Trending Games | World of Warcraft | Overwatch | Guild Wars 2 | Closers

    Facebook Twitter YouTube Twitch.tv YouTube.Gaming Discord
Register
Quick Game Jump
Members:3,795,152 Users Online:0
Games:981 

Show Blog

Link to this blogs RSS feed

The Theory Of

Here you'll find discussion of all manner of topics relating to the theory of multiplayer games. As I see it, anyway. A note to commentors: if you stray off-topic or if your reply contains ad hominem attacks, your comment will be deleted.

Author: JB47394

Videos of spaceship movement system

Posted by JB47394 Tuesday February 24 2015 at 9:48AM
Login or Register to rate this blog post!

Back around 2009, I was involved in discussions on the (predecessor of the) I-Novae Studios forums about spaceship movement systems - among other things.  The backdrop for the discussions was the assumption that spaceships could fly seamlessly from the surface of a planet, to interplanetary space, and then back again.  In such an environment, we were discussing how to handle ship rendezvous and interception.

The first cut was to say that as two ships came closer together, they would slow down in increments and eventually stop - the amount of slowing was dependent on their distance from one another.  Once stopped (they could still move around each other like spaceships), they would be close enough together to interact.  Shooting, trading, whatever.

The problem that I had with such a system was that it allowed something like a fighter to intercept and stop a battleship.  It seemed a bit silly to allow that.  But it was equally impractical to prevent a fighter from interacting with a battleship, whether it wanted to shoot the battleship or just enter its hangar to land.

One solution suggested was to allow the ships to interact on the move.  A lot of folks were having difficulty understanding how that would work in various scenarios, so I created a prototype that allowed people to try scenarios out for themselves.   The implementation was in 2D, with the simplest possible graphics, but it was full scale, real time and multiplayer.

Now, 6 years later, I've created three videos to explain how the system works.  I hope you enjoy them.

Basics.  How to fly a spaceship.

Intercepts.  How to make an intercept on another ship.

Jumping.  How to travel between stars.

Housing: Now You See It. Now You Don't

Posted by JB47394 Tuesday June 17 2008 at 11:45AM
Login or Register to rate this blog post!

Synopsis: Player character houses can travel with the character, being automatically incorporated into the layout of the nearest town as the character moves around the world.

lordaltay1 blogged on player character housing and I dropped a reply that got me thinking about a different way to implement player character housing.

In a nutshell, the idea is that housing is tied to your character.  If your character isn't in the world, then neither is its house.  If your character is in the Desert Region then your house is in The Town In The Desert Region.  If your character is the Forest of Blingo, then your house is in The Town In The Forest of Blingo.  Whenever you visit a town, village or city with your character, you'll find your house there.

As a result of this, your house must be dynamically placed into every town that it gets added to.  If the center of a town is considered to be the intersection of two roads, then houses are added at the corners of the intersection, then along the roads leading away from the intersection and so forth.  The number of houses in town is a function of the number of characters in the surrounding area.  It's a kind of dynamic citizenship.

The same would be true of a town by the shore of a lake, or along a river or any other location; the town grows and shrinks from a central point.

To take the idea a step farther, consider that the appearance of a town could change as more and more dwellings appear.  The appearance of the buildings could change, and the appearance of the roads and other paraphernalia that give a community its unique flavor could also change.  A village has a crossroads, while a town has a town square.  The village NPC blacksmith becomes the city's iron works.

The village by the lake acquires upgraded docks.  The village in the plains acquires a stronger defensive wall.

Imagine all such upgrades and downgrades happening smoothly and subtly instead of everything snapping abruptly from one appearance to another.

A step farther still would permit a house to change its external appearance according to the region of the world the character has reached.  A middle-eastern look when in a middle-eastern-themed region, and a european look when in a european-themed region.  That could be done to a character as well, but I'm fairly certain few people would want that.

The interiors of houses could remain the same or change as the exterior changes.  A chair in the european-themed region looks a certain way, while in the middle-eastern-themed region it might turn into a settee.  Perhaps bookshelves become scroll racks.  And so forth.

This would be an alternative to having banks serving as a place to dump things.  If you want something of yours, visit the nearest town (which is as large as the number of people in the region), go to your house, get it and get back to adventuring.

For crafters with shops, I can imagine their shops traveling with them as well.  The shops would be placed according to the zoning laws of the town.

