In last week's video blog, CT talked about trying to encourage "emergent gameplay", that is to say that the mechanics of the game promote things like taking the higher ground, flanking, spears versus cav. etc.
I have seen multiple threads about "flanking give 10% bonus damage" and the benefits of Rock-Paper-Scissors (or Lorry-Slime-Snail) balance. The trouble with these arbitrary bonus damages is that it feels strange. Did no-one else ever wonder why Pikes decimated Paladins, but got screwed by Militia in AoE:2?
I propose some ideas to create emergent behaviour which make the game more coherent and provide more selling points.

1) Range units on hills do more damage and shoot further.
Given that all projectiles are simulated more range on hills should not be a problem. In addition arrows do a set ammount of damage and an additional ammount dependant upon their speed. Simply put, if you fire an arrow of a cliff, it has more time to accelerate under gravity and thus will be going faster when it hits the ground than an arrow fired accross a field. This way the game uses the arrows speed to modify damage accordingly, creating an emergent reason to put archers and catapults on high ground.

2) Cavalry and Spears
Standard fantasy RTS balance dictates spears > horses. Sticking with this theme, emergent behaviour could be created that cavalry do additional "bash" damage dependant upon their speed (ideally most of their damage should be dealt this way). Additionally cavalry slow down when they take damage and spears offset damage from the unit (they effectively have a short ranged attack). Thus is cavalry hit a unit of swordsmen, nothing slows down their impact and they do full bash damage. However if a unit of cavalry charges into spearmen, then the spears do damage to the cavalry, slowing them and reducing their damage, meaning that the spearmen can now overwhelm the cavalry (this also means keeping cavalry static is bad).

3) Uphill and Downhill charging
Similar theory, if all units did "bash" damage on the charge and moving uphill was be slower than moving downhill, you get emergent behaviour where charging up a hill is not as wise a move as charging accross a plain. This means that hills are providing defensive bonusess emergently, rather than arbitrarily.

TL:DR
I argue (badly) that it would be good to take advantage of the physical simulation that the MOHO engine does to create emergent balance rather than arbitrary.

I really like your idea! Far better than the usual RPS system. It would need some serious testing first but I think this could really bring fantasy RTS games to a whole new level.

If you would make bash damage a percentage of the whole dps you would even get a nice additional way to balance units.

Wow, very nice! Good use of the physics and simulation to add depth! I don't know if they can do the physics of so much melee interaction though? The only melee damage in supcom was the crush damage on massive units.


Send an email to Chris with your initial post in it and the concerns other have about how it would work and see if he talks about it in a blog!

The funny thing, cavalry only does bad against pikemen in real life when the pike men are set for the charge. And even then, if the cavalry know they are going to do that, they can quite often maneuver around them even then. So yeah, pikemen did not automatically win in real history, just if they had a smart leader.

So the pikeman should have a stance/formation you can set (like you will be able to do with sheilded infantry) for horseback engagements where they either lose speed in return for better anti-calvary attacks or they become en trenched where they are to counter horses?

I see an actual use for the formation painter now!

I fully agree with your idears. well thought of, good sir.

I like #1, and while the others aren't bad ideas.

Adding a secondary damage variable to ranged units based on their speed is kind of emergent but you then have two variables.

If you base ranged unit's damage to be purely based on speed, from that simple rule you could get complex behaviour, such as controlling hills to gain range and damage bonuses, not because you designed it that way indirectly or not, but because that is the result of ballistic projectiles. p.s. due to ballistics, airborn targets will take lower damage than ground targets, due to aiming at a target higher than them.


several birds with one stone.

Similarly, if you tweak unit turn speed, their weapon damage, rate of fire, and range, you can get a nice complex behaviour set between units that aren't programmed into them.

I made a post about a similar idea a while back..
viewtopic.php?f=49&t=43976
Now that we've seen a bit more of the game, I doubt it will make it in.. But I suggested a 'stance/formation' ability that would (almost?) completely remove the mobility and DPS of spearmen in exchange for a massive buff to their defense. It was part of a bigger factional idea for a 'human' group that was individually frail so they had to depend on their massive formations of infantry to shield special damage units (like archers, mages, etc.). And they would have to move across the battlefield in a chess-like leapfrog pattern because if they were ever caught outside their defensive stance the enemy could cut through the disordered pikemen and kill the important units in the rear.

This 'mobile walls' style of play would contrast them from more dynamic/flexible forces of fast cavalry and beasts and whatever else.. and it lends itself to the cool mega units that can break through their formations with the 'massive' shove pathfinding.

