Edit: It appears like all size/volume/quality step up/step down are 1.5x multipliers for strength.
Standard/Battalion/Medium are being dead center values.
Looks like you get it and could thus answer the rest of your own questions in that post. At a point it's almost impossible to imagine the highest end because that would probably have to cover more than the width of a plot and on the low end, we have something so insignificant it's not really worth calling a unit.
Currently LE units get XP for subtypes, such as land and melee. This is why many of my cities build stick training (or however the building is called, for +3 melee xp). I meant that they will no longer get this kind of bonus, while retaining the ones you mentioned.
They'd continue to get land and melee as well because Land isn't a combat class and melee is in a category of primary combat classes that don't conflict with other combat class labels most of them have. Now, Archery is in the same category so if we had a Melee and Archery combat class on a LE unit, then we'd only get the best of the two bonuses from those sources.
This is also an option, and should be good. Not entirely sure if it will be effective later on, but should help in the long run. There is already a tech limitation on the higher levels of the promotion, so there won't be a need for major changes.
How about +1/+1/+1/+2/+2/+3/+4/+5/+7/+10/+15/+20 each one opening up with a new era?
I was under the impression that the trained units always start with the same quality. I never actually understood the quality down promotion line either.
Nope. A Brute has a much higher quality than a Swordsman because of the raw savagery that man had to possess in the Prehistoric to survive. Man was also not as populated and thus the median group volume was much lower. As the unit upgrades, its base quality and volume can change and the system adapts. The quality downs aren't something you'd see in the game often but maybe a disease might be able to cause this? Or a poison that befuddles the mind? (another good counter method for the super horde that hasn't been completed yet.)
It seems like a unit's upkeep is about doubled every time it's combined in the suggested form, instead of rising slightly (20%). This does seem to be more in line with the normal version, as you train 3 units and pay for 2 (instead of train 3 and pay for 1.2). At the worst case, it should be 1.5 times of the lower tier, to keep the upkeep per power the same. (I would consider 3 times the normal so merging won't be a tactic to avoid upkeep)
I want it to be a tactic to avoid upkeep. Larger units are more efficiently organized - some support functions take some basic overhead to initialize the support services for each unit and as a larger unit. I just worry about actual upkeep impact getting too severe. SM demands more units be trained and it might eventually be necessary to create an alternative production speed modifier for units on SM games as it is. I don't want this becoming something that tooooooo overburdens the gold balance of the game. I would be willing to let us play with the numbers and the number theory but try to take a gentle approach rather than a dramatic one and see how even a more gentle approach pans out first.
Without a way to re-train them, it seems like a major hit to these units, and may cause issues down the road.
That's really the point. To make it much harder to reach huge merged and extremely trained groups. If you think about it, the flaw in our current system is that we're not factoring in the dilution rookie trainee replacements would be bringing to the units they join. You hardly ever see such extreme combat mastered warriors in large groups in the real world BECAUSE so many of their comrades have perished on the road to them becoming the biggest badasses imaginable. Thus in the game you'd start seeing high high quality units struggling to find others as skilled to merge with to get their volumes up. This will help a lot to keep the super unit from happening often and would provide an ability for less built up civs to have some hope in facing such a unit when they do (try to at least do enough damage to get it to split!)
The actual mechanics may be strange, too. If split happens when below 50%, will terrain and artillery cause it? If not, will a single combat against half-dead unit will cause it to split, regardless of damage? It may lead to a lot of confusion why sometimes units are split.
Anywhere damage is dealt it would be possible for the unit, afterwards, to be at less than 50% health, at which point it would be split unless it is a unit that cannot be. Combat wouldn't be the ONLY way, and yes, other attacks like distance attacks could cause it as well. (Very important to maintain air superiority right?)
This will also mean that normal units will split into their lesser versions all the time, which are prime target for assassins (I think).
Yes, this would give the smaller, merge/split restricted criminals and strike teams and canines a more effective mopping up role that they were intended to have to begin with.
An idea to raise the price of losing a lot of hp- a unit that reaches red (say, 20% or lower) may gain a promotion that reduces its healing rate and combat strength (say, limit it to up to 5%/turn, herbalist included, and give -100% combat strength). That unit will now be either stuck in a city healing for a while.
The planned Critical Hits system might be a better reflection for this or at least it would be a major overlap. We also already decrease healing speed for higher group volumes to somewhat reflect the same thing I feel this would be an effort to reflect. Nevertheless, after some other balance steps taken I'd still be open to this.
Such a unit will get an additional option- rank down, resulting in relatively healthy unit without the promotion. This will get the unit back to combat a lot faster, but with the permanent damage of being a group size less.