I see guilds using a shared house instead of having a special feature for guild halls.  A shared house can appear in N towns simultaneously, with the selection of towns based on how many shares each owning character has.  If six characters hold 10, 10, 10, 5, 5, and 5 shares and the house is rated for three simultaneous placements, then the towns that the three 10-share characters are in would have the house appear.  If one of the 10-share characters logs off, then one of the 5-share characters is used to place the house.

There are issues around a character changing towns while there are people in his house, but that can be dealt with.  The behavior that would NOT be implemented would be to move the people with the house.  The choice is simply a matter of how tolerant the player would be of letting the characters continue to stay in the house.  A shopowner logging off while customers were in his house may well want the house to stay open so long as customers are inside.  A minute or two after the last has left, the shop fades from sight.

Other variations are things such as adding your house to a town only if you're within a certain distance of it.  If you're way out in the boonies, you might have the opportunity to 'make camp', which takes a little bit of time and produces a simple campsite.  Enter the tent and you have access to your house's contents.  Or perhaps some defined subset based on volume.  For example, in your house, there is a locker of a limited size that is your Field Pack.  Whatever you put in there in town can be accessed by making camp.

All of this is intended to make housing something that adds to the fun of the game.  If you see a house, you know the owner is nearby somewhere.  After all, he's a resident of the town where his house is located.  Of course, 10 seconds later that house might fade away to be replaced by another.  The first owner left the region and a new character entered.

Challenges

1. The permanent feel of a town is lost.  This is considered a feature.  The permanent feel of towns in MMOs today is that they are ghost towns.  Many houses with nobody around.  This system attempts to produce towns that at least reflect what's going on nearby.

2. The exterior appearnce of a house is not directly under the player's control.  This will undoubtedly be a sticking point for many players, but I'm not excessively worried about it.  During the upgrade and downgrade cycles, the exterior treatments of houses can be upgraded and downgraded as well.  Plant a shrub in the yard of your house and in the village it's an unruly shrub.  In a town it's a trimmed shrub with a border.  In a city it's a trimmed shrub in a planter.  The basics of the exteriors would be under player control, but they would be upgraded and downgraded automatically.

3. There's no way this can be implemented today, except in a very  basic form.  Upgrading and downgrading the appearance of housing as a site moves from village to town to city and back down again would require massive amounts of artwork as well as careful crafting of the code to make the transitions work.  Houses constantly appearing and disappearing would be an additional load on the network, CPU and graphics of all the clients.

It's a tough system to implement, but I'm pretty sure that I'd enjoy having it.

Crafting Systems - Misapplication

Posted by JB47394 Friday May 23 2008 at 12:15PM
Login or Register to rate this blog post!

There are many crafting tasks that are not suitable to an Artisan.  When a large number of an item is required, and there is no significant decision-making involved in the construction of those items, that crafting should be implemented to appeal to a Manufacturer.  Those systems either involve push-button manufacturing as is found in World of Warcraft, or NPC factory manufacturing as is found in real-time simulation games.

An example of this is arrow crafting.  Arrows do not have 'big picture' skills.  All the skills used involve fine motor control, which is very much ill-suited to computer games.   Fitting vanes, attaching nock and head and so forth are simply not Artisan tasks.  A game that permits a character to make its own arrows need only use the standard push-button manufacturing model.

Crafting Systems - Group Crafting Example

Posted by JB47394 Friday May 23 2008 at 12:15PM
Login or Register to rate this blog post!

EVE Online involves the creation of many objects, and spaceships are certainly a mainstream element of the game.  They are an opportunity to occupy Artisans.  Indeed many Artisans could be involved in the crafting of a single ship.  That is an unexplored experience in games: group crafting.  Imagine players beginning with the crafting of each component that goes into a spaceship.  Each engine could be crafted individually by one player.  Or even a team of players.

I can imagine an assembly game that involves the same basic process as crafting a sword; the players assemble components for an engine one part at a time, aligning and calibrating each piece.  Once the entire engine is assembled, the computer examines everything to see what its performance should be.  Just a few numbers to reflect the operation of that engine are stored, a standard graphic is used, and the engine can then be moved into the drydock.  Other players will attach it to the drive system of the ship.

Naturally, that drive system had to be constructed in advance.  As were the other engines.  And all other major components of the ship.  Assembly tasks have great potential to occupy scores of players as they converge on accomplishing a single large task.

