienced units from active service to form a training cadre...  anything unrealistic about that?

And again - it creates a choice. Use your experienced units in the field or use them to create the cadre of your future army.
Now we're talking strategy.

Spinning that thread a little further:  If such a city is attacked, a militia would be formed depending on how many figures from disbanding military units went to the barracks.
A second integer to save per city. We're getting excessive now!

It won't be some crack elite force but enough to drive off the odd wolf or bear.
This reduces the micromanagement of garrisoning but by itself won't stop or even delay an attackig army.
And you'd need barracks for that to work so it doesn't give you auto-defense when spamming lvl 1 cities all over the place.
And to get any such militia, you must have trained troops before and disbanded them, losing their utility as fighting troops.
If someone were to get flashy with such a system, the equipment quality of disbanded troops would somewhat reflect on the militia units.

Reminds me of Lords of Magic where you could have your champions stay in a recruitment building and "train" the people there, allowing you to recruit units at a higher level. The deference here is you must disband, rather than having them sit around.

Well, yeah. And MoM had champions with the Armsmaster trait.

But so far, all suggestions about "combining" troops were a nightmare in the implementation.  This system is simple enough to actually work.

you know it's funny you mention the problem of different weapons, i assumed you would have to be combining the same troop type. I wouldn't have worried about mixing apples and oranges. I like your suggestion for the xp, i would have thrown out the baby with the bath water.

I would not have a militia be generated by these training troops, I would consider that the risk of taking the time to train the troops into a new cohesive unit. Don't lose the town/ outpost where they are training. It's more about recycling the xp and the time you originally invested into training the original troops. I imagine the retraining time to be the difference of the time it takes to train 4 individual soldiers compared to the time it takes to train a pack of four soldiers, then add a turn or two.

edit * changed my bold into an underline i didn't see the bold that well.*

and that's as simple as i can imagine it to be - hence no militia rule that you suggested ( that just makes my head hurt) Don't get me wrong, in real life those soldier's would take up arms to defend that town i'm sure - but in a tun based game they would have to count as being in the factory and not available for the fight, what happens then if some of your 4 soldiers die, then the construction queue crashes.

Like I said it was just a potential avenue to further integrate the whole feature into the game.
Doesn't have anything to do with the main idea itself.

It would mean that your soldiers don't simply vanish and are reduced to a single number.

surely not too difficult to just limit it to identical copies of the same unit? galciv 2 did it like this with fleets.

but a good stop gap solution would be to give everyone squads for free at the start of the game. everyone always tells me to reasearch this first because it's "essential." to my mind there should be no essential techs. if everyone had squads from the start and was forced to use them for troops there'd be:

- less need to retrain units when you got new techs (because there'd be fewer to get)

- faster battles because there'd be fewer units on the map (i don't know why, but the ai doesn't seem to use many squads)

- more space on the battlefield for the same reason

- more "epic" look

- ai settlements would be better defended because the minimum unit strength would be higher

- slightly less need to focus on the warfare tree

of course, the numbers might need tweaking to make  this balanced. especially to make heroes competetive against squads. but arguably this needs to happen anyway.

What would've been nice is, instead of the huge jump from 1 to 4 to 8 man squads we had a gradual jump of 1 to 2 to 3 to 4,5,6,7,8. where each level would not be so huge as it is now, also a similarly armed 3man squad, for example, should have a decent chance agains a similarly armed 4man squad, say (1-N) where N=3to4 odds or 1-.75 = 25% chance of winning.

It shouldn't be so hard to be able to combine singles into a larger squad, if its a performance issue the computer can "erase" the four singles (as it does in combat) and substitute a 4-man squad.

I also agree with some other posts in another thread that if an individual guy dies in a squad that squad is permanenly reduced: i.e. one man dies in a 4man squad so the squad now becomes a 3man squad, of course it should still be possible to add a single guy to a 3man squad as above.

The game application should be capable of dealing with single guys that are armed differently too, by subtracting or adding to the gold/resource cost on the basis of what the single guys already have or lack.

there would need to be 7 more squad techs to support the "gradual" which is a good thing.

2 is still twice as powerful as one though. better to start with 4, then 6, then 8. or something. personally i think that more "obligatory" warfare techs is a bad idea. people already focus on it too much anyway.

There was no merging into larger units in GC2. Every single ship stayed one ship forever and could never upgrade to another base hull.
And upgrading was an integral part of the game, meaning that you could upgrade a ship anywhere on the map. You could also upgrade all ships of Type X in service to Type Y.
With the multitude of resources required for military production that system wouldn't work very often for WOM.

If you mean that Soldier v1.12 can only be merged with identical Soldier v1.12 units, then I think that's a bad idea.
What do you do if you only have 7 of them and the obsolete "design" is long clicked off?  You'd have to destroy 3 of them.

And even if you had 8, you'd have to micromanage them to move them all to the same place on the map for the procedure.

With my system, you'd get "the spirit" of these soldiers back if you only moved them to some city with barracks.
They'd become part of whatever unit is built there next.

