EverQuest's Good Design

I will make an argument for what I will classify as the good design elements in the first part of this critique. ‘Good’ is often subjective, but regardless these are all characteristics of the game worth discussing.

I will be making a lot of comparisons to World of Warcraft because the genre has become WoW clones. WoW is the measuring stick for all games of the genre because it’s such a behemoth. WoW is objectively a great game (at least earlier eras), and vanilla WoW is one of my favorites, however I’m eternally frustrated that WoW is copied and EQ is not. Game design often has pros and cons and there is no one best way to do things. I didn’t intend for this to become a WoW roast when I started writing but please indulge me.

Player Interdependence and Required Socialization

“I’ve always said, and this dates back to before Beta 1, that EverQuest is primarily a group-oriented cooperative game” – Brad McQuaid

EverQuest was deliberately designed to be solo unfriendly. While some classes could level solo fairly easily, others could not without much difficulty. (soloing became less feasible at later levels however) Natural health regeneration was extremely slow, and this ensured that healers were always in high demand and useful. This extreme dependency on a healer meant that healers did not need to have damage output on the level of other classes. WoW in contrast gives every single class a similar level of DPS because the leveling game is designed to be highly soloable and food plus drink were added to almost eliminate downtime. It’s only in WoW’s raid game where classes started to seriously differentiate.

In EQ your choice of character to play is tank, melee DPS, caster DPS, healer, buffer/slower/CC. In WoW, while leveling, it’s just melee DPS or ranged DPS. If you bothered to do a dungeon while leveling then two of the DPS classes transformed into a tank and healer for the run by clicking off a stance or form. Half of the subscribers in the early years of WoW never reached level cap so only a fraction of WoW’s playerbase made dedicated tank or healing characters. Class interdependence makes for more class diversity. EQ has more class diversity.

If players have the option to go it alone without drastically reducing their progress rate, then they will largely do so. This is not what you want in a MMOG. Players are often introverted nerds that need a push to socialize. (I am exrtemely introverted myself which may not surprise you) One reason I love EQ is that it forced me to make life-long friends. This can happen in WoW of course but it rarely happens while you’re leveling up in that game and I would argue significantly less so even at the raid level. One of my best pals I met in Unrest in my teens.

EQ’s grouping requires sending invitations. Later games tried to get players to cooperate without invitations, such as Warhammer Online and Guild Wars 2. While these are not bad games in my opinion, and I do think their public quest, invitationless player cooperation events were a noble effort, I feel that these have failed. In these games other players are just NPCs played by people. They could be replaced by real NPCs and you’d never know it. You don’t remember who was there and you don’t travel to other places with them. You’re not on an adventure in a dungeon trying to kill the mob with the sword your tank wants. The server just rolls on a bag at the end for you. This lack of communication also resulted in tanks and healers not fulfilling their class duties and everybody just mindlessly DPSing.

EQ’s interdependence that forced players to actively seek out others and contact them directly is the better design. A cleric or enchanter in EQ might have several groups begging them to join. There is no need to make friendships in a game that doesn’t require you to communicate with other players and/or who are all interchangeable because classes are homogeneous. Distinctive class roles in classic EQ meant that you could not replace the enchanter or cleric with another class and expect to do nearly as well. Befriending somebody of those classes, or a good tank, was valuable. If you liked doing the dead wing in Sebilis then knowing a rogue was valuable. The downside is of course that groups sit around waiting for a specific class before they can do much of anything but once your friend network is built up then this problem is mitigated. There are still lower risk lower reward options for suboptimal groups.

Making friendships in-game is important for player retention. If the players in the game are just NPCs you never speak to then quitting the game is easy to do once you grow tired of it. If quitting the game means cutting yourself off from your buddies and crippling their group because they were so reliant on you then you may stick it out longer.

A side effect of this design is that it results in, to some extent, a self policing community. Your reputation matters in a game world where knowing people is necessary to progress.

World of Warcraft “solved” EverQuest’s problems largely by suppressing the first two Ms in MMORPG. Modern MMORPGs are often just single player games with other people in them. For most WoW players you could make a solo server for them with smart NPCs that behave like players and they’d never know the difference. This does change in the raid game but most players didn’t raid. Single player games are fun however and WoW is fun.

Buffs and Bind

In EverQuest buffs are powerful. Very powerful. In WoW they are not, because they cannot be in a game where grouping is optional. If all classes do the same DPS then granting one class a 50% DPS multiplier buff doesn’t work. Powerful buffs create another need to socialize and act as a force multiplier for groups. A group of five adding an enchanter as a sixth will increase the capability of the group by far more than 20% just from buffs alone. Strong buffs mean the class doesn’t need DPS, healing or tankability to be effective. This is one reason why EQ’s classes were far more distinctive than WoW’s classes. (counterpoint: being a buff bitch isn’t compelling, but EQ can be improved upon)

Clarity and Mana Song were another large benefit to a group. One or both of these in a group could eliminate downtime entirely with good play. Slow mana regen is necessary in a game where wizards out-burst other classes by a large margin. The regen working while standing allowed hybrid tanks to cast their hate generation spells without downtime which would otherwise be required since they’re standing to do their role, so the synergy when them was significant.

Luclin however added the four hour duration clarity buff which was a bad decision because it allowed players to get the benefits of an enchanter without grouping with one. Most buffs should probably be limited to about 45 minutes.

Buffs need to have a significant cost to them otherwise they can’t be powerful. In WoW you can just buff everybody you run into, without stopping, because the mana cost is so trivial and it’s an instant cast. In EQ you might need to memorize the spell, stop to cast for four seconds and expend 30 seconds worth of meditation time to buff somebody. This cost allows players to charge money for buffs, and means raids will need more than one of the class in order to buff the raid. In WoW you could have one druid in your 40 man raid and there is no penalty for this.

The Bind spell was one of early EQ’s most notable dependence abilities. If you were a melee class or too low a level then you absolutely required somebody to bind you if you traveled long distances. While generally I like this kind of thing, the problem I have with it is that melee classes did not have much in the way of equivalent abilities that casters needed, so there was a demand imbalance of sorts that was unfair.

Slow Is Good, Actually

EverQuest is a game where it might take you multiple minutes to kill a single monster, with considerable downtime after. Sitting around waiting for mana to regenerate or NPCs to respawn was a large part of the game. The argument that some downtime is good because it allows opportunity to communicate with the group has been made before and I agree with it. This subject is a good example of why, in game design, one solution isn’t always the best. Slow games work and fast games also work. Honestly I just want to be able to take a piss without having to tell my group to stop for a minute and make them wait because the game has no natural rest periods.

A slower burn results in a higher climax. Accomplishments in EverQuest felt more satisfying because they required more effort to achieve. Merely reaching the level cap in the original game was a big deal. Some items were valuable for years, such as epics. I had a Sceptre of Destruction instead of my epic, which I quite treasured and used until I got a Blade of War.

“Patience is bitter, but its fruit is sweet.” – Aristotle

Long Camps

I would go further in that I would make the case that a game that requires you to sit on a spawn and kill an NPC every 30 minutes while you spent most of the time AFK is fine. The Paw of Opolla quest comes to mind. A game where you might watch TV or even play another game at the same time (I liked Street Fighter) is fine, just so long as the majority of the game is not this. Time sinks are largely what gives items value and if you can get all your loot in ten minutes then nothing will have value and you will have no economy to speak of. MMOG items are bottled time.

Paw of Opolla spawns were very rare because the item was strong and obtainable at a relatively low level and the spawns were in places groups would level. The alternatives are that all great items must come from the highest level content, or instead of AFK camping greens you’re mass slaughtering greens, or in WoW’s case doing the same instanced dungeon a dozen times in a row. Some people might prefer a nice Street Fighter camp once and awhile to those. Although I would say that it’s the job of the designer to make the time spent chasing items be entertaining, but having some optional camps that are AFK time sinks is not the end of the world. The Opolla camps were actually an opportunity for mid level players to camp something valuable while leveling. I don’t see this camp as being much different than camping lower level mobs for tradeskill materials.