This sort of a system would completely change the game experience of EVE Online, so it's impractical for that game as it exists today.  Ship turnover in that game is very high, and new ships are manufactured at a furious rate.  If this were introduced there, it would have to be reserved for specific types of ships that are used for special purposes.

One of my favorite game design ideas is to have an entire game revolve around the construction of a massive device that is composed of assemblies of assemblies of assemblies.  Just a massive beast.  Fighters would find bits of the machine in the hands of enemies.  Explorers would find them lying about in obscure places.  Traders would negotiate for pieces in deals.  Crafters would assemble some components from scratch.  How they all fit together is to be figured out collectively by the players.  But when a piece fits, you know it fits.

When the machine is powered up and turned on, the game changes in some significant way.  This might be the means of opening the next section of content, or adding whole new systems to the game.  For example, the machine might transport characters to a new gaming environment (a la the movie Contact), or it might create spaceships for the players to now begin doing in space what they used to do on the surface of their planet.  It could even introduce a magic, psionic or superhero system to the game.  Anything is possible.

Crafting Systems - Ship Salvaging Example

Posted by JB47394 Friday May 23 2008 at 12:15PM
Login or Register to rate this blog post!

While making a sword blade is cliche, there are many other examples of crafting systems that appeal to enthusiasts of other types of activities.  A good contrasting example is salvaging in EVE Online.  After a battle is over, there are wrecked ships drifting about.  A character with the right combination of ship, equipment and skills can salvage items from those wrecks.

How does EVE Online implement that?  Point and shoot, unfortunately.  As with fantasy MMO crafting, EVE Online eliminates the actual process of salvaging.  The real fun of salvaging is left to targeting the many wrecks, tractoring them close, then firing up an automated salvager on the wreck.   There's a modest skill to it all, which is good, but enthusiasts who see Salvaging in the game are likely to want to actually do some salvaging. 

I made a proposal in the EVE Online forums to introduce a new item to the things returned by the automatic salvagers: Salvage Clusters.  These boil down to being puzzles that are of the kind where there are many individual pieces that move in limited ways and the goal is to tease the puzzle apart to get to the special pieces.  This is very much what a salvager does when he enters a wreck.  He cuts this, he opens that, he pushes there, he pulls here, and out comes the item he was after.  That is an example of using a puzzle to entertain a crafter.

This is far less involved than the near-simulator that would be required for the sword blade crafting example.  It's not necessary that all crafting systems designed for Artisans be complex or overwhelmingly sophisticated.  The goal is to entertain, and different people can be entertained by different things.  This is why a simple puzzle could suffice for a salvager while those interested in crafting their own sword would look to a more sophisticated form of entertainment.

Crafting Systems - Sword Blade Example

Posted by JB47394 Friday May 23 2008 at 12:15PM
Login or Register to rate this blog post!

A game that entertained an Artisan swordsmith would do something like the following:

The player first gathers the raw materials needed for the blade.  That would be a block of metal.  That metal could be any kind of metal.  Tin, lead, gold, silver, iron, steel, and so forth.  Don't assume that all swords should be steel.  Suppose a magic spell needed a sword made of silver?  A deception of a well-armed military involved tin swords?

The block might not be pure.  It could be an alloy or have inclusions.  It could be in the wrong shape for the Artisan to begin working.  Not all raw materials are created equally and there may be some preparation time involved before the main task can begin.

How can a game do all that?  Procedural content.  Generate a random number for a block of metal, and all of the necessary information is derived from that.  Where the inclusions are, what types they are, whether the metal is alloyed, etc.  It's all a question of how well you can construct the procedural generator.  It is only when the Artisan begins fiddling with the block of metal that the game has to turn the procedural content into detailed information.  Until then, blocks can be bought, sold, traded, juggled and otherwise manipulated without ever deriving any of the available information.

How is metal worked?  By heating and hammering.  So an Artisan would have access to a heat source of some kind as well as a hammering mechanism.  This is the fun part for the Artisan.  He drags the lump of metal into the fire, watches it as it heats up, and notices the color of the metal, not a scale or timer countdown.  The point is to appeal to enthusiasts of metalworking, and that's how metal crafters did things.