There is good micromanagement that lets you refine tactics and orders... and bad micromanagement where you have to give units excessivly intricate orders to do simple and boring things.
"Retire unit at nearest barrack" can be a one button military ability - once the strategic abilities are no longer inaccessible.
You could still refine that order if you first send the unit to a location where it is closest to the "nearest barracks" that you prefer.

but a good stop gap solution would be to give everyone squads for free at the start of the game.

I tried to find a fault with that suggestion but couldn't.
I can see where "large" military formations would require some thought (aka research) but it doesn't take a military genious to get four guys marching in roughly the same direction.

True, but you can say the same thing about 4 & 8 man squads: a 4 man squad is 4x more powerful than a single .. but still it's only twice as powerful as a 2-man squad, giving some leverage there. It should not by an "auto" win for a 2 to 1 odds (assuming squads are similarly armed). even at 4 to 1 the single should have a very minute chance of winning, say 3%. "Obligatory" techs would not be so obligatory if 3man squads had a reasonable chance against 4man squads (as it should be), or 5 vs 7 for example, thus some players may opt to research up to 3man & then research something else whereas others would go to 4 before researching magic, for example. There's also a cost factor here, a 3man squad would be a little cheaper than a 4man squad, which would tie more nicely to an empire's resource availability & gold revanue. The cost difference between single guys and 4 or 8 men squads is *huge*.

The real problem with the combat system is simply the way it works: moving first as the attacker is wayyyyy too powerful. I always kill 90% of opponents on the first round of combat: between spells, archers, summoned monsters with boulder blast & fireball. The human player is unbeatable. The turn-based combat system in this game is broken. There needs to be either a radical change (such that the attacker cannot attack all at once), or a RT tactical system needs to be implemented which is resolved on "auto" & players have no control. I'd rather have that than the broken, un-loseable tactical combat system we now have.

I had this battle last night that was unbelievable. I declared war on both resoln & umbar who bordered each other. There was this massive group of resoln on the border which also happened to be next to a mountain chain. I moved one of my uber groups on a hill-road and attacked the massive resoln group on my 1.5 mp. At first I thought there was a game bug, because I attacked and there was another group, attacked again, another group .. I literally went through 30 minutes of combat wiping out a couple dozen resoln groups that were stacked up against the border of umbar & the mountains. I realized that what must've happenend was that these resoln groups had gotten "stuck" somehow and could not move. It was unbeleivable. There was a huge patch of very nice uncolonzed land with gold, crystal, food resources, ancient temples on the other side of the mountains, etc.. So all these guys (which included pioneers) were unable to move past the mountains & border of resoln because they got "stuck" somehow and the AI was not intelligent enough to circumvent such a basic obstacle as a terrain/border block. In my last game the sovereign of yithril got "stuck" in a 2-square area between my zone of influence and tarnith's. It was "stuck" there for more than twenty turns (didn't cast teleport, which I made note of). Eventually they "disappeared" when the zone of influence grew and the 2-square pocket disappeared.

Guess what I'm trying to say here is that this is a much higher priority from my POV than diddling with squad/types .. I mean getting the AI to actually "use" its resources in a halfway intelligent way. As I noted, this is not the first time I've seen the AI get itself "stuck". I've been able to manually cause the AI to do this with strategic positioning of a few single man scouts. You can block up a whole AI pioneer platoon with a couple of well placed scouts. The AI just keep builiding them & they go to the same square. You can literally block up dozens of AI pioneers this way & the AI town just keeps churning them out over & over turn after turn. It's funny in a way but also sad seeing such ineptness in a game AI. 

The way to win every battle in this game is simple: attack first. That's all you got to do. All you need in a group is a mage or two, an archer or two, and three or four summoned units preferably earth giants with boulder throw, and one maceman group. Don't even need an uber stack, as long as you attack first you've won. You need to make sure the opponent doesn't attack you first, which is easily done by having "wingman" groups. Two strong groups is actually better than one uber group IMO.

You seriously underestimate performance of modern PCs. It does not really matter whether you keep track of 100 units or 10000, esp. in TBS. Even with most simple data containers, you will start noticing any performance decrease only when reach 100k-1m ranges of items. Just look at the way Dom3 handles it, with keeping track of age, wounds, exp and all the stats of each of those thousands of soldiers you employ. It added a nice twist with most elite units starting very old and with high chance of them dying 20 turns later from aging.

this kind of tweak was never meant to fundamentally change the game, there are big tweaks that take a long time, such as ai, and little ones that take a short time to make, such as tweaking the way one tech works. It is just supposed to be a simple change to enable the crossover into bigger unit sizes less painful. small units take less resources and are faster to build than big ones. its easy to make single units that you can group together at your leisure when you want to go to war and take advantage of the defensive tech that raises unit hp  and minimizes the wages you pay out to your units.

     However, after having the experience of my high hp, high armour troupe  die to catapult fire  (60+ armour dieing to 25 attack  ) I am at a loss to how to fix that...

This might be true if Elemental runs flawlessly in its current form. But it doesn't (see all threads about lategame performance problems). Adding more stuff for it to keep track off? Seems like a bad idea to me.