Downtime reduction is a metagame in itself. This will often determine your optimal group composition, but also make the choice of location to set up at not a simple one. Tactically designating a non-tank puller and sending him out to pull before all the mobs in camp are dead is something that the better players would do. Downtime adds another dimension to the risk vs. reward calculus: even if the group could handle a dungeon with desirable loot or an experience bonus, an easier camp (like one where single pulls were easy) might result in more experience over time because the downtime and risk of death are lower there and this frequently made the best choice of camps not obvious. Forcing players to make difficult decisions (in all aspects of the game) is what designers should strive for.

EverQuest was game where achieving full BIS gear was more or less impossible unless you ran a very successful guild with an iron first or your DKP system was miscalibrated. This is not something players tolerate nowadays because they’re used to modern games, but I think it’s better that players not have everything. (perhaps not quite as strict as EQ was) It’s better to always have something to chase after, and that everybody’s gear profile is unique. It makes the raid drops more exciting and meaningful if they’re not common. If people get BIS then they will quit, and you don’t want people to quit.

Slow Leveling

When EverQuest launched, the entire end-game was two dragons and they weren’t even itemized until 3 or 4 months post launch. Fear and Hate were added mid-era. In EverQuest the leveling journey was the game and not some long tutorial before the “real game”. This seems quaint now, but in EQ you were expected to have fun in the leveling process and not want to rush to the level cap as soon as possible. The intent of EQ’s designers was for it to take the average player a full year to reach the level cap. EQ took roughly four times longer than classic WoW to reach the level cap.

EQ is not brutal and unforgiving just because leveling and camps took longer than in other games. I reject this assertion. The designers didn’t even know they would make an expansion when they were designing it. Modern gamers just don’t like seeing other players with more time advance faster than they do. They become envious. Also modern gamers want to play other games, and so they make excuses and say nobody has the time to play games like classic EQ nowadays. That’s silly because people work similar hours today than they did in 1999. My guildmates were all adults with jobs back then.

The first level 50s in EQ appeared in about eight weeks. These people were mostly guard campers, so were classes that could solo well, like druids, although Furor’s rogue was one of the first. The guards were later nerfed with higher MR and longer respawn timers. Their /played was roughly 700 hours.

In 1999 the level race wasn’t much of a thing. Naïveté is bliss. Players back then wanted to stop and smell the roses because the world was new and unexplored. This is why new games are important instead of replaying older ones endlessly. Slower leveling is not a drag if you have many diverse locations of hunting grounds and there is still some mystery remaining about them.

EQ is obviously a subscription monetized game. If players would have hit the level cap in 1-2 months and then had nothing else to do, they would have unsubscribed. This is why WoW has “cycles” with sharp drop-offs. So while you may call WoW more “mass market”, that market doesn’t stay subscribed year round.

In WoW you follow quest rails that take you on a pre-planed route around the world. Doing dungeons is entirely optional, for everybody. Most players don’t do the dungeons and even the veteran players will generally skip all but the higher level dungeons. This is a huge waste of content. Slower leveling encourages players to explore more of the mid-level game. It’s otherwise unappealing to attempt to get a group together and go into mid-level dungeons and accept a lower exp rate just to get a couple of pieces of equipment that are made obsolete soon after you earn them because you’ll level ten times by next week and pick up a green replacement in the auction house. In EQ doing dungeons to level was the norm, not the exception.

Even though I was somebody who liked to level fast, I now feel that level races are bad for the game and this should be curbed. During WoW Classic beta, Blizzard had a staggered level cap to force testing of content. That they had to do this is telling in and of itself, however I recall watching YouTubers talking about how they enjoyed this restriction because it allowed them to fully explore the mid-game content.

I am in favor of staggered level caps for an EverQuest-like game. It’s just too much of an advantage to rush to the level cap and monopolize content before the crowds get there, and incentivizes unhealthy playtimes which players will feel compelled to do because of the modern meta. A lot of players nowadays become resentful from seeing other players who play unhealthy amounts of time advance much more rapidly and instead they just choose to play servers where leveling is very fast. With staggered level caps you avoid that problem while keeping longer leveling times. How long these caps should apply is subjective, but certainly not anywhere near the length of the expansion/era. A serverwide cooperative task to unlock the next level tier might make it more palatable and fun. Softcaps are also an option to maintain a kind of neutered level race.

Slower leveling is also important so that you don’t outlevel your friends and lose the ability to group with them. EQ’s experience system was designed such that later levels took much longer than earlier levels (more so than typical) and this was probably a good idea. Particularly because you need to hook players in the first day and weeks and faster levels at the start is important for this. EQ also doesn’t give many skills to players in the early game so it makes sense for these levels to go quickly. EQ was also fairly generous as to the range of levels you could group with, and players could still do meaningful damage to NPCs well above them in level. WoW in contrast is designed such that players cannot do anything to NPCs that are five or more levels above the player.

Challenging Games Are Better

“We had a mandate from Brad that this was a game that would be so tough that you would need a group to do most of it” – Geoffrey Zatkin

“I never advertised it to be an easy game, that’s for sure.” – Brad McQuaid

Dark Souls is a game that is notorious for its difficulty, and yet it has had a massive impact on gaming and game design, having influenced games in entirely different genres such as Final Fantasy, Destiny, God of War and Path of Exile. “Soulslike” is now an adjective in common usage. But somehow having strict penalties in MMORPGs has been taboo since 2004. Sometimes I like to go on the Fires of Heaven forum and say something to the effect of, “EQ’s original gameplay was good, actually” just to see the resulting flames. Classic EQ is too “niche” to be deserving of a clone, we’re told, even though it was the top selling online subscription game for five years. I reject this assertion. We’ve been living on a moonscape of WoW clones for two decades and the genre is on life support because of it.

There is a famous Medium article from 2014 about a father who gave his young son older era video games before giving him modern video games. The child had to progress through the generations of older games before he could play modern games. The result was that the child ended up preferring the older games, which were more difficult, and became a video game prodigy able to win at the most difficult games. I consider this to be one of the greatest experiments of all time.

Hard From The First Mob

EverQuest was designed to be challenging. Other challenging games exist but I think EQ is somewhat unique in that it’s challenging during the leveling process and from the very first level. EQ can often feel more challenging in the earlier levels than in the later levels because the player’s kit is so limited in the earlier levels which means a lack of survival tools like run speed buffs. My first character was an erudite wizard which had night blindness in a forest starting zone. I learned quickly to pull mobs to the zone wall and fight them there.

In EverQuest the first mob you attacked had a good chance to kill you, depending on your skill rolls. Skills started at zero in the early years. (modern EQ starts with maxed skills at level one) The smart player would make sure to fight the first mobs near a guard before the combat skills had a chance to increase. Dumping a few points into your weapon skill at your trainer was also a good idea. That EQ is difficult from the very first fight is something I adore about the game and wished other games would do. I don’t want to play Darksouls MMO edition but nor do I want to wait many hours before I fight something that might make me flee for my life. In other genres this is considered normal, however in RPGs the trend is for early levels to be a long tutorial, which I think is unnecessary and lame. EQ itself proves it to be unnecessary. (EQ did have a small tutorial application that was separate from the game however)

Night blindness is something most designers are too afraid to add to their game. The argument against it is sound, however there are sound arguments against many cool things and if you always play it safe then you end up with a bland game. Night blindness works in EQ because it’s only really an issue at the earlier levels and if the player really hates it they always have the option to play a non-human race. It also jives with the lore and traditional D&D really well. Night blindness adds another way to improve your character by obtaining better light sources (e.g. lightstones) or getting equipment that grants night vision. The game actually looks better in human vision because infravision and ultravision colorize the entire game screen and the lighting looks better in human vision.