The metal reaches the desired temperature and the player moves the metal to an anvil.  He holds down the mouse on the spot where he wants his assistants to strike.  Again, this is how it was traditionally done.  The assistants are an NPC fiction.  They're just unskilled muscle labor.   They strike at the places that the Artisan indicates with the mouse until the Artisan stops the process.  All the while, the metal is cooling, making the hammer blows less and less effective.

Each strike will reshape the metal block.  That's key to the Artisan.  The Artisan is using his skill in working with the game systems to get that block shaped the way he wants.

For reasons of practicality, the artisan cannot make an arbitrary item.  The goal is to make a particular item.  He's making a specific pattern of sword blade, and how well he matches the ideal blade will determine the properties of the item that he is crafting when it is finished.

This is how A Tale in the Desert tackled its own blacksmithing.  Its system was primitive and could be done far more intricately, but the essentials are there.

The metal block might start as an amorphous blob of metal, as a purpose-built rod of metal or just a cube of the stuff.  Any way it arrives, the Artisan has to work it into the proper shape.  Note that the form of the raw materials will influence the desirability of those materials to various artisans, affecting the game's commerce system.

Now the Artisan has hammered out the blade and declares it complete.  The sword is rated by the computer (a simple calculation) and assigned certain numbers for durability and so forth.  At that point, the blade is ready to be incorporated into a sword or whatever other engine of destruction it was originally built for.

Note that I have glossed over the meat of the Artisan's activities, which is the shaping of the metal.  The fun of the crafting is in landing blows where the player wants them, and seeing the shape of the blade take form.  The blade should be straight or curved, of a certain length, perhaps tapered, perhaps with a specific type of edge, perhaps even the metal needed to be folded at the beginning.  The blade would have to retain the best properties of the metal it is fashioned from, and too much heating or the wrong cooling could damage those properties.  The game would lay all this out for the player, but the player would have to use their skills to ensure that the blade was properly crafted.

In order for a player to be entertained by all this, the player has to be an enthusiast of blade crafting.  While blade crafting is similar to any other blacksmithing task, it's different from whitesmithing.  Working with silver and tin to make jewelry and decorative pieces is a different game system.  So you might imagine that different players will spend their time in different systems.  Further, there is real player skill in mastering one of the systems.  Just as there is skill in mastering the level/loot system in most fantasy MMOs.

If you want a sense of having to master many individual systems, try playing EVE Online.  They implemented multiple systems that operate on different principles.  Manufacturing, trading, PvE, PvP, corporate operation and so forth.  Fantasy MMOs just don't do that.  All systems are about levels, classes and character skills.  So if you are having a hard time understanding what I'm talking about, or don't believe it's practical, try the free trial of EVE Online and get a sense of being overwhelmed by your gameplay choices.

Crafting Systems - The Artisan

Posted by JB47394 Friday May 23 2008 at 12:15PM
Login or Register to rate this blog post!

There are several perspectives on crafting systems.

1. Consumer.  This is the player who spends his time using a crafted item.  His greatest concerns are on the capabilities and appearance of the item, as well as what he has to do to get it.  The skills of a Consumer revolve around how the crafted item is used.

2. Merchant.  This is the player who spends his time buying and selling crafted items.  His greatest concerns are those of supply and demand, market pricing and profit margins.  The skills of a Merchant revolve around money and inventory management.

3. Manufacturer.  This is the player who spends his time cranking out copies of a crafted item.  His greatest concerns are similar to the Merchant, except that his needs are for raw materials, and he may be able to work with a Merchant to get the products out the door.  The skills of a manufacturer revolve around factory efficiency and materials management.

4. Artisan.  This is the player who spends his time making individual items.  In contrast with the Manufacturer, his goal is to work on a single crafted item.  The skills of an Artisan revolve around the crafting task, whatever that might be.

Players carry one or more of these perspectives with them when they consider crafting systems.  The perspective that is rarely accommodated by MMOs is that of the Artisan.  Because that is one of my primary areas of interest in gaming, I'd like to offer a few basic ideas to sate the desires of my fellow Artisans.

The Artisan is interested in sitting down to make something.  It may be used by someone, it may be sold, it may be powerful, it may simply be nice to look at.  However, the perspective of the Artisan is to focus on its crafting.  We all like to see something happen with our creations, but that motivation is apart from that of the Artisan.  The Artisan wants to be occupied with fashioning stuff from other stuff.

