I’m a player who tends to take games somewhat seriously; I’m someone who tends to think about games when I’m away from my computer as much as when I play them.  When I play I’m constantly testing things, trying to find optimal ways to play, as well as come up with new strategies.  This is fun for me.  As such I tend to play on higher difficulty settings and I’m worried that the combat system in this game along with the way difficulty is typically implemented in this genre could mix to become a disaster.

Let me explain my thoughts; this entire game to me seems very well designed, and I get the impression that there has been a lot of thought and care that has gone into the combat system and that every unit has been made with every other unit in mind.  This has lead to something I consider to be a very fun, delicately balanced combat system.

In this delicately balanced combat system every unit seems to have a counter.  If your opponent has knights you can deal with this problem by attacking them with pikemen.  Pikemen are a unit which has a low overall combat strength but they receive a large combat bonus when fighting cavalry.  If your opponent has pikemen a natural counter to this would be to send in a strong melee unit such as great swordsmen.  Since they have a higher combat strength than pikemen they are naturally fit to end a pikemen problem.

Now let’s talk about difficulty settings.  In this genre I.E 4x and strategy games as a whole, the way difficulty is usually implemented is simply to give computers bonuses, bonuses to production, technology, and unit combat strength.  This difficulty model is a reasonable one, since it does increase the challenge, and given how expensive, difficult, and time consuming, making a competent computer player can be, it makes perfect sense why so many games use this model.

Now let’s get to the point, why did I spend a morning making such a long post?

My concern is this:   the reason why pikemen are made is to counter knights, If all computer units have say plus five or even ten combat strength, pikemen will no longer be an effective counter to knights.

Imagine this, it’s the medieval era and for some reason or another you are at war with your neighbor.   You’re in a battle and they have knights.  This shouldn’t be a problem for you, since you were prepared and brought pikemen to deal with their knights.  You send your pikemen to attack their knights.  Problem, their knights have more combat strength than your pikemen do.  So what happens?  Your pikemen attack and the result of the fight is their knights lose something like 30% of their health, and have 70% left while your pikemen lose 100% of their health and die.  If this is the case, no one would ever attack with pikemen or even great swordsmen for that matter as attacking with melee is no longer viable.

Instead what would happen is people would simply build as many ranged units as possible since they take no damage when they attack something, along with maybe one or two melee units, or as they could more accurately be called “meat-shields” to block cavalry charges.   

My concern is this:   if higher difficulties in this game give computer players combat bonuses all of a sudden half of the units in this game are no longer going to be used, and as a result of this, hard mode will be a new, and completely different game then the one being carefully developed right now, which in my opinion is very damaging to the fun of this game.

In my opinion science and production bonuses are adequate to increase the difficulty, since they would indirectly make combat harder.  Higher production means more units to fight, and better technology means better units to fight.  A good example of a game doing this is Total War Shogun 2.  If you play it on legendary difficulty, you are constantly given incredibly hard, but winnable battles against armies that are simply much bigger than yours and the result is an extremely challenging game that still plays the way it was designed.

TLDR This game is special, and I’m worried that if difficulty isn’t handled properly it could be devastating to the experience of passionate players, and the result would be low player retention, which is EXTREMELY EXTREMELY bad for the game’s long term health.

We're thinking about some combat modifiers for difficulty, but these would be small modifiers (+5 would indeed probably be too much) that only kick in at the extreme ends of the difficulty settings (in favor of AI at the highest difficulties and in favor of the player at the lowest). We're still testing this, though, so I can't give any definite answer.

There are definitely some balance issues but I would like to note that a huge part of balance is also the cost of the unit. Yes, knights are unequivocally the most powerful non-unique unit in their era, but they also cost like 5x more than most units. If you attacked me with 8 knights and I fought back with 20 pikemen, I'd win and would have spent less industry doing so.

Eulogos wrote:

There are definitely some balance issues but I would like to note that a huge part of balance is also the cost of the unit. Yes, knights are unequivocally the most powerful non-unique unit in their era, but they also cost like 5x more than most units. If you attacked me with 8 knights and I fought back with 20 pikemen, I'd win and would have spent less industry doing so.

Indeed! in addition to the production speed/cost, knights also require Horses, limitting the production cap as well.

I was wondering how, based on opendev, you would rank the AI capability in battles? I initially focussed on the battle mechanics. After getting to terms with those, I felt the AI is probably not that good at managing its units. For instance in the hold the fort scenario I was wondering why the Khmer keep running around so much with the elephants instead of using them together with the crossbowmen to penetrate specific parts of the Oxford city defense. Moving the elephant around did not really come with any clear benefit and it allowed me to take out the opponents one by one without too much trouble. Did you see similar at times puzzling AI behaviour? 