EQ’s Core Principle: Risk Vs Reward

“It’s all about risk and reward. We are quite generally against reward where there is no risk” – Brad McQuaid

RvR was something mentioned a lot in the early years but later fell out of use. I am a disciple of Brad and thus I believe strongly in RvR.

RvR means that to get something great, you have to take a significant risk for it. In modern games this usually means risking your time and nothing else but in EQ that could mean experience loss or permanent corpse loss. EQ RvR meant actual danger, not merely an inconvenience.

EverQuest is designed in a way that punishes failure but also rewards skill in a more extreme way than a typical MMORPG, at least while leveling up. In a WoW clone, a really good player might level twice as fast as a mediocre player. In EQ an exceptional group of players might level five times faster than a group of mediocre players. This may be frustrating for bad players but skilled players enjoy it. Strategies like PBAoE groups can be extremely rewarding and satisfying or you may end up with less exp than you started.

Raid loot was also not unlimited due to the uninstanced nature of the game. If you had a Vulak item you might of been only one of a few people on the server with it. Nobody came close to having full BIS gear back then. WoW’s raid scene was quite challenging in a way that contrasted with the leveling game, particularly in late vanilla era, but instances meant that your gear was never really special. If Vulak is only killed a dozen times before Luclin, then everybody with Vulak loot has something to boast about in a way that doesn’t exist in instanced games. Competition for spawns adds another dimension to the difficulty, although I do feel that anti-poopsocking and griefing prevention measures are necessary for a healthy game and these are something EQ lacked.

EQ was made long before in-game shops existed. In the early days if you had some flashy piece of gear, it meant that you earned it. Ebaying was a thing before Kunark launched (I myself sold things) however the top end items weren’t generally available on ebay. This was largely due to no-drop tagged items but also because the practice wasn’t yet widespread and your guild wouldn’t tolerate you RMTing your Alendine robe. As far as I am concerned, buying “cosmetics” is not very far away from buying “power” in a very social game and it diminishes the enjoyment for others. Looking badass was a large part of the reward for succeeding in the game and if some idiot can just buy cosmetics that look as good or better then it reduces the sense of satisfaction for those who earned their equipment.

The Case for Experience Loss

EverQuest’s death penalty is actually fairly lenient compared to ’90s games. Permadeath was something games used to commonly have, even very grindy games.

Why is experience loss a good thing? Well, one might ask why hardcore modes exist in any game. Experience loss is merely a much softer hardcore mode. Victory is sweeter if failure has a consequence. Combat is more thrilling if there are consequences for death. A penalty limits how many attempts you can make at an encounter. This is actually important in a game where you can bind outside of Nagafen’s lair, but it also forces you to actually think about what you’re about to do and how to do it.

Experience loss adds another variable into the calculus of deciding where to set up camp when leveling. A camp may be doable but if the risk of death is too high there, then exp over time in that spot may not be worth it simply because of the experience loss, and particularly so if no cleric is available. This forces players to make hard choices. Skilled players will correctly gauge the ability of the group and select the appropriate location to set up camp. Mediocre or bad players will end up losing more exp than they make and get frustrated.

WoW got rid of experience loss then quickly discovered that they needed some kind of punishment and so implemented equipment repairs. Even Blizzard acknowledged that some kind of penalty was required. A difference between experience loss and gold repairs is that you cannot RMT your experience penalty away.

Experience loss also created a lot of demand for the cleric class, and allowed necros to get a resurrection spell without taking over the cleric’s job. This is another example of limitations and obstacles creating opportunity for class interdependency and avenues for character advancement.

In EverQuest you not only lose experience on death but you can de-level. This is important otherwise experience loss at max level would be meaningless, however there are some advantages to being able to reduce your level. One is that grinding AAs at level 60 is often faster than at higher levels due to the way the experience bonus worked. Another reason might be so that you can keep grouping with a friend who fell way behind. The trivial loot code was a reason. On PvP servers you might stay below a certain level to be able to attack certain players or prevent being attacked.

The Case for Corpses

“Some of my favorite memories playing MUDs and then later EQ are the corpse runs. Having friends and guild mates log on late into the evening and risk everything to save us - it was challenging, even frustrating, but in the end the comradery and team-work employed made for the most fun I’ve had playing online games.” – Brad McQuaid

Corpse retrievals (CRs) are the biggest part of RvR. In 1999 permanently losing your corpse was a real thing. More so in the early levels, but even late game this was possible– particularly in the planes, albeit unlikely. The idea that you needed a backup set of gear in the bank was actually a thing once. I knew somebody in 1999 who lost his corpse in Fearplane because no guild had cleared the zone before it poofed. Corpses poofed in 24 hours of online time and 1 week of offline time, so back before Fearplane was on farm corpse loss was a real risk in that zone.

Sony later added corpse summoning and stopped designing zones with very hostile entrances so this gradually faded as a risk, which I feel was unfortunate. Were I to design a custom content server I would nerf necro corpse summon to a short range and design some raid zones with serious risk of loss à la Fearplane. Part of the planning should involve how to retrieve corpses in the event of failure.

Some of you probably don’t like that. But consider that in an uninstanced game, you can zerg the crap out of content if you rally enough players. In this harsher CR world, one might make deals with other guilds just to get them to help recover your bodies. In EQ, Sony could be lazy and permaroot bosses with necro summoning in the game. But on a hypothetical server with serious CRs, bosses would need to be mobile and the location you pulled a boss to and fought becomes a big consideration. One might design the zone such that one spot makes the boss easier to kill but makes CR much more difficult than another spot with easier CRs but makes the boss harder to kill.

A lot of the arguments for experience loss also apply to corpses, but in this case a rogue or monk may be required to save you before the cleric does. Personally I love the visceral image of seeing an entire raid lying on the ground from failure. The time spent recovering from a raid wipe also allows for a discussion of what went wrong. In a game where players get a limited number of attempts on a boss, the fight can be easier and still be challenging because players get less practice at it, and it feels much less repetitive. Fights feel more epic if done infrequently.

During WoW Classic I was in a raid guild attempting a particular raid boss that we kept failing at. Because of the lack of a real death penalty, we were making as many attempts as we could on it without stopping to strategize or analyze why we kept failing, perhaps in the vain hope that we would get lucky in one of the attempts. Losing that many times in one night demoralized the guild to the point it literally disbanded.

A game with no serious death penalty results in what I call DDR gameplay. DDR meaning Dance Dance Revolution. If players can make a dozen attempts at a boss in one night, every night, then in order to make the encounter challenging and satisfying, the design must lean on forcing the players to make quick, sudden reactions to telegraphed or timed attacks which result in failure if not dodged in time. This is why WoW is full of fire-on-the-ground boss effects and attacks that would wipe the raid if one player wasn’t paying attention, which in turn makes larger raid sizes infeasible. In WoW raiders had homework assignments: watch videos and study boss fight mechanics before your raid. A harsher death penalty allowed for EQ bosses to be more forgiving, with simpler mechanics and attacks, and for larger raid sizes. Lack of penalties also result in cheesy gameplay in games. (“zerging” or “bind rushing”)

Corpses as a mechanic enabled some creative gameplay. Corpses would be used as a means of teleportation by dragging them, as a means of storing no rent items, as a means of bottling exp, as a means of dropping aggro on pullers, as a means of keeping your equipment safe before doing something suicidal. Corpses give players options; although some of these are debatable as to whether they should be allowed. This kind of open-endedness can be difficult to design around. Lady Vox used to drop an item with resurrection charges which enabled rogues and monks to resurrect a cleric deep inside a raid dungeon (with corpse dragging) which allowed raids to skip trash content, so this item had to be removed from the drop table. Corpse dragging through walls was also common.

Designers Must Err Toward the Difficult

“Did you know that the first Matrix was designed to be a perfect human world? Where none suffered, where everyone would be happy. It was a disaster. No one would accept the program. Entire crops were lost.” – Agent Smith

In the movie The Matrix, Agent Smith tells Morpheus that humans rejected the paradise world created for them. I think that is analogous to video game death penalties. Some kind of non-trivial penalty is expected, or the game feels wrong. Imagine playing the NES game Contra and by default you had unlimited lives and you didn’t lose your weapon ability on death. Even the most casual players would reject this.

