#mmog development

The reason that MMOGs continue to be made is because MMOGs are often actually quite profitable once they establish themselves. There is no real "romanticism" in building MMOGs - they are, by far, too expensive to build for the sake of being a pet project. MMOGs are a good vehicle for maintaining engagement. The social aspect of playing with many of the same friends over weeks and months provides a powerful incentive for players to continue to engage with the game. As I've said before, engagement is the most valuable metric that any game can have because it directly correlates to game revenue more than any other metric we have.

Let's examine your premises and see which ones don't hold up.

They require a massive constant influx of players to remain profitable

While this is true, there are adjustments that can be made as time passes. Profitability is a factor of both cost and revenue. As the player base naturally dwindles over time, so does the size of the team and the scope of planned future content. Thus, the cost of maintaining the game continues to scale downward with the decline of the player base, maintaining profitability. Eventually any MMOG will be placed in “maintenance mode” where no new content will be developed, but even then it can remain profitable for years as long as enough players are still willing to pay for the ongoing maintenance costs (as is the case with Ultima Online, which is approaching its 25th anniversary this year).

The graphics quickly become outdated

Cutting-edge graphics are only a selling point at launch. Many MMOGs today (WOW, LOTRO, FF14, SWTOR) are still going strong many years later despite not being the prettiest belles at the ball. Once the game gets its hooks into the player, they want to keep playing even if it isn't the prettiest game.

Most MMOs, barring a few exceptions, shut down relatively quickly after launch

The first year is where the majority of MMOGs die, primarily because they cannot grab onto a critical mass of players. The first year is also where the majority of games in general die for the exact same reason. However, if the game manages to survive its first year, the chances of a long lifespan are actually pretty good. For example, Defiance was a MMOG that was licensed and based on a Sci Fi channel cable television show that lasted for two seasons. The game launched in 2013 and didn't shut down until 2021, six years after the show had gone off the air. There are hundreds of smaller F2P MMOGs still running today because they managed to secure that critical mass of players.

I think most people don't realize that MMOGs are still a thriving genre, but have also evolved with the times. Most of them don't try the AAA big budget approach because such vehicles are somewhat obsolete. The only big AAA MMOG in the past several years I can think of was Amazon's New World Online, which had (and continues to have) its own share of problems. AAA MMOGs have evolved into more streamlined versions (a la Destiny, the Division, etc.), losing expensive systems like item trading, player housing, and so on, while retaining the social systems and retention/engagement elements. Smaller MMOGs have reduced scope and visual fidelity, instead focusing on content creation tools to save on development and maintenance costs. These gradual changes over time have changed the way MMOGs are built and maintained today. Making new MMOGs is by no means untenable.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

A scenario very similar to what you asked about actually happened. [Project Gorgon] is an indie MMOG developed by a husband-wife team, where both developers were extremely experienced MMO developers. Both have a lot of experience on games such as Asheron's Call 1 and 2, Star Trek Online, Everquest 2, etc. and decided to build their own small scope MMOG. Obviously, a lot of specific stars needed to align for this to even have a prayer of success.

First off, the dev absolutely must be a very skilled programmer who understands client server architecture and scaling. Without this core component, it doesn't matter what you can do - there is no off-the-shelf game engine (yet) that can scale to the number of users needed for a MMOG. This is one of the biggest problems with running MMOGs in general - a server that can handle ten players at once may not be able to handle a hundred. A server that handles a hundred players concurrently may not be able to handle a thousand. Scaling code to user demand can be incredibly difficult and expensive, which is why good server programmers are highly sought after. The game client would be built in an off-the-shelf game engine like Unity or Unreal to handle the visuals and front-facing elements.

Beyond this, the dev must also have sufficient design chops to create core gameplay that is both fun for players and sustainable to develop for. This essentially limits the types of content that can be developed - there are only so many hours in the day, so the ratio of developer effort to gameplay hours must be very high. This tends to exclude things like intricate story content and specific scripted encounters that will get consumed super quickly, and instead leans on sandbox gameplay and combinations of different pieces of content and systems in order to create the fun for the player. This way, any new content addition can potentially be used in combination with old content multiplicatively, creating a lot more possibilities for interaction. Leveraging and reusing systems and content for extended engagement is imperative or players will run out of things to do and drop the game.