Here's an idea; if we're going to let kings mount horses/etc, why not all units? Rather than produce set cavalry units, you produce the mount and the rider separately. Riders are just your normal everyday foot soldier units like archers. (but we can still have 'rider' units, like knights.

Once joined together, the cavalry unit has a hidden stat called... something, which represents the rider's stability on the mount. Taking damage causes this stat to reduce and return to normal quickly. If the value reaches zero, the rider is forcefully dismounted.

Riders and mounts can take damage separately, meaning you can kill the mount without harming the rider, and vise versa.

These are all interesting ideas, and I'm a very big fan of emergent gameplay myself..

But, I'd just like to point out that if you're designing a particular feature to produce a specific set of effects then it's somewhat debatable whether or not it should be called 'emergent'. But that's just semantics. The real point is that if we know the specific effect(s) we want then we don't necessarily need to use, and probably shouldn't use, the more complicated way of getting that affect.

This 'bash' bonus based on unit speed isn't part of the game physics any more than an armor multiplier, right? It's just a variable that gets applied in a particular situation. I'm not saying it's a bad approach, and I like the way it could interact with unit speed up/down hills.. But I just want to suggest that if GPG could do these things through a more 'artificial' and scripted manner, and if the artificial approach made the calculations more efficient, then that's what they should do.

For example:
Total War games apply a series of plus and minus bonuses to the two basic unit stats, attack and defense, based on a number of factors. In some sense this may seem very artificial.. It isn't really the result of anything 'emergent', but it works very well and produces much of what you're talking about.

They apply simple bonuses vs various targets, a +3 attack charge bonus for units marked with the 'charging/running' tag, a +2 defensive bonus for spearmen when they have 2+ rows of spearmen behind them, plus and minus bonuses/penalties for units depending on their morale, and a lot of other things..

You might call the interactions between the morale and defense and flanking systems emergent, since they the sum of their simple rules can have complex effects. (e.g., tiredness/condition affects fighting strength, fighting strength affects success, success affects morale, and morale affects nearby unit morale. If a unit routs then units nearby suffer a morale penalty, and the nearby units may be attacked from their now exposed flank. Flanking units get a +5 bonus to attack, which increases the damage they do, and shield bonuses aren't applied to rear attacks, and a unit that suffers a lot of casualties will take a big morale hit, so units that get flanked and rout can set off a chain reaction, causing once-confident units in the middle to rout because of all the mayhem).

I'm just saying this because fans of GPG and Supcom often seem to get caught up in the glory of the 'simulation' side to their gameplay. I happen to love it myself, but we shouldn't be afraid to have GPG use lots of intelligent shortcuts either. In the end, simple armor modifiers and "I'm on a hill so I get a bonus" bonuses are part of a simulation too.

Here's a link to the complex (and imho brilliant) set of combat modifiers used in Medieval Total War. I just found this.. wish it had existed back when I played the game.
http://forums.totalwar.org/wiki/index.p ... bat#Morale

This idea is made of win.

And Pinguin, I'd say emergent gameplay is moreso about sets of basic rules that apply to a basic thing the unit does which allow the player the freedom to utilize it in any way they can imagine instead of only in several scripted ways.

Take the bash idea for example, if you just apply the basic principles that a unit gains speed going downhill and does damage to something it may hit after having enough speed, you could have a scenario where a player finds a very tall hill next to the enemy base and starts flinging down hundreds of cheap infantry that would normally do next-to-no damage. But since they're on a very steep hill, they may end up being more like human wrecking balls that smash apart hovels and barracks alike.

Or perhaps the arrow speed affecting damage is implemented. A player might position their airships at the top of a very tall mountain so that they take even less damage due to the extreme angle the bowmen have to fire up at. There's no need to script, "if airship is on very tall mountain, +2 defense" it just happens because the basic framework is already in place to let it.

That simple framework of arrows being speed-dependent already accomplishes things like having units ontop of walls stronger than ones underneath them, or tall hills becoming fortresses, while still allowing other intriguing uses of the system.

You've never played Dwarf Fortress have you.

For the sake of completeness, I feel it prudent to point out that while Dwarf Fortress (particularly as of the latest major version) does use weight, and thus momentum, as a means of determining weapon power (along with other things like sheer strength), there are also a lot of random dice rolls involved, mainly for determining whether a weapon or projectile hit and where it hit. While this shouldn't discount the system as a whole, it isn't quite as natural as it may look at first.

For the most part I was talking about the world, what it's made of, and how it can be interacted with.