In concrete terms, that means that the Artisan wants to be using the controls of the game to manipulate raw materials into a finished product.  The Artisan wants that process to occupy him for hours.  This is the same attitude that any other player has; a player wants to do something that is fun for hours at a time.  Monster bashing, perusing markets, exploring; whatever the interest of the player might be.

What can Artisans be doing for hours on end that could be any fun?

The activity will vary, depending on the type of crafting being performed.  As with all activities, they should be implemented into a game in ways that appeal to enthusiasts of that activity.  Swordfighting systems focus on factors that appeal to swordfighting enthusiasts, while gunfighting systems focus on factors that appeal to gunfighting enthusiasts.  It makes no sense to make swordfighting like gunfighting or vice versa.  In the same way, each form of crafting must be designed to appeal to enthusiasts of that activity.

The classic crafting example is blacksmithing or swordsmithing.  In standard fantasy MMOs, players fill a recipe with raw materials (which can be an arduous process), then press a button to make the item.  That moment when the player presses the button is the moment that an Artisan is disappointed.  That is the very moment when the Artisan's game would normally begin.

Seamed (Multi-Application) MMOs

Posted by JB47394 Monday May 19 2008 at 1:10PM
Login or Register to rate this blog post!

One of my starting assumptions about MMO design is that the player is looking at a screen that shows his character in a 3D virtual environment.  The player uses that character to seamlessly visit each experience in the game.  To craft, the player uses their character at a forge.  To fight, the player uses their character near monsters.  And so on.  I recently decided to start exploring the what if exercise of departing from that assumption.

The prevailing mechanism has the problem of guiding all activities into the same mold.  I see a character in a 3D environment, and I think a certain way.  Characters in 3D environments do certain things, and those are the things I expect.

But what if you didn't see a 3D environment?   What if you saw a web-based 2D flash application for crafting?  What if there was an FPS game for one type of combat, a class/level system for another type of combat and a turn-based email game for yet a third type?  All affecting in the same world.

Chaos, you say.  Variety, I say.  The FPS game happens in space.  The class combat happens on the ground.  The turn-based combat influences the political game that spans whole planets.  The crafting game affects only individual items used by any of those other systems.  The systems don't necessarily interact directly, which dovetails with what I was talking about in Designing Roles for MMO Characters.

Why do all this?  Because different players enjoy different things, and what they enjoy can interrelate in ways that provide greater entertainment for everyone concerned if they aren't forced to use the exact same user experience.

I'm back to playing EVE Online, which is a game that caters to different types of players intent on different activities.  Yet there is constant leak-through where one system impacts another even though the players of the respective systems aren't interested in those interactions.  It makes no sense to create a single user experience that is a vast compromise across the spectrum of player agendas when an isolated treatment of each is what players would really like to have.  When systems are most entertaining while intertwined, intertwine them.  When most entertaining while separate, separate them.

Note that switching between games doesn't require character travel, which is a long-time gripe of mine.  If I'm in the middle of combat and decided to do some crafting, I can just close down the combat side and switch over to the crafting side.  Or do both at the same time, for that matter.  I switch from a FPS experience to a 2D flash experience - or however the crafting game wants to be structured.

Different people have different temperaments, different schedules and even different game devices.  Is there any reason why players of a game shouldn't be able to play through their cell phone and affect the exact same game world that others are playing through their virtual reality rooms?  The cell phone guy may only be able to place a sell order on something that he owns, while the guy with the virtual reality room can get an adrenaline rush from combat or do complex design tasks, but they'd be affecting the same virtual environment.

As you advance your own designs, think about how your game can have many faces for many different people.  Don't assume that all your players have to be interested in killing monsters and gaining levels.  Make that only one bit of the entertainment juggernaut that you can assemble.  Look at an activity that you're going to put into your game and then figure out the ideal interface for the people who enjoy doing that thing.  Take everything into account, including how long they'd want to be actively involved in the game per session (10 seconds?  10 hours?), where they are likely to play it, whether they're likely to play by themselves or with friends present (thank you, Jimmy_Scythe), whether they want mental or physical exercise (q.v. the Wii) and many more factors.  Let the consequences of their games affect each other without the way they play the games affect each other.

Designing Roles For MMO Characters

Posted by JB47394 Thursday May 8 2008 at 12:05PM
Login or Register to rate this blog post!

Synopsis: An MMO design can be thought of as being a set of character roles that relate to one another.  Each role is supported by a complete entertainment system, and the interactions of those systems establishes the way the roles relate to one another.