MMORPGs must err on the side of being difficult and with significant penalties when designed. There are several reasons for this which I will outline.

First is that players hate things being taken away from them, and this includes making content more difficult after they’ve first experienced the easier version. This is human nature. On the flip side they generally approve of and are even happy to see content made easier, within reason. The hardcore players won’t like seeing their content made easier after they’ve beaten it but the casuals respond well to this. (endeavor to get it right the first time however) If the designer unintentionally makes content too easy or penalties too lenient then tuning them upward later will alienate all players. You simply cannot put in EQ’s death penalties after launch and expect your players to stick around, but you always have the option of lessening them later. EQ’s experience penalty until nine weeks after launch was actually double what they decided to keep it at. Once upon a time in EQ, if you died in level 46 you lost half your level.

Second is that MMORPGs naturally become easier over time because of mudflation, spoilers, power creep and more higher level players in the world. In EQ’s case computer hardware and UI improvements were also a significant factor. Cheat programs unfortunately are another factor. EQ kept getting easier with every expansion to the point where, in PoP, Sony had to just massively increase the NPC power to keep the game interesting and this was well received because it was new content.

Third is that accurately gauging difficulty of raid encounters before players actually attempt them is not really possible. The designer needs to have a very deep understanding of the game’s combat and how difficult the other raids are to even attempt to estimate how difficult a new raid will be. Vyemm was notoriously more difficult than Vulak but that wasn’t intentional AFAIK. I made a few custom events for The Al’Kabor Project and I intentionally designed them to be somewhat harder than where I was targeting for and they still ended up being easier than I had intended. Players will often surprise you with what they are able to overcome.

The Plane of Fear, My Favorite Zone

“The Fear Plane is now on-line. However, before adventuring to the Fear Plane, we advise you to heed the following advice: Traveling the planes is not like visiting a dungeon. People die in dungeons. People die hard on the planes. Dying on a plane means that you might die in a place where your body may be extremely hard to recover, or in an area that you have no idea how to get back to. Big, nasty monsters will jump you at inopportune times, spells may work differently and once valid tactics might go right out the window. The planes are dangerous places to adventure, and massive player slaughter will happen on a regular basis. We are not saying this to goad you in to going - this is an honest disclaimer. So don’t say we didn’t warn you.”

– July 26th, 1999 Patch Note (my favorite patch note)

In 1999 I had a little obsession with this zone. The concept of a harsh break-in before getting a bunch of goodies intrigued me. I would zone in naked and run around the zone with SoW just to explore it. I made a map of the zone which is still found on Project 1999’s wiki. This zone really separated the men from the boys back in the day. You needed a plan just to stay alive inside, let alone kill something.

Fearplane’s design was very simple: strong see-invis mobs with huge aggro radii and players landed inside a NPC spawn area from a one-way zone-in. The amount of difficulty and depth of strategy resulting from this very simplistic design is elegant.

The general strategy for this zone ended up being, one person zoned in to train away gate mobs, the rest of the raid zoned in and moved into the southwest corner and possibly camped out first to clear aggro. The raid then sat there and waited to be attacked, killed as much as they could before being overrun, then evaced out when too many mobs showed up. This process was repeated until the gate mobs were dead and fear was broken in.

The only other raid zones which had a hostile zone-in were Hateplane and maybe Veeshan’s Peak if the wurm happened to be there. I wish Sony had kept making zones like Fear and Hate because I feel they are good design and the most memorable. Fearplane embodies RvR in a way no other zone does, although the big flaw with it was that other guilds could take advantage of your break-in and get their armor without earning it.

Consider that somebody camping Cyndreela for exp could accidentally zone in and lose his corpse forever back then if no guild cleared it. 1999 EQ was a special time that should be replicated in another game.

Zone Keys

Sony implemented zone keys for Kunark. Sebilis, Charasis and Veeshan’s Peak required a key for entry and originally soulbound items didn’t exist. This lasted for about three months before they implemented soulbound keys, but boy were those some RvR three months. (this would be an example of erring on difficult) This is not particularly great design but the attempt to make the dungeons risky is admirable. I could make the argument that intentionally making content unforgiving early on with the full intent of softening it later is actually not a terrible idea in some cases, as it allows your hardcore players to overcome difficult obstacles for a period of time while not excluding the casual gamers from content forever.

Skyplane had in-zone keys that poofed when you zoned out. Sky is probably the least popular raid zone in the game, and for some good reasons. I do like the general concept of a raid zone with staged progression like that however. I will discuss my opinions of possible improvements to the design in a later section.

Veeshan’s Peak was intended to be a brutal raid zone in a spirit similar to Fearplane. Sony in fact had a policy to prevent GMs from assisting players in the zone, because back then you could sweettalk GMs into CRing you. Here again I admire the attempt but the design was poor. Player teleportation spells were disabled in this zone and the zone-outs were all behind the raid bosses deep in the zone. More about this later.

Trivial Loot Code and the Yield Rule

As early as May 31 1999, Brad posted on Usenet about “annoying camping” he had witnessed in-game:

The annoying camping to me was by players waaay too high level for where they were at. They just sat there, typically alone, with impunity, waiting for the rare NPC or NPC with the rare item to spawn. No challenge of combat, no challenge of forming a group to handle the area.

This of course offended him because his core tenet is RvR. He then proposed the idea of what eventually became the Trivial Loot Code which was implemented during Velious for the Warrens zone on March 14 2001.

I agree with Brad that this is a big problem, perhaps more so because this prevents lower level players from camping the spawns, but my solution would be different. I think the Trivial Loot Code probably would have been tolerated had it been implemented before the game launched in March 1999, but as I mentioned previously, gamers really hate having things taken away from them and they will revolt on you if you put in strict limitations after launch. This is why you need new games instead of replaying the same one forever.

My alternative solution to this would be two fold:

One, green cons should assist like blues do. It was unwise to make them not assist in the first place. In dungeons they should also attempt to flee early and try to call for help and swarm the player. EQ actually did have flee-early behavior but it strangely was limited to mobs under level 18 or so. Those would flee at 50% if they were green. Players being able to single pull their way through the dungeon and in the named mob rooms is largely why they could solo the mid-level dungeons. Using smarter AI to handle this is much more seamless and just makes sense from a logic standpoint which makes it much more tolerable.

Two, if the NPC had anybody else on the hatelist that is low enough in level to be able to receive exp, then the loot transfers over (not exp) to them and their group if a high level player gets kill credit. This would be a server enforced implementation of the “yield rule” so GMs would not have to get involved. It also allows for higher level buddies to help lower levels to do the camp which the TLC would deny. Some sanity checks might be needed for this to prevent griefing.

The TLC is somewhat heavy handed. It feels bad if you went to the trouble to level up and you couldn’t enjoy the benefits of your progress. Also at some point mid-level dungeons become empty after the server becomes top heavy, so trading for mid level loot becomes problematic in that scenario, although one could argue that this creates a good reason to make alts which might increase player retention. Players may actually appreciate doing camps “legit” as it’s more entertaining than camping greens. Players used to have fun making super twinked level 52s to do Nagafen and Vox with.

The “yield rule” is something Brad proposed and was almost implemented but wasn’t due to criticism. They at one point were planning on having GMs kick out higher level players from dungeons if they were camping something a group of the appropriate level wanted to camp.

No Guarantees

EverQuest had a “no guarantees” ethos in its design. Perhaps nothing exemplifies this more than spell fizzles. Fizzles are something modern games wouldn’t even consider implementing and players obviously dislike but there are good reasons for the game to have it. I would presume that the original intent of them was to add an element of risk to the game (you think you have enough mana to do X, but you really don’t because of randomness) but there are other benefits. Fizzles make macroing and botting significantly harder, which is a big problem in a slow paced, grindy game. Fizzles create a reason for spell skills, and something for stats to improve. The game feels bland if the only number that goes up is your level. Fizzles add a reason to use lower level spells over higher level ones.