Eulogos wrote:

There are definitely some balance issues but I would like to note that a huge part of balance is also the cost of the unit. Yes, knights are unequivocally the most powerful non-unique unit in their era, but they also cost like 5x more than most units. If you attacked me with 8 knights and I fought back with 20 pikemen, I'd win and would have spent less industry doing so.

I’m very sorry.  I miscommunicated, I never intended to convey that I thought knights were overpowered.  I actually believe that knights historically were extremely strong military units and I’m glad they are represented as such in humankind.

What I was trying to say in my earlier post is that mechanically melee units are worse than ranged units.  If a warrior attacks another, they both take damage; while if an archer shoots a warrior, the archer takes zero damage.

With that in mind, my fear is that if the computer’s units get a flat boost to combat strength, melee units will no longer be usable, as in, if you were to attack any melee unit with another, yours would do very little damage and die in the fight.  If this is the case, then the only time anyone would ever build melee units, knights included, would be to defend ranged units such as archers.

While I want the higher difficulties to be a challenge, I also want to still be building units like pikemen and great swordsmen.

I feel like in higher difficulties of Civ, while certain units such as roman legions and Civ 5 infantry are very good.  I feel that for the most part melee units as a whole are an endangered species.  My hope is that if they want to give the computer combat bonuses, they’ll do it in such a way that it’s still reasonable to attack with melee units.

reich238 wrote:
Khmer keep running around so much with the elephants instead of using them

Owlexander wrote:
would move elephants off a hill for seemingly no reason.

There's an important question regarding that: Did the elephants shoot before moving. What I have seen the Khmer AI do is move the elephant onto a hill and shoot, then next turn use their special ability to shoot from the hill before moving out of range/LoS, and start this process again the next turn. But it is of course perfectly possible that in your case, the AI did not use the elephant like that and just ran around.

Yeah, good point, I think they indeed mostly used their elephants for shooting. To be more precise I think the puzzling behaviour was that they did not concentrate their fire on specific units. As it has been drilled into us that it is better to finish off units than to damage them, I think the target selection instead of the movement is the riddle here.  

The-Cat-o-Nine-Tales wrote:

reich238 wrote:
Khmer keep running around so much with the elephants instead of using them

Owlexander wrote:
would move elephants off a hill for seemingly no reason.

There's an important question regarding that: Did the elephants shoot before moving. What I have seen the Khmer AI do is move the elephant onto a hill and shoot, then next turn use their special ability to shoot from the hill before moving out of range/LoS, and start this process again the next turn. But it is of course perfectly possible that in your case, the AI did not use the elephant like that and just ran around.

That’s exactly what happened!  The elephants would move, attack, end round, attack and move back.  I got the impression that this was arbitrary rather than a tactical plan because:

1) When they moved back, they moved west, downhill rather than north uphill, where I would have had a much harder time attacking it.  Throughout the opendev I noticed that the AI seemed to understand how important terrain was when they attacked, but they didn’t seem to be averse to ending a round next to a hill.  At first I thought the AI, simply acted arbitrarily, but now I’m wondering if I simply didn’t understand what I was seeing.

2) For some reason I was under the impression that elephants had 3 movement.  If they do have 3 movement, shouldn’t they move, attack, move back, end round; instead of move, attack, end round, attack, move back. If I’m wrong and they only have 2 movement, then this makes sense.

Owlexander wrote:
2) For some reason I was under the impression that elephants had 3 movement.  If they do have 3 movement, shouldn’t they move, attack, move back, end round; instead of move, attack, end round, attack, move back. If I’m wrong and they only have 2 movement, then this makes sense.

As far as I recall, they actually have 4 movement, but their ability only allows them to reverse the order of actions, from the normal "Move, then attack" to "Attack, then move." It does not allow them to split their movement. It's an easy mistake to make given the ability description and we should probably improve that.

But yes, we can probably still improve the movement priorities of the AI, but it's difficult to say exactly what needs to be tweaked in this case without a save or video.

Ah, ok thank you very much for explaining that.  I really appreciate all the effort, you personally as well as Amplitude as a whole, are putting in to staying in touch with the community.  It’s a very good feeling.

 When the game does eventually come out, please include an in-depth tutorial.   What I found myself doing to try to understand everything was to go back to my experience playing the original Rome Total War, as well as Total War Shogun 2, as I felt like combat in the game was somewhat similar to those.  While I was able to do ok, I’m very worried someone who’s never played a real time tactics game will get frustrated and quit rather than learn through trial and error.

With all of that said, the combat as a whole feels really, really good in this game.