This article is about examining MMO system designs by considering character roles and the way that they relate to each other.  The purpose of looking at system designs in that light is that it underscores the multiplayer aspect of the game genre; if players are not interacting, then the value of having the players in the same virtual environment is squandered.

The examination begins by considering a supposed role for a character.  As an example, the Crafter.  If you want a design to contain a role for Crafters, then you must have a complete entertainment system devoted to crafting.  Crafters must be able to be using that system every moment that they are in the game.  They should not be obligated to do things unrelated to crafting.

But what does related to crafting mean?  Part of the challenge of design lies in finding the set of activities that someone enthusiastic about crafting wants to experience.  Critically, the designer of the game is focused on ensuring that crafters have a game experience of their own.  That experience is not polluted with warrior tasks or political tasks or animal husbandry tasks.  It's about crafting.  A player who enjoys crafting can enter the game and do nothing but the activities of crafting and be happy.

That's the key to creating a role: that there are players who enjoy doing a specific set of activities.

Now comes the second part of our examination, which is the way roles relate to each other.  It's great to have a crafting role, a political role, a fighting role, a shipping role and many more roles besides, but unless they are related to one another then the multiplayer element of gameplay is not fostered.

To relate roles is again an exercise in design.  No magic recipe exists that I know of to ensure that any two characters with different roles will interact.  However, there are some natural ways in which character interact.  There is the producer-consumer interaction.  There is the cooperative interaction.  There is the competitive interaction.  Undoubtedly there are others, but use those three as a starting point.

Returning to the example of crafting, we'd pretty quickly look to a producer-consumer interaction.  A crafter creates things that characters of other roles consume.  At that point, we have to do something very important: we have to make sure that the producer wants to produce for the consumer and that the consumer wants to consume what is being produced.  If bakers are making bread for warriors and warriors don't want to be bothered with the maintenance task of eating bread, then the relationship between the two roles is not a sustainable one.  Players with warrior characters will attempt to minimize or eliminate that relationship every way that they possibly can.

If two roles are going to relate, we want them relating in mutually enjoyable ways.  Crafters who make warrior tools are liked by the warriors.  This is the case in Eve Online where the people piling all those goods into the economy in support of the corporate wars are appreciated by those fighting the corporate wars.  Both sides are entertained by the interaction.

The last part of the examination is making an interaction sustainable.  If a certain character role involved providing a one-time service to another character role, we don't end up with a sustainable interaction.  We want an interaction that players will continue to use time and time again.  Eve Online implements a marvelous structure where corporations chew through equipment in their bid to play the corporate war game, causing a constant demand for harvested goods as well as crafted goods.  That is a system that has all three elements of a good role

1. An entertaining system for the role
2. Mutually-entertaining interactions with other roles
3. Sustained interactions with those roles

I should note that interactions between characters do not need to be live.  Being able to drop off an item with an offline crafter so that he can repair it later may be perfectly acceptable.  Selling crafted items on an automated market does not involve live interactions, but the interactions between the members of the different roles are certainly present.

If you are hoping to create a distinct role for a certain activity in your game, be sure to devote a full entertainment experience to that role.  Players who enjoy that activity will spend time there, and you will have created that role in your player population.  By relating their activities to other players' activities in a sustainable way, you will find that your player community will strengthen.  This is true whether the interaction is designed to be cooperative, competitive or some other variation.  So long as both groups enjoy the interaction, you're golden.

Damage, Stealth and Magic

Posted by JB47394 Thursday April 17 2008 at 1:18PM
Login or Register to rate this blog post!

Synopsis: Eliminate classes, and make the success of actions dependent on player skill, character skill and various ambient conditions.  The influence of the ambient conditions tends to channel characters along class lines without the use of a straightjacket.  Damage generation is removed from magic, turning magic into a means of altering the ambient conditions that influence all actions.

Daedren's recent article "My Kingdom for Different Archetypes" got me thinking about how I'd like to see the general notion of classes tackled.  I started to reply to his article, then decided that I wanted to go on a bit.  Ergo, article-o.  Most of what you'll read has been suggested by many people through the years.  This just happens to be my twist on the overall package.

There are no formal classes.  Players can use their characters to do anything that the game has implemented.  However, you'll find that certain combinations of activities simply cannot exist for what I hope will seem thoroughly practical reasons.