Two simple rules that can have interesting effects when players divise a means to use them would be water and magma. Magma that comes in contact with water hardens into a tile of obsidian, a very basic rule of the game. There is also a rule that has any unsupported tiles fall to the floor and crush/hurt whatever is underneath. Two basic rules that don't at first glance seem related, right?

Then some clever players found a way to get a finite amount of water and magma to meet in midair above a designated hallway. The magma hardens into a tile of obsidian as intended, and the new tile of obsidian (which is not supported because it hardened in mid-air) falls to the floor and hopefully hurts the bad guys underneath it.

There is no way the creator of DF predicted such a use of the system, but because the basic rules are not based upon a set of specific conditions and reactions, (except maybe the water/magma thing, but you get what I'm trying to say here) there doesn't need to be any pre-emptive scripting or coding; the game already supports it.

Same with countless other things in that game, like luring magma men onto a tundra to harden them into a stone statue, or boiling forgotten beasts composed entirely of water and letting their vapors drift away.

EDIT: I'm sorry, but this was so offtopicly ontopic that I felt I needed to show it:
http://www.bay12forums.com/smf/index.php?topic=64447.0

They're designing a computer, using simple mechanical devices inherent in DF, for the sole reason of making a fully-solvable Rubik's Cube in-game. God I love DF.

Oh, and this has got to be the mother of all examples of emergent gameplay.

Okay, fair enough. Incidentally my personal favourite is the bridge-catapult.

As a general model, DF does show ways interaction can be made out of simple rules. Isn't that a selling point of SupCom (and TA) too?

Well yes, of course, and hopefully KnC. I like the speed idea. Really neat. I'm not sure if having it make armoured soldiers turn into human wrecking balls (as opposed to just killing themselves and not doing much damage) would work, or even be desirable, but the other consequences seem believable and desirable.

K-lord: you should write that to crackedout .

I have indeed played DF. Was that directed at me? If so, then I think you missed my point.

Some of mine may not be practical for a RTS game but I'll just put them out there, see what people think. I was revisiting a few accounts on the battle of Agincourt last week, and a few things came to mind from further reading today:


Could there perhaps be areas on the map with muddy terrain that reduce the speed of heavy units? This might affect how and where battles are fought, and encourage exploitation of certain areas of the map that would otherwise be ignored.


Units passing over corpses (before they disappear) could incur a speed penalty. 70% is a bit extreme, but it stood out to me as I haven't seen something like this done in a game before. Would that be too annoying for people? I personally love a bit of chaos in a medieval war, I'd love to see charging infantry trip or stumble over dead bodies, but that might be too costly to pull off.

I like the ideas posted in the OP. It'd probably need some reworking of the system to pull off. As it stands, GPG currently has weapons set up so that projectile speed and weapon range is hard set with the unit and projectile BPs. What they should do is get rid of range and speed values in the BPs and replace them with "mass" (weight) value in the projectile's BP and weapon firing "force" in the unit's BP. Then, upon game initialization, it can calculate and store projectile speeds and weapon range (for level, flat terrain) based on the projectile speed, 45 degree angle, and gravity. Then all they need is some sort of quick formula (if quick is possible) that can recalculate range on the fly with height variances based off the "normalized" range.

I also like the ideas posted above by Winter-Dragon.

I think this system would encourage those sort of emergent phenomena. If the spearmen are spaced out or disorganised, they will not slow down the charging cavalry as much as if they were in phalanx. This also promotes flanking as cavalry slamming into the side or rear of a spear formation will do far better than a unit of cavalry in a head on charge.



I'm thinking spears on the outside and double-handed weapons in the center. Spears slow and trap, massive axes kill?



I think you make good points. When I thought of this I was trying to create the effects, but thought about how these effects occur IRL. Arrows do more damage and go further if fired from elevation because of ballistics, MOHO does ballistics anyway. Cavalry charges are dangerous because of the additional momentum from the mass and speed of the cavalry. Barring biological/exotic varieties, I think that the idea of any weapon is to transfer into the enemy more energy than their body can handle, be it an edge, a blunt face, a spike, fire, lightning or some teeth. The system I came up with, by no means perfect, there are good ideas to improve it, tries to recreate the "over-load with energy" weapon philosophy. I don't think that it would be very complex to add, MOHO calculates unit mass, velocity, location etc. From those you can pretty much do any Newtonian mechanics you want.

Supreme Commander had its own sort of Emergent behavior, such as using TMLs to shoot down Soul Rippers.

Or suiciding CZARs to kill commanders.

Or using Strategic Bombers to kill Czars, using huge Scout clouds to deflect nukes, using Novaks to rape any nuke or game ender, etc etc etc.