Techno and Magics

I wasn't talking about treason. In your post you said one faction will use technologies, and one would use magics. Suppose only the techno faction has firearms, this means that if you are in the techno faction, you will never face an enemy attacking you with firearms. Since anyone holding a gun is an ally. By the same token, if you are a magic user, you will never fight against a flag that uses magics. So the techno users will never need to think about defending against firearms, and the magic users never need to think about defending against magics. It may also mean that some artifacts are only useful for one side. Is this preferable?



Spies

On the other hand, double playing will indeed piss off many people. Probably won't implement that. Having spies alone is quite risky too, ideally, members of each faction should feel safe to talk about strategies. That probably won't happen, but at least the players shouldn't have to deal with the constant stress of being backstabbed.

So in this game spies won't be about deception. Probably would rather sacrifices this dimension to safeguard the overall trust and loyalty required for the Flag system to work. It is important that the players in the same faction can trust one another.

So 'spies' correspond more to 'scouts' than 'traitors'.

Spies having skill of both sides is ok, but if spies are not played as traitors, then it probably won't be necessary.



Berserker

Delaying the damages taken is a good implementation. A similar way of implementing it is to increase the Hitpoint temporarily when the berserker effect is active, and have the added effect taken away when the berserker effect is gone. In your way, the berserker is invincible while berserked, so that will probably motivate the player to do some more crazy stuffs (suicidal attacks, since after activating berserker, the character has a 30 second of invincibility)

---

Damage Model

Damage model is always a difficult topic. In reality, human is very fragile. most high resolution damage models create a downward spiral (i.e. after you got hit the first time, your performance decreases, making you more likely to get hit again).

Note that there are already ways to slow an enemy down using magics, and the various debilitating magical effects already do what a higher resolution damage model can do.

Another related question is whether the combat should gear more toward skill-based or stats-based. The more skill-based it is, the harder it is to use tactics and strategies. It also make it harder to evaluate strengths of the different flags. But each flags can keep a win/loss ratio, so it should be implementable. Currently, combat is a mix of skills and stats, but skill does not correspond to hand-eye coordination or click speed. Skill correspond to positioning your character, your choices of abilities, and your timing in executing weapon skills and combos:

Combos:

1) Your fellow teammate unleashed a special weapon attack on the enemy, with the right timing, you unleash your special weapon attack to combo with it to deal bonus damage.

2) The mages begin casting a huge spell that takes 15 seconds to cast, the melee commander jumps into the enemy, and unleashes an area of effect weapon skill to freeze them and make them vulnerable to ice magics for 3 seconds. During those 3 seconds, the magic spells that took 15 seconds to cast land and the enemies get nuked.

3) The mages begin casting in battle instance 2 between the city wall and the siege weapons. Catapults in battle instance 3 fire. The magics in instance 2 adds to the catapults shots and turn the shots into fire dragons, and direct them to hit the city gate in battle instance 1.

FlagA in Instance 3: Yo B, blizzard incoming
(Translation: enemy mages started casting the nuke 'blizzard' from instance 3 to instance 1, these are long range nukes, think battleship)

FlagB in Instance 1: we are ready, tell us when
(FlagB is going to counter it by using the combat mages to form a reflective dome)

FlagA in Instance 3: fire in the hole
(FlagB received the signal and the combat mages formed the seal. The blizzard shattered around the dome and dealt an AE damage to the enemies in Instance 1)

FlagB In Instance 1: OMG
FlagB In Instance 1: that was tight, took screenshot
FlagA In Instance 3: neat

These are the types of tactics that can be implemented with the current damage resolution. It already take some coordination and skills. Note that these tactics relies on a certain level of deterministic behavior. Too much variables may also make strategies unreliable and turn the battle into chaos.

So my current position is to assume that the character have the high resolution damage model, and see what happens. My hunch is that higher resolution damage model is equivalent to a larger set of combat abilities. (i.e. the weapon ability Legsweep that induces a special effect is the same as hitting the enemy on his legs in a high res dmg model). The difference is in a FPS, there are no such thing as 'abilities', the only difference is whether you aim at the torso or at their leg. By selecting different part of the body to hit, you are selecting the different 'abilities'. This make the gameplay more hand-eye coordination type skill-based.

and doesn't THAT sound you difficult to handle?