Interesting... it gives the impression that the unit sorta leaves behind those too damaged to come back quickly, pretty much forcing the others into retirement and taking on replacement recruits (assuming the rank down is a quality step down), which actually does happen a lot in modern armies. I'm not against this but at the same time I'd like to see how the halvsies rule would play out. This might be a great way to allow a similar style effect for units that can't split.
This is more about training than merging, and it isn't related to my third point, so this is about standard size units. The idea is: If you train a squad unit (usually in Prehistory), even a clearing in a forest might be enough, but training an entire batallion there to be a cohesive unit is out of the question. You already get better training ground buildings during the game, but I think their main advantage over earlier buildings is the bigger area they provide. So a batallion-sized unit should perhaps require a dedicated training building, with even bigger buildings (and exercise areas) in further eras. They wouldn't ever be big enough (unless perhaps in Galactic+ Era) to house and train millions, but I don't think there is a human unit with that kind of standard group size, is there?
It sounds like you are talking about an element of the third suggestion you made, creating a limitation factor on how large a group size shift your city could choose to train. If that's what you're gunning for I think it's a good idea and would be easily done with a new building tag once the other system is in development as well. Would blend right in, provided we had the building review nicely established for such barracks building lines such that nearly every era (or perhaps every other era) had its own step up in that chain.
Unfortunately I'm rather inexperienced in Python (especially with the "special" Python that seems to be required), but I think you would need something like an additional drop-down box for either a text-field or (perhaps better) the group-size icon, after making a choice the unit list would need to be adapted regarding cost, strength and order of units - the last one because the non-merging units remain unchanged and might now be cheaper/more expensive than certain other units.
Then the pedia entry should get a dropdown box like the one in the city screen, unless we want
up to thirteen entries per unit there, and even then we would only represent group size - with two dropdown boxes we could show the strength of the units depending on both group size and quality (at least unit size is fixed
) - including their natural vision/invisibility properties.
I'm thinking of a simple up arrow/down arrow system with the volume shift # in the middle, so you'd see a +1 or whatever between the up and down arrows under the label 'volume adjustment' or something. Then when you hover over the unit info, it gives the unit as if the group volume was that many shifts different from the normal (0). This would cut down on valuable screen real estate a lot.
@Toffer90 : Do you think if we did this that this would be something we could find room for on the city screen somehow?
Cool... pm me your email address and I'll attach it to you on an email. Should be the fastest means to transfer. The cloud stuff would be the best way for us to work on things BUT it's kinda slow to process as you go with it and is thus a bit of a hindrance as well. Plus, this much info tends to stress google.
I see, then what about changing this:
I like it enough to have already just adjusted the code. You're learning how to get me to do things the way you mathematically want them done. I often resist certain suggestions simply cuz I would struggle to create the math for the effect suggested. By laying it out like that, it's all too easy to implement that good idea. This allows for some more extreme modifications without a complete bind up.
Sure, what about a 30% increment for both quality and group volume?
<iExperiencePercent> for group size -
- from:
80 ▬ 60 ▬ 40 ▬ 20 ▬ 0 ▬ -20 ▬ -40 ▬ -60 ▬ -80 ▬ -100 ▬ -120 ▬ -140 ▬ -160
- to:
120 ▬ 90 ▬ 60 ▬ 30 ▬ 0 ▬ -30 ▬ -60 ▬ -90 ▬ -120 ▬ -150 ▬ -180 ▬ -210 ▬ -240
Yes. With your suggested code adjustment this would work very nicely imo. Do you think these might be good numbers for the upkeep modifier? Maybe shifted down one rank across the board so we don't get into less than -100% numbers? I don't think the previous method would translate well for unit upkeep due to the way it tallies things up at a player level rather than a by-unit level.
0 ▬ 25 ▬ 50 ▬ 100 ▬ 200 ▬ 400 ▬ 800 ▬ 1600 ▬ 3200 ▬ 6400 ▬ 12800 ▬ 25600 ▬ 51200
1 ▬ 1.25 ▬ 1.50 ▬ 2 ▬ 3 ▬ 5 ▬ 9 ▬ 17 ▬ 33 ▬ 65 ▬ 129 ▬ 257 ▬ 513
?
hmm... I'm not really sure. Maybe you had it right the first time and I'm just finding it shockingly expensive at the outer ranges due to the multiplicitive effect. Your math is sound on that and maybe that's the way to go. You understand my reservations on allowing units to become this expensive to maintain though right?
).
Cause you know when you get 
fighting against the "Windmill" gets harder. 