There is basically zero bandwidth for visual/audio asset creation on top of creating the game-specific content and writing all of the game code, so all game assets will either come from the asset store or is outsourced from contractors. As such, the look of the game is almost certainly going to be patchwork at best with very little visual theming unless they spend a lot. UI is probably the best place to invest for custom work since it probably needs the strongest visual identity.

That’s basically what you need to start a small MMOG - expertise in getting the technology requirements working, expertise in game systems and a content creation pipeline that is extensible and efficient in terms of developer time spent compared to user engagement, and enough money to buy and commission the assets you need. If you can assemble these core requirements (and manage to find an audience), you could build a sustainable indie MMOG.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Today we'll act as system designers and break down your question into a set of smaller questions that are easier to answer. First, let me establish this as the premise - a MMOG service is like an all-you-can-eat buffet that's open 24 hours a day, 365 days a year (minus server maintenance times). The expectation from players is that they will be able to log on to the game at any time of day, and play without having to wait for too long.

We can break down the sort of gameplay engagement from players into two buckets for this - solo gameplay and gameplay that requires other players. For solo gameplay, we're good - the game service automates all of the NPC spawning and loot table rolling for us, so all of the solo content should be fine on its own. For content dependent on other players, we need to consider the ways in which players can interact with each other within the scope of the MMOG's gameplay systems. These activities can include things like:

Player to Player Trade (direct trade, auction house, etc.)

PVP Group content (e.g. Battlegrounds)

PVE Group content (e.g. Dungeons and Raids)

Zone/Faction PVP participation

PVE World Events

We can break further break these down into two smaller buckets - real time community needs and asynchronous community needs. 

If a player logs in during their preferred play time and can't find other players to run a dungeon with or finds the PVP zone to be empty in real time, they aren't going to have fun. This is the "real time community needs" bucket - a player needs other players right now to play with. To keep these systems from feeling frustrating, we need enough players engaging with these gameplay systems such that the wait time is not too high (i.e. usually < 5 minutes or so). We can usually mitigate the frustration by allowing players to engage with other game content while waiting for their queue to pop, populating empty slots with bots for queues that take too long, and so on. 

Similarly, if a player logs in and the auction house is empty or the faction-wide quest hasn't advanced enough since last week, the player isn't having fun. This is the "asynchronous community needs" bucket - a player needs other players to have engaged with the game within the past N hours/days/etc. and not immediately. For these systems, we have a much longer time frame where we just need to make sure enough players over a period of hours/days/weeks/etc. will engage with the system to keep players from feeling frustrated. This can also be supplemented by generating content for players (e.g. having a bot auto-sell items on the auction house if there aren’t enough of those items to buy).

Usually, this means we want around at minimum of around 10 x (number of players per group) players seeking real time group content at any given time in order to do the matchmaking within a reasonable time frame. This also doesn’t include further subdivisions among the player base (e.g. waiting for a tank to queue, or how level 20 players and level 100 players probably aren’t going to be seeking the same group content). For asynchronous content needs, you need enough players to engage over that time frame (usually a few hundred at minimum) to keep things stocked and working. Exact numbers will, of course, vary with factors like player turnover rate, game length, backfilling players who leave games, mitigation features like bots, etc.

PS. If this talk about gameplay time frames reminds you of [primary/secondary/tertiary gameplay loops], you earn a cookie for thinking like a game designer.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

That particular feature isn't super widespread to my knowledge. When I saw it, the system only generated common high-volume auction house items like crafting materials and consumables if there weren't enough of those items up for sale already. So, for example, if there are fewer than 10 Wombat Leather for sale on the auction house for at least 8 hours, the system kicks in and generates enough new Wombat Leather sales attached to randomly generated names to make up the difference. Only commonly traded items (that is, items from a designer-defined list, no rare stuff) would be generated this way.