Well, ok then. I surrender, and follow you there.
Let's have coordinated attacks!!

Since it is so difficult to get a full 3D MMORPG done, what about adapting the design to suit a web-based text-based Game?

That post was mine. It seems to be a pervalent problem that the site logs you in as an AP.

Addressing foreseeable design flaws

At this stage, it is beneficial to think about flaws and pitfalls of other strategy games or Warfare RPG games. What problems do analogus designs have that should be addressed before beginning the prototype?

The previous post addressed these problems:

a) Cities far from a frontline become ghost towns
(being solved by the use technological and production levels, and their importance for defensive (adventure) style gameplay and level advancement)

b) Enemy being able to pentrate into enemy territory, dissolving the importance of a Frontline. This is a common problem in space strategies where the enemy can potentially wrap to anywhere, destroying the notion of terrain and distance.
(Being solved by non-increasing stamina within enemy territory, implementation of watch towers (radars))

c) It takes too long to assemble a battle
(Being solved by allow the characters to teleport to the Flags, and allowing the characters to join a battle after the battle has already begun.)

Maybe we could take a page from the book of other Text-based games?

I know for sure that Marcoland (www.tiipsi.com) handles them in no way. I've been paying it for more than one year, and still have to find a challenge. Or a RP element, at that...

The individual battles are handled by personal mana, and personal PvP mana (yes, they are separated). You are expected to use your skills and pray every day in order to advance the stats of your character faster. Praying is rewarded by random stat increase, smetimes negative.

You have PvP mana, which allows you to fight in PvP mode which obviously kills the opponent, forcing him to spend money and resources to revive before being able to earn XP, plus the fact that, if you log in earlier than others, that is right after reset, everybody is still alive but not active, and you have more potential targets.

You have personal mana, which allows you to fight with creatures, which rewards you with XP, money and scarce resources. Resources can be spent to increase mana total, to an extent...

But there is also the guild city part.

In this city where you belong you can use the healer, and other auctionning facilities. you can also have access to city run armoury, where you can borrow weapons and armor pieces.

The city wars are handled out of control of most of the soldiers. Only the top ranking staff members are involved in the decision of making war or not. once this has been decided, then everything is computer based, and the calculation is started. You almost instantly know if you lost or won. No strategy apart from two different styles of wars, including or not siege weapons and legions (groups). A Third option is caled invasion, and it is used in order to steal resources from the defending town.

As you said, each town has levels of XP. Every level, it is allowed to spend more resources to build different facilities, which in turn, wil provide different effects. Walls to increase defense, factory, which will provide different siege weapons depending on the level of the factory. Some other facilities have been proposed, but none has been implemented yet.

The biggest problem with this game, is that there is NO STORY to it WHATSOEVER. Therefore it lacks all RP elements to make it an enjoyable experience set apart from the grinding machine. Plus there is little to no strategy to make it interesting in the city wars part. It has a strong community nonetheless.



B the way, have you ever played a tabletop game named Confrontation, by rackham? It has some nice rules I though could be used here, in the case of a simulated interactive Massively Multiplayer tabletop game.

Each unit has a different size, and therefore its base is of three different standard size. This can be used to attack a single bigger unit with more than one smaller unit. And the more soldiers you have to attack one single unit, the more additional attacks you get. In fact, it is often simpler to kill a big creature with smal characters than a small character with only one small character, or a big character with only one big character. If we could manage to simulate these squares of different sizes, I think it could be used.

Also, the characters each have different abilities and different stats. Their speed stat tells how much they can run each turn, and how fast they can attack when in close quarters. Therefore, what you have is not Warhammer's slow paced player to player tur by turn action, but a rather more involved and faster action. Sometimes two or more characters have to act in the same second, and only a dice can decide which one begins. Yes, the more I think about it, the more I feel it could work.

There are also two other games I played on the net which were more interested in Big time warfare, which "demange", a game in which you played either an angel or a demon, and had to fight the other side in order to push them further away from yours. Unfortunately, the farther from your side, the slower you got, and the opposite was true. And since you only got TWO actions per turn, and that was a day, it highly lacked interactivity to make it enjoyable. Plus the flag element we discussed did not exist. Nevertheless, it DID have that aspect of multiple attacks being more effective than single attacks.