Character success in any action is a consequence of player skill, ambient conditions and character skill (the PAC).  A weak character cannot do strong things.  A player who cannot solve puzzles handicaps his character's success in puzzle-related actions.  The presence of a wall means that a character must walk around.

I'm not overly-concerned with the relative impact of player skill, character skill and ambient conditions on the outcome of an attempted action.  Their balance is critical to the resulting system, but for now just keep in mind that all three can play a role.  I particularly want to stress the influence of ambient conditions.

Characters use physical weapons when they want to do harm.  From daggers to longbows to clubs, the PAC will influence the outcome of whatever aggressive action the player is attempting.  A crosswind or even nearby combat may alter a bowshot.  Uneven terrain may unbalance a swordsman's stroke.

Characters can attempt all sorts of activities that traditionally involve a specific class.  Stealth is the quintessential example.  If a character wants to sneak, it sneaks.  Its success, as always, is dependent on the PAC.  It is the combination of those factors that causes players to decide how much effort they want to put into making their character able to move in a stealthy fashion.  Don't wear noisy armor.  Don't carry a lit torch at night.  Don't wear bright red clothes in the middle of a crowd of blue.

Climbing is another example of a standard activity.  There need be no "climber" class.  Simply establish that the PAC influences the attempts at climbing, and let players decide how important climbing is to them so that they can configure their character accordingly.

When a character is not well-configured for a given activity, it may be a simple task to reconfigure.  Climbing while carrying a heavy pack may be completely impractical.  But leaving the pack behind may be acceptable.  This is part of the value of having ambient conditions play a role.

This brings us to magic.  I completely divorce magic from damage.  As I stated, weapons are for damage.  Magic is for fooling with ambient conditions.  Magic is something that anyone can do, subject to the usual limits of the PAC.

If a pack is too heavy for a climber, the climber can leave the pack behind or have someone who can use magic make it lighter.  In combat, a mage can make the ground slippery or create magical barriers.  A mage might even be able to alter a character's skills in the tradition of buffs and so forth.

We could argue that many things that a mage can do could easily result in injury or death to an enemy.  However, because I want people to stick to traditional medieval weapons to do damage, magic is artificially unable to do direct damage.  A mage might be able to levitate an object or heat it up, but the mage cannot do the same to a beating heart to directly kill someone.  It negates the role of conventional weapons in the world.

Remember that the effectiveness of magic is influenced by ambient conditions.  As a result, magic may be more or less powerful depending on circumstances.  Perhaps there are lines of force in the world.  Perhaps pools of magic.  Perhaps temperature influences all magic.  Classically, perhaps the presence of certain substances (e.g. metal) interferes with magic.  Perhaps even the time of day can influence certain magical activities.  Anything could, really.  Influences might vary between characters.

So magic is about influencing ambient conditions, and ambient conditions influence magic.  That means that one mage can magically affect the conditions that impact another mage's magic.  This can lead to magical combat where two or more mages are struggling to control some kind of ambient condition.  One group wants an area hot while another group wants it cold.  They can duke it out magically.

But one character in the struggle might decide to pick up a club and go beat some heads in on the other side.  Characters are referred to as mages when they spend most of their time doing magic.  But they're perfectly capable of climbing, sneaking and swinging a sword.  Perhaps not as capably as other players and characters per the PAC, but they can do the basics.

Naturally, any character can affect the ambient conditions that influence magic.  If metal interferes with magic, a heavily-armored warrior who runs near a mage might mess up the spell being cast - or that is being maintained.  That might produce comical results as the unthinking warrior in the group walks near the mage who is actively lightening the group's climbing rogue.  The magic is interfered with and the rogue falls.  The warrior's act was completely accidental.  Probably.

I like the idea of this system because it permits players to play around with the systems, to experiment and explore in order to create a variety of effects.  It also returns damage to the hands of those wielding physical weapons, causing battles to return to the employment of relatively standard tactics.  Magic will have a definite impact, but not in the sense that it does in current MMOs.

One last tidbit that I'd throw in is that magic's ability to influence ambient conditions should be applied universally (and modestly).  That is, if the ground is slippery, it is slippery for everyone.  If a magical barrier is placed to block attacking enemies, that magical barrier will also block friendly characters.  By making magic into something that alters conditions, we end up with an essentially medieval environment - with modest variations.