A max level player could not blindly autorun across East Commonlands and expect to never die doing this. This was a zone fifth level players leveled up in. EverQuest had high level, aggressive NPCs in zones which would run over newbies and even higher level players. This made the world feel dangerous and made the player stay aware of his surroundings. Seeing my first hill giant in West Karana was a fun experience for me. This is good design and should be replicated in other games.

Tradeskills had a 5% minimum chance of failure for all recipes. No guarantees. On the flip side it also had a 5% minimum chance of succeeding.

Invisibility

EverQuest granted players invisibility skills from a very early level and to many classes. In most other games invisibility is considered too powerful to do this, so one must consider why EQ does. I think the answer is that the game was made to be difficult and they didn’t want CRs to be impossible. It’s also because MUDs granted players invisibility at an early level and perhaps for the same reason, and EQ was largely based off of MUD design.

I think EQ handled invisibility well. It adds another dimension to travel with a nice RvR factor: the invisibility spell could fail at any time randomly so it adhered to the “no guarantees” principle. Strategically you have to consider which NPCs will see invis and which won’t, which often is not predictable as many dungeon NPCs had a random chance to see or not see. Unpredictability is often a good way to balance powerful abilities, and powerful abilities are fun to use but problematic as they can trivialize the game if they have no downside.

Sneak was another form of invisibility that required you to stay out of an NPC’s front arc to remain hidden from it. Mobs that could see spell invis might not be able to see through sneak and vice versa. Rogues could hide while sneaking which granted the benefits of both at the same time, plus invis to undead starting in late vanilla.

Brad wrote a post about his opinion on rogue stealth in July 1999:

“The Hide skill was actually much more powerful in beta, so much so that it was very abusable. Because of that, we limited it significantly. I am more than willing to entertain ideas as to how to make it more effective yet STILL balanced and NOT abusable. Keep in mind that a Rogue traversing the depths of a dungeon with impunity may be acceptable in fantasy literature and such, but it is not acceptable in a game like EQ.” – Brad McQuaid

I include it here because he’s right, at least for higher level zones, and because there is controversy over how good rogue stealth was before Shroud of Stealth AA in PoP. Admittedly I consider the issue unsolved (speaking as an emu dev) and is one of the top items I’d like to see a detailed explanation of, however rogue stealth was nowhere near SoS in classic.

Immersion and Environment

EverQuest, being the first 3D MMORPG, made a considerable impression on those who played it. Being the first gave it an advantage as the transition from sprites to polygons is poignant, but I’m going to argue that EQ excelled in immersionness more than typical MMORPGs and this is a core reason why EQ was so great. We need more of EQ’s design ethos of making a world that feels alive.

Immersion is something that is very intangible and unquantifiable. For this reason it is often disregarded or back-burnered. “Powergamers” don’t write forum threads about how great the trees sway. I would argue that this is something that is subconsciously appreciated even if the forebrain is focused on slaying the dragon, and that ignoring immersive aspects of a game is a mistake. Immersion will prevent people from thinking that your game is just another clone of something else.

First Person View

EQ was played in the first person view in the early years. Starting in Velious I started using the top-down F9 view because third person views are advantageous but particularly so for tanks. Mid-PoP Sony added “EQ Cam” which is the mousewheel camera most people know about and use today.

I think it’s a shame the genre moved away from first person view. First person camera is much more immersive and pulls the player into the world in a way a floating camera does not. Looking into the face of another player’s avatar from eye level makes encounters with player characters feel more personal. The world feels much bigger in first person view. The hill giant is scarier when you have to look up at it.

A consequence of third person cameras is that it makes the developers make enlarged structures to accommodate the camera. All the buildings in WoW are gigantic. Roofs are 30 feet above the player’s head or more. WoW can get away with this because it’s very cartoonish but I hate this in other games. I want more Qeynoses. I want small cities. More Rivervales. Small is charming. I want more packed into a smaller space, because it feels realistic and it means you’re not running a mile just to get through town. Running through Orgrimmar takes all day if you’re not mounted. Just try to find the Karana faction guildmasters in Qeynos though. The city is small but it doesn’t feel nearly as small as it is. Unfortunately because it’s now costless to make large cities and structures, developers make them. Luclin’s stupidly large buildings sucked. The entirety of South Qeynos could literally fit inside Lcea Katta’s building plus courtyard. (I checked) Look at pictures of old European or Chinese villages. They’re very quaint and beautiful. Make these in games.

Weather Is Awesome

I am a huge fan of weather in MMORPGs. This must be in at launch of a new game. When I think back to what I loved about the very first days of EverQuest, I think about hearing the rain and thunder in Qeynos while I was fishing.
Weather has a large bang for the buck immersion wise. WoW added it after launch and I do really like what they put in but it could trigger more often. Feralas is my favorite WoW zone (OK, maybe tied with Stranglethorn) largely because of the rain there.

Nighttime and Fog

I went over night blindness already. It’s a good thing and I hope more games do it. I want the environment to make the game harder at times. This fits well with the “no guarantees” ethos. These are magical fantasy worlds but a certain level of realism is needed to pull people into them. MMORPGs should strive to pull players in and not be an arcade game. EQ’s motto was “You’re in our world now” after all.

A lot of EQ zones had short clip planes. I do not believe that this was because the trees lagged the zones, at least not in all cases. (GFay maybe) Sony could have easily enlarged the clip planes some time later when PCs got stronger but they chose not to. An October 1999 patch allowed players to enlarge clip planes but not for zones with ‘heavy fog’ and mentioned forest zones by name. Sony probably set short clip planes for immersion effect but also to make the zones feel larger and more challenging. Toxxulia is actually rather small and if you could see far it would feel small. Short view distances also happened to make the game more difficult in a way I like, but it was rather unfair to Erudites. Dark Elves at least had ultravision and a better populated zone. Feerrott fog makes sense because at very low levels the stats ogres had did make a significant difference in combat and having a harsher starting zone is a way of giving Ogres a drawback. Same for Trolls.

Toxxulia could have been improved by adding some juicy camps in the zone (easily solo pulled, low risk) to compensate. Put some of them along zone walls that are easily hugged. The zone lacks paths as well and should have had some. (poor Erudites) Also put in some nice newbie drops and a decent exp quest somewhere so Erudites don’t immediately hop on the boat and create a reason for humans to take the boat.

Feerrot had roaming Bouncers that made the zone dangerous to non-evils. You couldn’t see them in the fog. This was great design. If you did CT in the day, you had to zone over to avoid Bouncer Hurd many times. Good races should not be roaming the Ogre newbie zone without concern for their safety. Ogres are supposed to be dangerous.

Underwater ‘fog’ was also great. I lament it being removed in modern EQ. Kedge is one of my favorite zones in part because the fog there made it really spooky and threatening. The Phinigel room felt terrifying. Excellent zone. The atmosphere in there was amazing.

Flavor NPCs

MMORPGs should have NPCs that exist solely to add flavor and immersion to the world. Not every NPC should be involved in a quest, be a skills trainer, be a merchant, be intended to level from or drop loot. Even the ones that provide some function, like city guards or merchants, should occasionally walk around and say dialog. Make NPCs that chit chat.

Qeynos was the first city built for EQ and the developers seemingly had more time to work on it, because it has the most “flavor” of any city. Guard captains tell weaker guards to stand up straight. Merchants and trainers walk around. Differently named guards spawn at night. There are NPCs that exist in this city that don’t do anything other than to make the city feel alive. It’s a big shame other cities and other games didn’t do more of this. Freeport’s bar however actually had a NPC vs NPC fight script that triggered periodically which was sadly removed in the zone revampt.

Flavor NPCs add mystery because they leave the player wondering if they are involved in some quest or have a purpose. By having them you can fool players into thinking the game has more quests than it does. Mystery is a good thing to have in a game.