The other was almost a clone, except that you played in armies, and the map was more interestingly drawn. It had also more features. Possibly it was either the original work, or simply a fan dedicated to his game that decided to improve it on his own. Both communities were not particularly active, since there was little to no possibility to make it interactive with the game.

Exposing Potential Flaws

Grinding:

In BattleCry, Grinding is not necessary for enjoying Flag battles. Because there are Flags that the player can join that boosts the minimum level, and Flags that boost specific classes. So if you are an archer, an archer Flag will welcome you pretty much no matter what level you are, the Flag can probably boost you to a functional strength. The importance of Grinding is lessened. The overall system will encourage XP gain from battle instead of from Grinding, because obviously the player is taking more risk going into battle than killing creeps in safe, allied territory. So the XP bonus gained from battles will not only reward the player for taking risk, but also reward the player for collaborating.


Offline characters:

We already know the current implementation, but just reiterating this to highlight these implementations as solutions to potential flaws in other games. The proposed solution about offline characters is that they are represented by NPCs. So when a player logs on in the wee hours, Flags and targets are not unguarded. The implementation may make the NPC counterparts harder to kill than normal for defensive purpose. The Flags can't use the NPCs representatives to initiate assualts. There is an abuse that a Flag can lure the enemy to attack an NPC Flag. But it seems to be a reasonable punishment for stupidity.


Cities and Rations:

There will still be auction halls in cities. There might be separate armory for cities and for Flags. Individuals of a Flag carry their arrows, and the Flag also carries arrows. And the Flag restocks at cities. Weapons are done in similar fashion, but the most obvious resource to get depleted is arrows (and in the newer version, gunpowder and bullets). Flags with crafters can create supplies, but usually won't be able to continually supply the Flag battle after battle.


Decisions about battles:

Flags start battles. But Flags only exist because of the supporters, and any one can rise as a commander. So there is an interdependency, that a Flag is a representative of the will of the people. This is a subtle feedback system, because a commander cannot rise in power without supporters. You have to do what the people want, while adding your own vision and leadership to it. So the 'high ranking officers' are not those that bragging about their ranks making you jealous of them. They are part of you. You are part of them.



Story:

The conflict and warfare will have a story, but much of the actual story that the player will experience will come from his interaction with the Flags. In this design, the game player aims to provide many opportunities for COLLABORATIVE strategies that cannot be executed single-handedly. This is the basis of emergent dynamics and stories. The Flags are used to centralizes these events, so that the dynamics are significant, so that the players won't feel that they are just making stories that no one else will know about or doesn't affect anything else. It should be obvious that no one is going to care about a single player's story out of like 5000 players. But players will know the Flags and what the Flags accomplished. The players will know the Flags because they have fought alone side with other Flags and experienced how different Flags interact. Through the Flags the players meet. Examples of emergent stories:

You and you buddy Flag had been fight along side, in one encounter your buddy Flag was destroyed. You want to sneak in and res them but your buddy Flag told you not to, that it is too dangerous.



Many against one:

I haven't played Confrontation. For the current implementation, the game mechanics favors Big Flags against Big Flags, but at the same time does not favor the forces to concentrate so much into a few Flags (since everyone under the same Flag travel together). So in an actually battle, Big Flag vs Big Flag is better than a number of small Flags vs a Big Flag. However, strategy wise, mobility and uncertainty (for the enemy) is important, so the Flags may not want to move together until a battle is about to begin (you don't want the enemy to know where you are concentrating your forces).

Quote:

Each unit has a different size, and therefore its base is of three different standard size. This can be used to attack a single bigger unit with more than one smaller unit. And the more soldiers you have to attack one single unit, the more additional attacks you get. In fact, it is often simpler to kill a big creature with smal characters than a small character with only one small character, or a big character with only one big character. If we could manage to simulate these squares of different sizes, I think it could be used.

