Hi there, I recently began writing a series of theoretical MMO game design articles. The first three chapters are up at VaporGame.com. Thanks for all of the comments. I'm going to be gone for a few days but I intend to read and respond to everything. I chose to remove the huge block of text here - I don't have time to reformat the updated version before leaving. Chapter 1: Several Giant Leaps For CombatChapter 2: Downtime Does Not A Challenge MakeChapter 3: Like Shooting Orcs In A Barrel Thanks for reading! [Edited by - Renwolf on January 10, 2007 10:21:09 PM]

Thanks for that. It is mostly applicable to a 3D action game, but the design points are pertinent to almost any game.

I don't know about converting believers, but you get a nice big "hear hear" from a fellow believer :)

While I am eschewing combat for the time being in my game, a lot of these themes can carry over to many different areas. The idea of having combat "rounds" sounds pretty good to me, and you could variate the length of rounds based on what kind of combat experience you want. Longer rounds would be slightly more strategic, and still keep the depth that makes most of it fun. While a lot of it seems to be focusing on SLIGHTLY more twitch based play (several times you mention making combat quicker paced), it could also be tweaked for slower games. Another thing to consider, again refering to fast combat (of course that is like 1% of your article, so I dont know why I'm focusing on it) is the fact that if combat is faster, you will go through enemies faster as well. You will have to balance how fast combat is with how many enemies you are expected to advance. But that's true for any game. I think myself I would prefer slower more strategic fights, maybe consisting of multiple enemies, and not having to kill 20,000 bunnies hehe.

Can't wait for your next article!

Thanks for your kind words. They really go a long way in encouraging me.

Quote:

While I am eschewing combat for the time being in my game, a lot of these themes can carry over to many different areas. The idea of having combat "rounds" sounds pretty good to me, and you could variate the length of rounds based on what kind of combat experience you want. Longer rounds would be slightly more strategic, and still keep the depth that makes most of it fun. While a lot of it seems to be focusing on SLIGHTLY more twitch based play (several times you mention making combat quicker paced), it could also be tweaked for slower games.

Yep, I was thinking much the same. And it should be possible to offer both extremes - lightning rounds and strung out strategic attrition fights - and everything in between, all in the same game.

Quote:

Another thing to consider, again refering to fast combat (of course that is like 1% of your article, so I dont know why I'm focusing on it) is the fact that if combat is faster, you will go through enemies faster as well. You will have to balance how fast combat is with how many enemies you are expected to advance. But that's true for any game. I think myself I would prefer slower more strategic fights, maybe consisting of multiple enemies, and not having to kill 20,000 bunnies hehe.

There's probably some confusion as to what exactly I meant by "faster paced" combat. And I'm confusing myself a little, too. What I intended to convey here is that there would be less "dead space" in combat, where the player's mind is idle as he waits for something to happen. So in that sense, the "fast pace" is independent of the encounter's duration. I'd appreciate a little help with the language here to better illustrate what I'm trying to say.

Quick question - need some advice on how I should post these. I'd probably end up spamming the forum if I posted each one separately. But I don't want to make this thread impossibly long, either.

Not going to post them until I get a suggestion on what's best.

Also, Chapter 3 is up at the site.

nice write up ^^ look forward to reading more :)

A very good article. I agree with most of it (and I am attempting to implement some of these very ideas in a game I am currently designing).

Quote:

I'm visualizing a combat system where the player has three primary duties to keep himself alive. One: pay attention to what his enemies are doing. Two: be able to recognize what a particular enemy's more powerful abilities are, through actual real-time experience fighting the opponent. Three: know how to use the character's abilities to counter or mitigate those attacks, and do it.

The way I am attempting to implement these ideas is to eliminate the "Random Hit Chance" that is used in many games. Once the Random Hit Chance has been eliminated I had to use some other mechanic for Armour and defences.

For defence I used "Damage Reduction" (DR). This is that when a character has damage done to them the actual damage applied is reduced. To avoid having character able to eliminate all the damage, I have it that the DR is applied in layers and that the DR is only a percentage of the damage that makes it through to that layer.