EverQuest actually had some NPCs that spawned in multiple zones, to provide a sense of them traveling the world. (e.g. Sir Edwin Motte) I an unaware of any other MMORPG that does something like this. It’s cool.

NPC Behavior

EverQuest NPCs had behavioral quirks not found in other games. NPCs turned to look at you if you stood nearby them, even if you did not interact with them at all. When hit in combat NPCs would be pushed backward like a player would. Some NPCs picked up items off of the ground and did an emote when this happened. NPCs had collision checks with world geometry to prevent them from walking through walls. Dragons would get stuck in small doors and this is why dragon model NPCs were immune to magic from outside of melee range, because in 1999 players would pull them to doors and get them stuck there.

Melee push in particular is unappreciated. It was removed because nobody saw the value in it. 1990s Sony themselves may have put it in as some sort of immersion thing, but it had important benefits to the game. Nagafen and Vox AoEs were centered on the target and they were not PBAoEs. (which is a mistake, but anyway) So it was possible to put the tank in a location where players who were meleeing the dragon would not get hit by the AoE. If melee push isn’t in then avoiding their AoEs becomes trivial because the dragons wouldn’t move out of position. Melee push is also a nice anti-boxing mechanic and keeps otherwise unsophisticated melee combat more engaging. Instead of removing it they should have just halved it on raid bosses with more than 32k hp so it didn’t get excessive.

All EverQuest players know what the H button does. In EQ you had to actually talk to the NPCs and this process started by hailing them. You had to try and talk to them to get more dialog out of them instead of being handed some dialog tree with response options. This is the kind of thing modern games don’t do anymore. EQ’s text based MUD lineage shows here. I think this process of NPC interaction is superior to modern dialog trees. It’s much more mysterious and realistic. Exclamation marks are kinda needed when your leveling system is a quest grind instead of a kill grind but if your quests are primarily for items and immersion then making your quests super obvious is a bad idea. The number of quests in EverQuest is actually substantially underestimated because they aren’t highlighted in the game. I would argue it’s actually a good thing that a handful of things stay undiscovered by players even after many years.

Fake Walls and Traps

In Qeynos if you made a rogue, your trainer was behind a fake wall that you had to discover. Actually, here’s a little challenge for you: go log into EQ right now and try to find the zone to Qeynos Catacombs where the Shadowknight trainer is from South Qeynos. See how long it takes you if you don’t already know where it is. What other game had the gumption to do this? Fake walls and floors are a fun thing EQ had and that modern games would reject. This is another example of a very low tech way to add some spice into the game. The nice thing about very social games is that if a player got stuck (in this case, can’t find their trainer) they can merely ask in OOC or equivalent channel for help and get a quick response. In a game intended to be explored without some mad rush to the level cap, you can get away with these things. I don’t want every game to be “mass market”. I want games that do stuff like this.

Vanilla EQ also had traps that actually came out of the wall and crushed or stabbed you. I wouldn’t say these were implemented particularly well, however they look cool. A trap like these that actually killed players outright if a rogue didn’t disarm them first would be neat but would have to be rare and well placed.

Mysterious Design

EverQuest was designed to be more mysterious than other games. A prime example of this are character stats. AC and ATK are two of the most important character stats but players had no idea what they represented. Players often debated how useful stats and equipment were. Some correctly figured out early on that stats in EQ didn’t do very much. (more about this later) I am a proponent of hiding things from the players and making them figure it out. (I don’t think all games should do this, but some should) MUDs used to not show damage numbers and instead output text like “Your slash <*><*>EVISCERATES<*><*> the mob!” which I think is cool. EQ in beta didn’t show spell damage numbers. Sony rushed DD spell damage notices just before launch. In 1999 when a caster nuked a mob, it would output “Mob says ‘Hit By non-melee for X damage’” before they later added a proper spell damage filter.

The caveat to hiding stats is that players still need to have some crude way of determining if an item is better, especially if they cannot parse numerical damages from logs. EQ failed to do this and the stats actually fooled players in some cases, for example the AC hardcaps were not represented in the character AC number.

World of Warcraft has a log system that records a lot of information about a particular raid encounter, and there is a website that can read these logs and generate a kind of combat replay which I actually think is really neat.
Basically I think games should go in one of two ways: WoW’s way, or the opposite. Both design choices work for a game and create different kinds of atmospheres. WoW’s atmosphere is extremely competitive and players try to min-max their DPS output for world-wide rankings. I think this is really neat however I strongly advocate that some games go in the opposite direction and try to make a game that is uncompetitive and is more of a friendly good time with buddies slaying things, without the stress of trying to hardcore min-max and finish the raid in record time. Hiding things from players is how you make a more relaxed game.

EverQuest’s experience bar is rather symbolic of EQ not showing the player things. I think it’s cool and I want a future game to do this.

One thing I have grown to strongly dislike are spoiler websites. Sites like Thottbot are the enemy. Developers of a MMORPG need to resist data mining attempts and not facilitate scraping of game data, like addons that capture items and upload them to spoiler databases. Ideally you want players to find stuff out by word-of-mouth and in-game social networks. It is impossible to prevent spoiler sites but you don’t have to help them and you shouldn’t. There are design choices that can be made to make spoiler sites much less helpful like randomizing drop locations, and I think much effort should be put into this.

Faction

Faction in EverQuest is actually more sophisticated than in any other MMORPG I’ve played, granted I’ve not played a whole lot of them. In WoW, faction is just a progress bar for repeatable quests unless you’re trying to get the pirate hat and parrot. Faction systems have a lot of wasted potential. Velious’ three primary factions were great design but you could take that further: imagine if a three faction system all had a roughly equal number of raid bosses for each side, and with roughly equal quality loot but choosing one faction locked you out of doing the other bosses. This would force guilds to coordinate and agree to separate factions and when to swap them. This would provide a neat way of implementing a kind of raid share policy without GM enforcement or hard coding a rotation system.

I do think that faction hits should be harder to avoid however. Some kills should force the player to take the hit even if the player died, logged out before the mob died, etc. If you participate in slaying King Tormax then you should be forced to suffer the faction hit. Forcing hits allows designers to create more elaborate and consequential faction designs.

12 Races, 12 Starting Cities

EverQuest’s original design vision was to create a large world with a lot of content in it. This included having many races and those races having their own starting cities which were spread out. This was a bold decision considering the amount of labor required, but it was a large factor in making EQ the brilliant game that it was. In EQ the content was made knowing that most players wouldn’t see all of it. That is how MMOGs should be designed. Not everybody has to see everything and that’s OK. Nowadays this is considered foolish and a waste, but I strongly disagree.

If you’re making a game which has a subscription fee and is supposed to keep people engaged for years then it needs to have a lot of content in it, obviously. When it comes to world size larger = better so long as it’s not empty. (although some places need to be like plains areas) Designers should strive to make areas of the game that are the virtual equivalent to Yosemite: there just to look pretty even without some player advancement reason to be there.

Larger worlds allow for trade routes. In 1999 players used to farm items in one region, travel to a far away place, and sell them there for a higher price. Qeynos area didn’t have smithing ore available at launch, (which was probably not intentional) so players starting there could not raise smithing and make banded, creating a large market in that area for smithed items or just selling the ore. Items like minotaur axes and dervish rings were common drops and desirable, so they were often transported for sale.

Less Is More: Why EQ’s Technical Unsophistication Resulted in More Depth

EverQuest was made before sophisticated game development tools were around and by a team of two dozen people. These constraints are a significant factor in why the game ended up being what it is. The game was actually intended to be less of a sandbox compared to contemporary games like Ultima Online, but since later games leaned so hard into the theme park model, EverQuest became known as being sandbox-like. Sandbox design results in significant depth without a lot of developer made content because a lot of the gameplay is emergent, although game balance becomes difficult.