Since the Flags have a hierarchical structure, this is still true. Translating this to Flags, this is what it means:

It is better to fight a Rank5 Pure Cavalry Flag with a combination of different units than with a Rank5 Pure Infantry Flag. In terms of implementation, the difference is caused by the fact that units of different types can combo better than units of the same type. In this particular case, a mixed flag with archers, infantry, and engineers can access a lot more strategies and deal a lot more damages than a homogenous pikeman flag. For example, the mixed Flag may be able to use the archers to fire screaming (high frequency) arrows to stun the horse at close range to stop the cavalry charge, and allow the infantry to charge the cavalry instead.



Strategic Saturation:

This potential flaw is not mentioned before. There is a point when a RTS renders into a click feast after the strategies are pretty well known and there is no more to learn than to execute. How do you prevent this? Does it need to be prevented?

Oftentimes a Flag beats another Flag not because of strategy, but leadership and experience. One Flag may just run smoother than the other. The game should let such Flags to exist, reward them with strategies so well executed. So with good on-battle leadership and command, a Flag should be able to offset the strategical advantage of a poorly commanded Flag. And then there is the will and faith of the players, that may also play a part in the outcome. This is presented through situations such as when the commander got killed, the Flag needs to know to stick together, and believe in each other's strategy, instead of everyone doing their own thing.

Saturation can be lessened by providing more objecives, and interchangeable strategies for each given situation. And there is the whole dimensions of putting your strategy in disguise. Disguises is only possible if there are forseeable patterns for strategies. In general, this means that each strategy must have some unqiue (or distinctive) setup or requirements. And those setups should be perceivable by spies. For the example above, spies should be able to tell that the Flag is carrying a type of specially engineered thin material. The spies may not know what the hell they are for, and the enemy Flag may only learn by experience. It is a game of deducing what strategies the enemy Flag may use on you based on their composition and inventory.

(For spying, when a spy got 'a piece of information', the spy doesn't know what kind of information it is until he reports to a commander. This way, the spy must physically escape the enemy territory instead of just telling the commander about the discovery through chat. The idea is that the secret shouldn't escape if the spy got killed on the way out. A spy can gather more than one 'piece of information' before returning to base. For engineering and technological information, the information needs to be opened by a commander, and interpreted by an engineer of appropriate level.)

agreed on spying. It would make things more intense.

I've got a strange idea, this morning. What if you had different weapons, which, instead of dealing bigger damages, dealt damages against specific defenses? Like "Rock-paper-scissors"? If you add the range stat, it could make it more interesting.

Let's say that you have pikemen. Pikes are made to be used in the second and third ranks of your platoon. This is where they deal the more damage. In first rank, you need swordsmen with shields, to absorb most of the damage while the others in the back ranks deal with the enemy (phalanx strategy). But if a pike can touch at best at three squares of distance, and deal only 2/3 of max damage at 2 squares, it only deals 1/3 of damage at close quarters.

But an archer can deal max damage in a range between 5 to 10 squares of distance. Below this, damage is lowered, and beyond, it cannot touch. A Swordsman can probably deal max damage at close quarters, while dealing half or 1/3 damage at two squares distance. Axe-wielders can deal damage only at close quarters, but damages are massive, same thing with maces. There should be different kinds of protections which would match the different kinds of weapons, in order to diminish the weapon's damage by a factor. Chainmail is extremely efficient against slashes. It would probably diminish swords and axes damages by half, but would have approximately no effect against piercing or blunt damages from arrows, pikes and maces... On the other hand, plate armor is very good against blunts, and diminishes fairly well the damage dealth by slashes. But piercing goes through. And leather and wood stops arrows and pikes like nothing, but don't protect much against blunt and slashes...

With magic, it should be different. It should take one turn to cast the spell, and one to see its effects. In this respect, shooting a fireball which has an area effect should be a problem of aiming.

In order to make it interesting, I think it should remove the FP aspect entirely, and replace it with a sort of TBS aspect. Turns happen each 15 seconds, say, or 8, or what have you, and each time a turn passes, there is a refreshment of the window. You can see the action going "live". This way, players who are not active can only act in a defensive way, and those that are active may coordinate their battles beforehand, while having to react fast during them, leaving them little time to actually talk meanwhile.