For instance, if the player has a shield and armour that each reduces the damage by 10% and 100 points of damage is done to the character, then the shield will first reduce the damage to 90 (10% of 100 is 10 so the damage is 100-10=90), then the armour will reduce the damage to 81 (10% of 90 is 9 so the damage is 90-9=81). Note that if it was taken form the initial damage then the armour would have reduced the damage by 10, not 9.

As the player will always take damage from a hit, actively dodging the attacks of their enemy becomes more important. You can't just sit there and know that your high defence value will stop you being hit. All your defence does is to reduce the damage you take.

This means that if both players are not dodging, then the character with the highest DR will win, but if one of the characters is dodging, then they will win as the other character's attacks are doing no damage to them.

Also, I can give the player's action that use weapons to block and increase the DR (by adding in another layer of DR) and give special actions for a shield (normally a shield spreads the total DR to the front and one flank of the player, but you have an action that concentrates this DR effect to the front of the character but leaves their flanks open to attack).

So if you have 2 attackers attacking you, having the shields extra bonus to your flank would be an advantage, but if you saw one of the attackers do a powerful attack, you would want to concentrate your DR in that direction and maybe even forgo your own attack to use your weapon to block (further increasing your DR).

This example uses all 3 of your ideas:
1) Pay attention to what the enemy is doing
2) Recognise an enemy's more powerful abilities and their activation
3) React using the character's abilities.

Quote:

The common problem encountered by most games in step two of the combat decision making process involves the constant consultation of the UI's ability buttons. The implementation of cooldowns - timers that limit the frequency with which you can use an ability - is more prevalent than ever in modern MMOs. I'd argue that it's artificial to impose these precise time limitations, and in most cases behavior becomes robotic and requires very little decision making on the part of the player.

Here is where I start to disagree. I don't think that cooldown timers necessarily create Player UI focus. I think it is the design of these UI elements that create the focus, not their existence.

If you hid the timing display and only reported when and what abilities are available and then use a visual cue (flash or such) and/or audio cue (a "bing", character speech, etc) to notify when a change has taken place (like an ability has finished cooldown), then this would not require the player to constantly focus on the UI elements for the cooldown.

As you say later in the article, the player will know aspects of their character, like (as you present as an example) the increase in costs for reusing an ability, or the rate of diminishing returns for frequent use of an ability. So why can't a player also keep track of time, or respond to visual cues or listen for audio cues that indicate cooldown status.

Bad UI design does not mean the mechanic that uses the UI element to report it's status is flawed.

I do agree that the current UI reporting of this mechanic is flawed, but I disagree that the mechanic is it's self flawed.

The reason that the current implementation of this mechanic can seem a robotic repetition is because there is because you have very little control once the ability is started. In the system I am using for my game design, the player has the ability to cancel (at a slight cost) an action after it has been started.

I use 4 states of an action:
1) Idle: This is the default state an action can be in. If an action is in this state is can be selected for activation/use.
2) Warmup: This state is when the action is being prepared for use. With spells this would be the casting of them, or with a weapon this is the wind-up before the actual swing.
3) Active: This state is when the action is taking place any effects from the action are applied here
4) Cooldown: This is when the action has taken place and is the follow through. It can also be where the character is "off balance" after making an attack and can't immediately make another attack or action. With spells, it is a temporary drain of the character or the mental readjustment needed after unleashing arcane powers. There is also a global cooldown where all actions become temporarily in the cooldown state (but only briefly).

Now during the Warmup and Active states, the player can cancel an action. When the player cancels the action in warmup, only the cancelled action is effected. This effect is to place it into cooldown.

If the player cancels the action during the active state, then this ends the active state and enters the cooldown state of the action as normal. This might trigger a global cooldown in the action would normally trigger a global cooldown.

This relates back to what you were discussing earlier about reacting to the other character's actions. Player A might activate their most powerful attack, Player B seeing this will react and activate their defence against this attack. Player A then cancels their most powerful attack upon seeing player B react. This would place the powerful attack into cooldown, but they could then change their action to something that bypasses the defences that player B just activated (player B can then react to this new situation and so on).