Why Instancing Is Terrible

Instancing did not exist in EverQuest until LDoN. I happen to think that LDoN was one of EQ’s worst expansions. I hate instancing and wish games didn’t do it. It’s the laziest solution you could come up with to the problems of a shared world. World of Warcraft is like a MMO/single player hybrid game. It can work as a game, and work well – obviously single player games can be brilliant – but it’s not a pure MMORPG in my mind. Diablo 2 had eight man groups but it’s really a single player game. I consider myself fortunate to have played in the time before instancing existed.

Uninstanced MMOGs are much more complex and dynamic. In these games you have to concern yourself with players who are not in your guild. What these players do impacts your experience in a game, for good and bad. This adds another dimension to the game world. In uninstanced games the players become the content in some ways. It’s up to the game developers to make sure that the vast majority of player interactions are positive.

In an instanced game, other players just don’t care about what you’ve done inside your instances, but in EQ people wanted to know who killed what even if they were not looking to kill it themselves. You’re simply going to be more invested in knowing who is on your server and what they are doing. And in many cases you can go watch what they are doing in real time and I don’t mean on Twitch.

The Ring War event actually let anybody on the server participate in your guild raid. The Ring War was fantastic design and it’s depressing that it didn’t inspire many more raids like it. Imagine making a war event like this and there were mobs in the event that dropped some relatively desirable mid-level loot which made mid level players grateful to your guild for starting the event. Or imagine giant faction players getting to attack a band of dwarves that spawn somewhere that let them complete some quests but that wouldn’t impact the chance of winning the war so as not to create animosity.

Uninstanced game worlds allow for designs in which servers may differentiate themselves from other servers by player actions. Imagine a NPC war which randomly occurred every 1-2 months and the side that won took over a zone and the spawns in the zone during this time were different than when another side wins, and players can decide which side wins by aiding one side or the other. Hollowshade Moor did something like this but on a much smaller scale. Imagine if Lower Guk were all undead frogs for a month and had drops that were only available in this eventuality.

As a MMORPG player, you want to see the same people running around the world everyday. If everybody you see running around is different from day to day, as it is on Classic WoW’s layered megaservers, then they might as well be NPCs. You can’t form relationships with other players if they’re always different and hidden away in their inaccessible little alternate dimensions.

In non-instanced games, player reputation matters. If you’re a jackass then players can refuse to interact with you. Behavior can be policed in a natural way like this. If you’re using dungeon finder in an instanced game however then there is nothing to stop you from being a jackass.

Classic WoW almost omitted dungeon finder in WotLK, because even they realize that these tools are destroying server community. But as I’ve repeatedly mentioned, players hate having things taken away from them. So while many players did like this move, there was a vocal non-trivial percentage of players who were up in arms and so Blizzard backtracked. Instancing plus dungeon finder might as well be a single player game. An AI could play the other characters in the group and the player would never know the difference. People don’t even talk to each other in these groups.

Being the first to kill raid bosses on a non-instanced server is much more thrilling and praise worthy. In WoW you’re in competition with the entire world to kill things first, but in EQ you were in competition with your local guilds. In WoW if you’re the first to kill an instanced raid boss on your server but a dozen of other servers had done it first then nobody outside your guild will care in the slightest and they probably won’t even hear about it.

In MMOG design, the goal should be that players will run into each other and form relationships without even trying. I.e. not having to check /who and send tells out of the blue then have them teleport to you. One thing I refuse to do in these games is write and submit a resume to a guild to attempt to join one. I find this to be demeaning. The guilds that I join are by invitation or simply happen with little discussion at all. If players bumped into each other inside dungeons they will team up, CR each other, merge when some of them need to log, etc. and in this way get to know one another.

EVE Online is one of my favorite games that I never played very much. What I love about it is that it’s a single global uninstanced server. That is what MMOGs should strive to be, although fantasy PvE RPGs will have a harder time achieving it. A more realistic goal would be EQ but with a lot more zones so fewer servers are needed.

Ultimately instancing separates players and that is the opposite of what you want to do in a MMOG. Putting large numbers of players together is the entire point of an MMOG.

Unique Game Experiences

A shared, uninstanced game world results in more unique experiences for those inside. My experience in EverQuest was unique and not like any other player’s. WoW’s leveling game in contrast is very similar for every player. This is a reason why you will find some old EverQuest players defending the game vigorously while others will deride it and call WoW a superior game. If you started playing EQ ten months after launch, your experience in the game was largely dealing with overcrowding, higher levels preventing you from camping mid-level dungeon loot, and the game being largely spoiled by then taking out much of the mystery. If you played from launch day you got to experience the game world before any druid or wizard could teleport, and exploring Norrath in that time was highly memorable.

I used to solo in Lower Guk and Solusek B on my wizard at the zone lines by rooting and nuking mobs, which would not generally be possible later on due to zone crowding. Reaching higher levels first had large advantages in terms of availability of spawns but large disadvantages in terms of nobody being there to save you. A friend of mine fell off the Lower Guk bridge and needed to get the first level 50 player on the server to get his corpse. Later in the era you could just walk down there uninvis because the zone was so camped. I got to experience the game the way it was intended, which was fighting your way through a dungeon and evacing my group out when we pulled too many. Classic EQ’s real problem was that they needed more servers, or a larger world– not instances. Ideally dungeons would not have more than half a dozen groups in them.

In a cooperative, uninstanced MMOG, who you met in game largely determined what your experiences were. I consider myself fortunate because I really liked the players I met in game, and the guilds I was in. We were highly successful. I would not be writing this if I hadn’t of met these people because they were a large factor in the game being enjoyable. Some players who were unlucky and didn’t find companions to progress with will be far more likely to criticize the game. People shouldn’t call EQ a failure and “niche” just because some players didn’t meet other players they meshed well with.

Uninstanced games require more content, because you cannot make copies of zones. More content means more variety, more options, and reasons to choose these options. Where do we level at? Blackburrow? Everfrost? West Karana? I hear South Karana has a low level dungeon, but it’s a long run. Why would we spend the time running to Paw? Because other places are crowded. Some amount of crowding is actually beneficial because it motivates players to explore. In WoW it’s either do Wailing Caverns if you want to group, else you go find Mankrik’s wife.

Alternatives to Instancing

I don’t know when ‘poopsocking’ became a term but eradicating this without compartmentalizing or mudflating your game is the MMORPG designer’s biggest challenge. Poopsocking is when players will sit on a spawn point (typically a raid boss) for many hours waiting for it to spawn so they can get it before another group of players can. Clever design can be utilized to heavily mitigate poopsocking without resorting to instancing. If you as a MMOG designer resort to instancing then you have failed in your job as far as I am concerned.

The best way to mitigate this behavior is to design the game such that inter-guild cooperation results in the best loot or progress. We’re already designing the games to reward player cooperation at the group level, so it should make sense to have cooperation at the guild level as well. If the guilds maintain good relations with each other then they will share spawns without fighting each other for them. EverQuest’s design unfortunately rewarded selfish behavior. Design the game to reward good behavior instead.

Examples of this:

Make rare spawning (once a month?) outdoor raid bosses that require multiple guilds to slay. If guilds resented each other too much they would refuse to work together. A designer might otherwise make a world event where the guilds are in different zones and need to cooperate in unison in some way.

In WoW there were rare recipes that not every guild had due to rarity. If your tradeskill system is designed in a way that allowed lower tier guilds to craft items that the top end guild wanted, this creates an interdependency.

If your game has a high risk of corpse loss like Fearplane as a recurring feature then this may result in guilds making arrangements with other guilds for CRs. Generally speaking if the content is very unforgiving then players will tend to band together.

You could even have desirable items that only drop for each specific guild, that are relatively common drops for these guilds such that they will end up rotting if they don’t give them away. The guilds would then get to decide which other guilds get to loot these items, by inviting non-guild players in to pick them up or simply by trade.