I know that Real-Time-Turn-Based-System is not exactly telling much of what it is but when you find yourself playing it, it is quite enjoyable. I know of a space game which only displays text and numbers that is quite fun, although I don't remember the name right now. I think it is Space Conquest...

Cont. Potential Flaws

---

Rock-paper-scissors

The RPS situations are actually implied, since it is the norm of RTS games to have units with different damage types and armor types. Range stats are built-in in Real time settings. All weapons have maximum and minimum range. However, damage may be scaled by the effective range, but it is not crucial. An archer should be able to deal the same damage firing at a knight at point blank, that is, if the knight doesn't block or stop such obvious attack. When a bowman is firing an arrow, he is very exposed and vulnerable to attacks, this will be the reason why a bowman does not fire right in front of an enemy. Not because of minimum range, but because of the risk.

It is not necessary to scale melee damage by distance. In a real-time combat, when the enemy is close to you, he is already getting extra damage, because not only you are attacking him, your buddies are also attacking him. When he withdraws, you will reach the enemy before your buddies do (due to the actual geometry). This is why the damage is reduced as the 'distance' increases. If you implement a damage bonus on top of this, you will be squaring (^2) the effect. Magics will take time to cast, and some AE spells require prediction. Some AE spells do not required prediction. For example, the Nuke may be casted targeting the commander, when the casting is done, the Nuke will deal damage to the enemies near the commander.

---

Turn based aspect

The global warfare is pretty turn based, because stamina is practically the 'time bar' of the Flags. Each battle takes stamina, and the Flag will wait for enough stamina (and supplies) before starting the next battle. So it is turn-based. When you move toward the small scopes from grand strategy to combat, from planning to execution, the gameplay becomes increasingly real time. So battles will be real time.

The original reason why BattleCry exists, is because click feast is stupid. The idea was to let the players play the real time units and let them carry out the operations in real time. If you move away from real time and back to turn base, the players will know that the commander actually has enough time to commander the whole army by himself, this means that the participation of the individuals is unnecessary. The game still need to deliver the central idea of interdependency and cooperation through gameplay.

The idea that a battle is not won by just impersonal strategies or a few, but the sum of the efforts of everyone involved. The game does not want the individuals to think that they are just serving the Flag, and following orders. Their judgements and improvisions are also very important, it is the type of emergent behavior that cannot be easily achieved through AI or micromanagement. The Flag has to trust the commander, and the commander has to trust the Flag.

I am not arguring that a RTTBS can't be fun, but it does not meet the global objective as much as RT does. During a battle, the Flag commander is the quarterback of a football team, not the coach. There is nothing wrong playing a football game the way a coach sees it. But the idea here is to present warfare not from above, but from within. When you are in a battle, it gives you a strong sense that you must trust the ones that can see, and make sure that you protect the buddy next to you, and trust that your buddy will protect you. Pretty local, but it all adds up.

I agree that making it a close-to-real-time-but-stillturn-based-strategy-game would loose a lot of the flavor. But I thought we were trying to translate this game design into something that could be actually done by a group of people as a hobby, not a holistic design which would be great but will never see the day. I thought it would be actually be more feasible to create a game which has very lttle graphics and has to compute the position of each character every eight seconds instead of one where it has to compute it every milli-second, leaving skill-less those with less connection speed.

What I had in mind for this trasnposition was something closer to a tabletop game, but where the turns would be limited to 8 seconds. In this amount time, you cannot "play" more than one character effectively. You then need the others, just as you would in the MMORPRTSFPS we had designed... Here again, the flag will see the actions of his teammates, but won't direct them. Not directly, at least. A flag still could defend a place alone with NPC-turned characters, I supose, but the NPCs will only benefit from the bonus of the flag, not from his strategies. they will only act defensively, and it should be relatively easy to lure them into defeat. If a flag only is there to defend, even with lots of NPCs, he can resist for some time, but not counter-atack, and NOT make innovative moves. therefore he is bound to die. No strategy.