As you can see, just the simple act of using an ability comes with a lot of tactical decisions. Timers, like cooldown do not unnecessarily limit the choices of a player, current implementations might, but this is not the mechanic that does it.

Quote:

Let's talk about the Refresh. Despite its hastily and poorly conceived name, the Refresh is a great thing. Each time you complete an encounter of the scripted stage of an encounter, you get one.

How would one determine (mainly in an mmo, but it also applies to other games) when the "combat" ends. Sure, if the player kills the monster then the combat ends, but what if the player or monster attempts to run, when does the combat end?

Also, this leads up to an abuse of this by exploiting it. If two players are engaged with a battle, one of the players can run from the battle to refresh, once they have refreshed, they then return and then the other player does the same. As the monsters are still in combat at all time, they can not refresh and the players are constantly refreshing. Using this, even low level characters would be able to take on enemies far more powerful than themselves (and the more character's you have the more powerful the enemy you can take on). If you design the game so that player shave to do this, then why is it put in there in the first place and you should just make it that players can either heal themselves or there is no need for a refresh (ie: they don't experience and increased costs to use abilities etc, or even take damage).

Quote:

The second purpose of the Refresh goes back to how we're going to kill those two previously mentioned, and invariably doomed, birds. The "demanded attention to UI buttons for cooldowns" bird and "over-simplified combat" bird will be feathery red smudges on the pavement of gaming advancement.

It doesn't really eliminate the need. They player will watch their UI so that they know the time they have to run from combat so that they can refresh.

Changing the mechanic, but not changing how things are represented will not eliminate the need to watch the UI. If you report things directly to the UI, then they will watch the UI.

What this means is to stop the reliance on the UI, you must change the UI. You can have mechanics that don't need to report as much to the UI and reduce the amount of "things" that the player needs to track on the UI, but this will not eliminate the reliance on the UI.

The UI is a necessary evil, but we can change the way information is portrayed on it. Instead of having timers that the player can watch, only tell the player when the timer has expired. Let the player judge when it would be about to expire, have this a skill for the player to develop, or even show it visually on the character.

Other games like FPSers rely on the player to be able to judge cooldowns and such (of their weapons) and this can also be relayed by the animations (the reloading animations etc). So why can't this same method be used in MMOs?

This UI focus problem is not a mechanics problem, but it is a problem of UI design.

Quote:

Current games don't take full advantage of the human dexterity to maximize a player's control over his character. Let's examine what's being done, and what I think should be done.

I agree that these designs can limit what players can do. But, the most flexible method would to allow a player to map any function to any key or combination of keys.

Most game do allow this (not fully but close enough), however, they seem to fall back to the WSAD and mouse. This indicates to me that there is a reason that this is used. It could be because that is what they are familiar with, or it could be because it give them the best control.

In the game I am developing, we are using a combination of control schemes. First off there is the WSAD and mouse look. But you also have the option to change the control scheme slightly in certain circumstances (like combat) and leave the mouse alone.

You do this by selecting your target and then activating the "combat control" scheme. As this is a player decision this will not place the player into confusing as they have to decide to do this.

Once the combat control scheme is activated the character will attempt circle strafe their chosen target and will automatically focus attacks against it. This frees the player up from having to concentrate their attention on tracking their opponent and frees up a hand for activating abilities and such.

The player can choose not to use the combat control scheme and just use the standard control scheme (or even switch between them) to use the best system for a particular task. In certain situations (like combat with multiple opponents) one of the control schemes will be better, in other situations (a 1 v 1 duel) the other control scheme will be better. It is a choice that the player must make for themselves.

The main problem with a "key only" interface is that it is not intuitive. This means that beginning players will have some difficulty learning the controls of the game making entry into it harder. This means less players and less income. As MMOs need income to survive, having this entry barrier could be a big problem.

The system you presented is not wrong, and neither is it bad. But I don't necessarily think it is "better" (its not worse). It is better in some ways, but not so good in other ways. It is different and thus will appeal to different types of player. A player more used to player FPS games will probably find your system more intuitive. New players or players familiar with different genres might find it harder to understand.