You could give guilds an option to spawn bosses for other guilds somehow. If the guild could decide when and where the spawn of a raid boss would be, then they could give this information to another guild or guilds and deny it to a hated guild, or they could get the option of not spawning it at all. In this system the guild that spawned it would not be allowed to kill it themselves. It could be as simple as a token that dropped in a raid zone someplace that when given to a specific NPC spawned a boss somewhere in the world where a friendly guild was already be setup because the two guilds made arrangements. This boss would banish anybody from the guild that looted the token. There are other ways to do this.

Anyway you get the idea. This topic could be its own 10,000 word essay.

There are other design choices that would mitigate the problem without creating guild interdependency. The overall solution will require a combination of design choices. One design trick isn’t going to work by itself. If the solution to this problem were simple then it would have been figured out by now and done in games before. Here are some more ideas:

1. Synchronize respawn timers. If every dragon in Kunark respawned at the same time, then it would force guilds to pick one and allow other guilds a shot at something without one guild monopolizing them all.

2. Announce respawns. If there was a world emote that announced the respawns, then keeping a guy sitting on them would be pointless.

3. As mentioned above, segregate raid content into factions and force guilds to stick to one faction. Velious on steroids.

4. Spread end-game content out more. Why is all end-game content in a single zone like Veeshan’s Peak or Vex Thal? Once again Velious did this better. Even Vanilla had two planes zones.

5. Prevent quick travel in your game world, then force guilds to “live” in a specific area of their choosing. They could then only monopolize the content in that area and would have to move periodically. Lesser guilds would simply move into the areas the top guild was not in. This would require a rather large world however. (large world sizes are preferable in general)

6. Increasing spawn rates could be done as well however this results in more items entering the world which devalues them. (mudflation) I think spawn rates could scale with server population in a way such that the number of items entering the world is roughly equal per capita across the servers. I.e. a server with twice the population would get double the spawn rate. But if one high population server just ended up with a single uber guild then this becomes undesirable.

During Luclin I noticed an interesting dynamic on my server. My guild was the top guild, so we moved on to Luclin raid content and started ignoring Velious content, including North Temple of Veeshan, after about 3 months. The guilds behind us in the pecking order however ended up ignoring the harder Luclin content and focused on NToV long into the Luclin era and we ended up not having much competition at all after this. The lesson of this is that designers should intentionally not obsolete the previous expansion’s raid content and allow it to be the casual player’s content in the next expansion. In this way the raid content will be experienced by a larger portion of the playerbase, it spreads out the server population into more zones, and there will be less fighting over spawns.

As for casual, single group content, my suggestion to prevent FBSS camps would be to not make it the only source of haste in the game outside of raids.

Kill Grinds Vs. Quest Grinds

World of Warcraft started the quest grind trend. Quest grinds put the player on rails and more or less force players along a predetermined path through the game world. Unrepeatable quests grant much of the player’s experience progress. This is very much a matter of taste but I am going to make the argument for kill grinds, which I prefer.

Killing NPCs for exp gives players much more freedom as to where to go in the game, obviously. If you don’t like one spot or it’s too crowded you can simply pack up and move. This could even determine how you killed your mobs. In EQ outdoor zone strategies to kill things were very different than indoor zones. (kiting, reverse kiting, animal spells etc) Games with quest grinds require a lot more development time which could be spent on other things, like making more zones to explore. Generally I would say that half the server population leveling in the same zone at the same level range is a bad thing, and it’s desirable to want the population spread apart. Quest grind worlds need to be small because of the amount of work needed to make all those quests.

Quest grinds require a lot of running around. WoW is a lot of running to and from a quest hub. EQ is criticized for the amount of camping in the game but WoW has the opposite problem: you are essentially awarded exp for running. If I’m running around a lot I want it to be because I’m exploring the world or moving items around to trade, not back and forth from the quest hub over and over. WoW has become a metagame of pre-planning a quest completion route to the point entire massive addons are made just for this purpose which is silly.

A huge problem with quest grinds is that it becomes very likely that you will become “out of sync” with your friends and end up being unable to group together because your friend had done quests ahead of you or you ahead of him. This is a very bad thing in a game intended to be highly social.

Generally speaking I don’t think quests should award exp at all. Why does collecting animal poop on the ground (Furor liked to make quests like that, which is admittedly amusing) make me swing my axe better? It’s illogical even for a video game. Instead killing for exp could be made less monotonous by granting diminishing returns from killing the same types of mobs, encouraging the players to travel, explore, and grind in exotic locations. In EQ camping loot was a secondary goal to earning experience. Design games such that players are working toward another goal while they earn exp and it won’t feel like a grind.

WoW’s quest system feels more artificial and less immersive than EverQuest’s even though it’s more technologically sophisticated. Nobody read the quest text and instead they installed addons which sped up the quest dialog. It even shows the rewards before you complete them. There’s zero mystery about it.

EverQuest Zone Aesthetics

EverQuest’s original zones were made with a tool so primitive by today’s standards that it didn’t even have a copy paste feature. Later on they updated their world building tool for Luclin and these newer outdoor zones had a rather uniform look to them. This later motif was: hilly forests with numerous oddly placed large boulders, slightly crooked trees, a generic reused grass texture and overly large stone houses without roofs.

EQ’s original zones had much more character to them. I could put up a screenshot of Dagnor’s Cauldron and every EQ player would recognize it even if it didn’t show any water or NPCs. Even the forest zones in vanilla EQ had some distinctiveness to them despite the trees being super low poly. The Faydarks are immediately recognizable. Toxxulia has a tan colored fog and is significantly brighter than Nektulos or Kithicor. Nektulos has a more vibrant grass texture than the others. I could tell these zones apart in unlabeled screenshots.

Guk – both Upper and Lower – are beloved dungeons. These had the character they have because the zone editing tool used was so primitive. This forced or encouraged Geoffrey Zatkin to make a very irregular zone with highly distinctive rooms and hallways. Sometimes fancy tools just get in the way of greatness.

In contrast the outdoor zones since Luclin look rather similar. Newer more sophisticated tools enable faster prototyping but less interesting designs. In modern games the focus is put on realism but this can be a bad thing sometimes. Pre-Luclin zones had more charm and character to them– even though the tech used to make them was primitive, they had fewer polygons and they used stock textures. It’s ironic. Newer games should put a lot more thought into making their regions look distinct.

Lost Dungeons of Norrath was the worst example of how to make zones. They copy and pasted highly linear rooms and hallways all over the place. It was like the antithesis of original EQ in more ways than one. LDoN is why you need more than six months to make expansions.

A Game of Band-Aids That Worked

A lot of EverQuest’s mechanics were not anticipated in the design doc stage and were later additions after designers observed player behavior which differed from expectations, or as a result of a problem that was not anticipated that needed a solution. EQ was an immersive fantasy world that needed a lot of band-aid fixes. EQ proves that you don’t need to overthink problems and still have a functional game by deploying very simplistic solutions.

Examples:

Problem: twinking made low level players far too powerful and ruined the early game. Solution: just make high level items not grant the player full power at lower levels under the hood. Simple and effective. No need to add a minimum level on items. (although they later did that) It allowed the player to twink but kept it at a sane level.

Problem: warriors had trouble keeping the attention of tanked mobs. Solution: make weapons with hate procs.

Problem: kiting mobs utterly trivializes them and that is entirely unacceptable for raid content. Solution: make mobs summon the player.

Problem: NPC summoning made pulling mobs very difficult and dangerous. Solution: NPCs can’t summon until they are under 96% health.

Problem: solo wizards were making trains then casting AoE spells on them and killing a silly number of mobs all at once. Solution: limit targeted AoE spells to four targets.

Problem: clerics casting Complete Heal in a coordinated staggered cast rotation trivialized healing tanks on raid bosses. Solution: make raid bosses hit a lot harder and assume this tactic will be used on everything.

These are all imperfect solutions and likely would not be designed into a game from the beginning, but they all work and actually add depth to the game. Instead of the warrior just hitting a button to generate hate, he must instead look for a weapon with a good proc, gear for dexterity and perhaps even tell the raid when the first proc or two hits.