As for the visual aspect of it, I thought that the screen could display a two part window. On the left part, the "action" part. It displays the vision you have of your environment, that is ten squares around you when you are a soldier, fifteen if you are an archer, twenty if you are a spy (maybe not, though, you seem to spot the flaws more easily than I, so weed them out.), PLUS the set of actions you could take. An eagle-eye view would be more easily done than a semblance of 3D done with poor visuals, because it would require your computer to figure FROM WHAT THE SERVERS TELLS IT where the others are, and what he should display. Not counting that developing an engine, even open-sourced, that will accept this kind of thing is, in my opinion, a waste of resources on a hobbyist project. And If you don't like the idea of squares, we can still fall back on hexagons, and give it a feeling of TBS even bigger, even though it will still be technically a close-to-real-time one...

The right part of the screen should display the chat window, but this window should only show what has been said in the area you can see in the other part of interface. If it could manage to modify the size of the font depending on the distance between your character and the other, it could be great. It would involve more roleplaying in chatting, because you would have to manage what your teammates can hear, and what the others can too. You would have the possibility to speak louder (bigger size) or to murmur (lower size). The officers and sub-officers would have access to the forum part, where they would develop their strategies. Maybe these strategies can be developped quite quickly if the forum is very active at the time.

And I would like almost everybody to be able to view an overall map of the area, but if it could be done in flash animation, it would be better. The map would be pretty small-scale, and it would require passing the equivalent of a magnifier lens on it to actually see something. Maybe it could be folded in order to limit the area you are exploring with your magnifier, in order to see quickly what lies ahead of you. this map would of course NOT display the players, only the landscape.

Well, would this make it acceptable to you? Or is this going too far away from your dream?

I wasn't thinking about translating the game design, I was thinking about prototyping it. The original discussion was theoretical, 64 players per battle instance is about the norm.

The turn-based version

- There is a grid
- It is turn-based with 8-second turns

What happens when the characters try to move into the same cell? What are the dimensions you are imagining? (how big is the board and how many players on each side?)

For the original idea, when a PC Flag attacks an offline Flag, the NPCs of the offline Flag do not necessarily behave defensively inside the battle instance. The game balance will tune it such that the PC Flag will have no advantage of attacking an offline Flag. While it is true that the offline Flag won't pursue the attackers outside the battle instance, the attacker Flag will lose more stamina than the defensive Flag. It is not true that the attacker can just repeatedly do minor damage to wear out the defensive Flag, it will be the other way around.

Quote:

An eagle-eye view would be more easily done than a semblance of 3D done with poor visuals, because it would require your computer to figure FROM WHAT THE SERVERS TELLS IT where the others are, and what he should display.

You still need to do the same for turn-based. 2D or 3D representation has nothing to do with the bandwidth. You have a server, and the clients can run the game using whatever interface/mod they like. Some might run it like a text-based game, some run it with a 2D map, some runs it with isometric view, some runs it with 3D simulator of the actions. It doesn't matter. It is all about sending the minimum information, so the client still has to figure from what the server tells it where the others are.

Quote:

Not counting that developing an engine, even open-sourced, that will accept this kind of thing is, in my opinion, a waste of resources on a hobbyist project.

I am not sure what you mean by 'this kind of thing'. The bare minimum of 'such engine' is just a client chat program connected to two channels, and servers connected to two 32-player channels. From a player's perspective, he is connected to a 32-player channel.

Command message should be heard no matter how far it is. No reason to make the players increase traffic by forcing them pass around command messages. Note that the real way commanders 'chat' will be mostly through macros. The individuals in the Flag shouldn't need to talk a lot during battle.

Depending on the traffic, conversations may not need to be filtered. RP chats and taunts are fun and both Flags should enjoy it. The players shouldn't be concerning about, "I want to move closer so I don't miss the funny things." If the chat window is flooding then filter.

Murmur (or /em) say, and shout are probably expectable. Usually they are not done with different size but with different colors, where /em is usually purple. Font of different sizes can be tiring to read.

In general, the Flags should be prepared for certain situations before entering the battle. The commanders will convey tactics through tactic names instead of the how-tos.

Everyone can see the battle area map.

Quote:

Well, would this make it acceptable to you?

The things I said were not preferences. The consideration here is that the 'turn-based' implementation is very different from a normal 8 player or so multiplayer turn-based system. When you let 32 players take the same turn at the same time, there are some continuity issues, such as the distribution is change in such a way that it might take more than 8 seconds to interpret the new situation.