The issue with creating a system like this is that it spends resources treating a symptom and not the root cause of the main problem - that there aren't many players engaging in the gameplay to keep the auction house stocked. Most of the time this is due to a low overall server population, and that tends to bring a slew of other issues with it. We can spend our dev resources building systems like this, but a better overall solution to fix this (and all of the other problems at the same time) is usually server merges. That would allow the auction house to remain full while the devs who would have built this system get to work on something else instead.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Raid participation among players really isn’t super significant. Based on the information I’ve gathered on MMOGs currently running and my own experience working on a pretty well-known MMOG, the numbers seem reasonably consistent. Around 20% of all max-level players will engage with raid content at least once. A significantly smaller number (~half that number or less) have actually completed a raid (i.e. defeated the final boss), and an even smaller number than that will actually return to do a raid more than once. For the vast majority of players, raids are something they either do not engage with at all or engage with once to see the story and then don’t bother again - even if we make it really really easy for them to engage with it (e.g. WoW’s raid finder tool, significantly lower the difficulty, etc.). 

One of the hardest earned lessons I learned from working on MMOGs is that certain elements of content just aren’t for everyone. Putting more development resources/time into building better quality, more polished, or even more plentiful content of a particular type (like raids) does not result in significantly more players engaging with that type of content, even if we make it super easy for players to engage with it. The biggest result is that making better/more raid content does not really get many more players to raid, it mostly makes the existing raiders a little happier. This general applies to other types of content as well, not just raids - pvp, crafting, mini games, achievement hunting, and so on. It might seem counter-intuitive (it certainly was to me when I started!), but that’s really how it is. This is why content updates are planned to include something for everybody. We want to tick as many checkboxes across the total player base so that everybody has at least one cool new thing to come back and check out the new update.

[Join us on Discord]

The FANTa Project is currently on hiatus while I am crunching at work too busy.

[What is the FANTa project?] [Git the FANTa Project]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

There are two key concepts that designers are thinking about when deciding how much and what loot players get. The first concept, regarding how much loot you get, is pacing. The second concept, regarding determining what loot you get, is anticipation. These are both important for sustaining the long-term health of the game. Let’s go over each of these concepts in a bit more depth.

Pacing (in this context) is the rate at which we dole out the goods. Our worst case scenario here is that players run the content once, get everything they’d ever need or want, and then stop playing because there’s nothing left to aim for. It devalues the content that we just spent weeks or months designing and building, because players will only engage with it for hours, rather than weeks. The best way to keep things interesting is to present players with an engaging and rewarding challenges that ramp up at the player’s pace (up to a reasonable maximum) over time. It is our hope that the time period affords us on the dev team enough time to build the next content update for players to enjoy. By controlling the rate at which gear gets doled out, we can tune and stagger challenges for expected player gear levels and construct an expected player progress schedule that we can tweak for how long we need the content to last. This also allows players to feel the slower build more as their characters get stronger and become able to tackle more difficult content. The sense of progress is very rewarding when you are finally able to defeat an enemy you previously weren’t able to.

Anticipation (in this context) is the excitement and interest level players feel when loot drops. Loot is a reward and needs to feel like it to players. While devs like me are sympathetic to those who end up on the wrong end of the RNG (back in my WoW days, we went for over a year of weekly clears on a particular raid boss without ever seeing a specific drop that one of our players really wanted), we also don’t want to err too far on the other side - only giving currency to players. By reducing content to what’s effectively a paycheck, it removes any anticipation the players may have for defeating the content and seeing what the rewards are. If you already know what the rewards are going to be, you’re just there for your weekly/daily/etc. paycheck. That’s not fun, it’s a chore and causes more players to churn out of the game from boredom. Generally speaking, we want a mix of the two elements. We want some form of RNG mitigation in order to prevent too many cases like mine, but we also want enough compelling stuff that is only available via RNG to keep the loot spicy and exciting. This way we get fewer extremes, while still avoiding the problem of churning players out because the game feels like another job.

Remember, the ultimate goal from all of this is the game’s long-term sustainability. Players want to keep playing the game. We want players to keep playing the game. However, we can only build so much content over time and it takes a lot longer for us to build it than it will for players to consume it. Our only recourse is to space out the content such that players don’t consume it all at once and spread it over a long enough period that we can build more content in that period. This means that we have to dole out enough loot and we have to provide compelling enough loot over that time frame to keep players interested. This means holding some of it back as well as utilizing techniques to keep it interesting and exciting.