Quote:

Original post by Renwolf Quick question - need some advice on how I should post these. I'd probably end up spamming the forum if I posted each one separately. But I don't want to make this thread impossibly long, either. Not going to post them until I get a suggestion on what's best. Also, Chapter 3 is up at the site.

You might check with someone else (a mod), but it seems like a good way to do that would be to take a small piece of your chapter (an abstract), and post a link to each chapter as it's complete.

I think posting links with maybe a short paragraph summary, would be the best way. People generally don't like reading really long posts. You already posted a link to your site, so that might be enough. I've just read the next 2 chapters, and am still processing them. The scripted encounters from Chapter 3 reminded me of Neverwinter Nights, which had a system for defining an encounter which would spawn the appropriate number of appropriately leveled beasties to match your party.

Although for me, the perfect game would have no spawns at all. There would be "families" of creatures (some aggressive and some not so) and as their populations dwindled they would go off to hiding places and repopulate, or call in reinforcements from other locations. Instead of having spawning, which to me is one of the most boring and unimmersive things in a game, there would be populations of monsters that would dynamically be altered to fit how the player population is acting. A dungeon would just be the home of one of these families, and it would have their possessions. If you beat them down and steal their stuff, some of them may take their most valuable possessions and escape. Later, the dungeon will be full of their extended family, and be more protected. Or, you may see the survivors from the raid build a new fortress somewhere else, which would also be more protected and have better loot.

But sometimes, a particular "family", may actually be wiped out by the players and go extinct!

I don't really like the idea of spawning encounters to match a given players abilities, because that can too easily lead to a bland experience, where as you increase in skill, your fights generally end up being against similarly skilled opponents. I like having areas that I go into and then die because I'm not ready for it yet :)

Although that can be accomplished with dynamic spawns as well, you just have to be extremely careful about it. Since you are not developing a game, it's easy for you to say, "yeah, dynamic spawning, down with static!" But in the real world, a poor dynamic spawn system is often going to be worse than a poor static spawn system, and even a good dynamic spawn may bring down other aspects of the game.

You definitely make some good points about the game pushing back though. These games, while they ought to be MORE dynamic than single player games, often break the immersion when you can always pick and choose a safe encounter, by finding the enemy who's name is green. You need to be attacked from things around the corner or in other rooms, that you either run like hell to escape or muster up your courage and face. That happens often in single player rpgs, but I've only had it happen a few times in MMO's, and it was usually when I was in an area I wasn't supposed to be.

I haven't read all the post yet, but i initially wanted to reply to the no healer mention (before I forgot)

I'm wondering if a healer based class could be used to "restore" the winning side or at least slow the process of injury. A look at Americas Army. I haven't played it in over a year but I remember when they first introduced the medic. Now a medic/healer in a realistic first person shooter at first i was skeptical. How could they heal and still have it realistic?

Well when shot, the player would bleed and start to lose health. Now usually it was by a stray bullet or a bad shot in which the player was then able to find cover. But he was now bleeding, it was usually 1 in 30 where the player would actually bleed to death (because he was usually shot or healed before that point). But he was still losing health. The medic could come and patch his wounds so the bleeding would stop. The medic could never restore health, just stop the active loss of it, until the person was hit again. (usually that was the killer one).

Now I'm sure this has to be experienced before people truly understand what i'm talking about but it did a few things.

#1. It reduced the need for myself and other players to not look at their health unless they were bleeding. Most of the time you were involved in the 3D action.

#2. The healer wasn't removed completely, they just had a different strategy. They could act as soldiers until called for as a medic.

Now this can be used in an MMO, the healer could either stop the bleeding of fellow members, the health would not always be watched as once they are hit, they are hit for good (until say they are no longer in battle, and some time has passed before they recuperate again.) All they need to look at their health for are signs of bleeding, which can be just a glance, or whether they think they have enough health to go up against the next foe.

The healer could also be a valuable resource as they could raise or increase the health restoration after a battle. whatever is deemed as "end of battle".