[Join us on Discord]

The FANTa Project is currently on hiatus while I am crunching at work too busy.

[What is the FANTa project?] [Git the FANTa Project]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

I guess it depends on the extent of how anti-social you are. Players who don’t want to interact with anyone else at all are definitely in the minority, but those who want to play in a world populated by others, but not necessarily directly interacting with them might be surprised to find out that they are actually a significant percentage of the population. They aren’t always the most active or loudest group, but they do exist in extremely large numbers.

Contrary to a lot of popular belief, most MMOG players don’t spend the majority of their time in group play. It’s actually much more similar to going to a coffee shop or shopping mall scenario - people often hang out at coffee shops or the mall to be around people, but not necessarily to directly interact with them. It’s a concept called being “alone together” - where one can get the aggregate benefits of generally playing with a lot of people (such as the ability to buy, trade, and sell resources, services, or items to each other), but not necessarily needing to cultivate interpersonal relationships with each other to do so.

Content aimed at these types of players tends to be things like new collectibles, daily quests, seasonal events, or new story lines that don’t require grouping. Different games have steadily moved to embrace these types of players in different ways. SWTOR, for example, targeted these players with their Knights of the Fallen Empire expansion by switching the main storyline of the expansion essentially into an episodic single-player experience, in addition to the usual multiplayer content. Despite what the hard core players say, it was Bioware’s most successful expansion to the game up to that point.

MMOGs tend to be a lot of things to a lot of people. Most MMOG dev teams spend a lot of resources tracking their player numbers and related metrics. The devs are most definitely aware of who their players are, and roughly how many of them there are. Solo play is definitely one of the most popular ways to play MMOGs, and has been ever since World of Warcraft conquered the world in late 2004. It’s definitely one demographic that most AAA MMOGs I know of keep their eye on.

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Actually, it isn’t really “development cost vs value of content” at all. The development cost for the return is the reason a lot of MMOGs moved away from raid content, and towards other activities in general. The main reason that MMO devs haven’t moved towards scalable content is that most MMOGs are designed around each player having a specific role in groups, and groups having a certain makeup. These are expected genre conventions that players have come to expect, and scaling group size throws a bit of a wrench into it.

Normally, designers create content that’s tailored for a group of players to consume. They create something interesting for the tanks to do, something interesting for the damage dealers to do, something interesting for the healers to watch for, and it’s a fun encounter. We can add interesting mechanics that test the player’s skill and knowledge of the game. The fun for the group comes from doing their respective job while coordinating with their teammates, like a band playing a song together. The music is written with each musician doing something specific in mind.

When you have a scalable encounter, then you can’t guarantee anything anymore. There might be no tanks, one tank, or five tanks. There might be twenty damage dealers, or just one. There might not be any healers. This means you can’t design with specific roles or group makeups in mind, because there’s no guarantee that it is even possible. This means we can’t create the encounters to be played a certain way, at least not with specifics. Nothing that requires a stun or an interrupt, because the group might not have one. Nothing with a tank rotation, because we can’t guarantee there will be more than one tank. Instead, we build content that takes the number of players into account and uses common mechanics that are accessible to all players and roles - hit this button, activate that doodad, run to this area, transform into a monster with a different set of abilities, etc. It isn’t impossible to do, but it provides a very different player experience than the more carefully-crafted encounters - more gimmicky and less mastery. Think of this more like a sing-along, where you can keep adding more people to the group, but the parts aren’t written for specific people. 

Most MMOG developers started building toward group roles because World of Warcraft popularized it 13 years ago and it’s evolved into genre convention. Several MMOGs have gone their own ways since then because of low engagement numbers with raid content. FF14 chose scalable raids. SWTOR explored role-agnostic content. Other MMOGs like ESO, Guild Wars, and Wildstar stay with a more traditional paradigm because it’s consistent with how they built the rest of the game - their fans expect and want it. Look for MMOGs still in development like Star Citizen or Crowfall to continue to shake up the formula.

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr