Monthly Archives: October 2014

Balancing randomization in MMORPG combat

Roleplaying games are pretty much built around the use of random number generators (RNG). The history of this derives from the tabletop world of course, but there’s a reason randomization has stayed a part of the genre throughout it’s years of moving to the digital world of computers: player skill isn’t what is considered important, but that of the character – hence the term roleplaying game. The player does not play themselves but a character with a distinct set of skills and abilities.

Another reason for the existence of randomization is that it’s an easy and cheap way to make gameplay more varied. In the tabletop world this point isn’t as crucial because no campaign will ever be exactly the same, even if run by a the same game master, but fighting for example the same monsters in a computer RPG would get boring much more quickly without RNG as every fight against a particular monster type would be exactly the same as the one before. Here we step on the main point of randomization: it spices up the gameplay.

While previously pretty much an unquestioned part of the genre, as competition has become a regular aspect of the multiplayer CRPG, especially the massively multiplayer online variety of it, random number generators have come to be a subject of debate. Especially in games that involve player versus player combat the subject is seen as one of great importance. It of course also affects another form of competition, the competition between raiding guilds to down encounters faster than others, but this is  to a lesser extent and hence I will not go into that in this post.

Its easy enough to understand why randomness is regarded as such an important subject in PVP-focused games – after all, continuously losing to another player due to bad rolls can with fair reason be called lame. It makes one question whether a game is suitable as a competitive platform to begin with.

With that in mind, it is surprising to me that, from my subjective experience at least, in games where losing in PVP can present a far greater penalty, RNG isn’t discussed nearly as often as in some games where losing to another player presents little to no penalty at all. I am talking about two distinct game types here: the open world, free-for-all, full loot games such as EVE, Darkfall or the bygone Shadowbane, where death can mean the loss of your entire property at worst, and games where just about all PVP is instanced and rather sterile, and where defeats represent nothing but a mere loss of time and rating – for example the Arena minigame in World of Warcraft.

At first glance it seems weird RNG is discussed as little as it is in open world PVP games in comparison to ones where competition happens in a far safer environment in terms of what one can lose. But what it maybe comes down to are two things.

First, because open world PVP is already not artificially limited by a set of rules that would, for example, limit the amount of participating players, there already exists such a great level of randomness that the addition of randomized damage ranges, hit tables and whatnot does not seem like such a big deal.

The second reason is social: the game types appeal to different kinds of people. From my personal experience, players in open world PVP games tend to have backgrounds in the RPG genre, where as people into instanced PVP often have at least some background in esport-type games – FPS, MOBAs and the like – and often little to no background in roleplaying.

While in instanced PVP randomness is especially frowned upon, just about no game that offers this feature has completely gotten rid of it. There’s has to be a reason for this. What that reason probably is is that developers realize having at least some randomness to combat is actually beneficial even in a competitive game. The question to ask is: what is a suitable amount of randomness to spice up the game without making it too random.

Once again it comes down to what a game’s playerbase expects from the game – what sort of player the game appeals to. The competitive arena player who watches esports streams on Twitch probably despises randomness and wants things to be decided based on player skill. That being said, while some would claim otherwise, having a level of randomness doesn’t completely eliminate player skill, not even close, and despite numbers sometimes being more in favor of another team, in the long run randomness always balances out. A reasonable amount of randomness in a game of this sort can make it slightly more interesting to watch and play without completely deciding matches – critical hits, damage ranges and the like. Pseudo-random systems are not bad either, and many MOBA games utilize them despite these games being very balanced.

Contrast to arena players and the like, players of open world PVP games tend to be more used to traditional RPG mechanics and expect them.This alone makes them more prone to be acceptable of randomization, but in addition to this, world PVP is already almost always unfair in one way or another. Of course a randomly generated streak of missed strikes causing a death and loss of property is annoying, but just about equally annoying is getting killed by a group of two players who happened to stumble upon you as you were adventuring alone. In such an unbalanced, unsterile environment RNG does not get to stand out in a negative way very often due to the huge amount of other factors that affect who actually wins a fight. This, I would conclude, is the reason RNG is so little discussed in open world PVP games.

Now, if we did want to completely remove randomization from combat, one could ask: what harm would that do? The problem is that as for PVE, random number generation is almost necessary, because encounters are normally tightly scripted. Without randomization, PVE would quickly become extremely dull, because the exact same strategy without the slightest bit of variation would work every single time against the same encounter.

Few games have a completely separate combat system for PVE and PVP, so the two must overlap – if randomization is part of PVE, it must also be part of PVP. That has to be accepted by the developer. But the amount of randomization should probably depend on the type of game they are making, and the type of playerbase they are aiming for – different types of players expect different things.


Carebears playing PVP games

Inspired by recent small-scale discussion in certain blogs on the subject of PVE servers in games centered and balanced around the idea of free-for-all PVP and player looting, I came to think of trying to put my finger on what makes a game of this type appeal to players who don’t do PVP combat at all – the carebears. The fact is, just about every game of the described type has those players who never fight it out with others – what is it that draws them to these games?

The economy game

Its well known there is a vocal group of people who are very strongly against any type of unconventional PVP in MMORPGs. I believe these people to be in the minority. Much like the PVP advocates who feel every game should be all about free-for-all PVP combat, I view this group as a niche. The majority of genre players probably fall somewhere in between the two extremes, as is usual with divisive features.

Just because a player does not want to engage in PVP does not mean they dislike it. EVE is a perfect example, as has been pointed out time and time again – warfare drives the economy, but participating in the economy in a very meaningful way does not require direct participation in combat against other players.

A meaningful, well balanced economy is interesting, and to a lot of people, reason enough to play an MMO. In the case of games where killing other players is possible and looting or partially destroying their stuff equally so, PVP helps to both drive the market forward and balance it, to the benefit of also the merchant who might not have an interest in direct conflict at all.

Ignoring the bad for the good

As gamers know all too well, the perfect game does not exist. Even our favorite games have their flaws which we look past because the rest of the game draws us in. Especially in the current market this is more true than ever, for the genre is not new anymore and people more or less know exactly what they want. Combine that with the fact in recent years the supply of MMORPGs has mainly consisted exclusively of the post-WoW themepark -subtype and we’re left with a situation where if one wants to play a new game, it is likely they’re going to have to look past a lot of features they personally dislike to get a piece of the good stuff.

ArcheAge is a very good recent example of this: a lot of FFA PVP-minded players are known to be playing it despite the game having many features they dislike and lacking others they would find suitable, simply because there is not much else for them to play out there in the current market.

Just like the free-for-all PVP crowd playing ArcheAge have to settle for “good enough”, so must many high sec EVE players. To a different extent of course, but in the end, FFA PVP is just a feature amongst others – just because it exists in a game doesn’t mean its the only thing to do.

Players will play a game the majority of which they like, even if they dislike some of it’s features. However, if some feature is especially offputting to a certain group of players, the features that group actually considers interesting should be of particularly high quality. Take Darkfall Online for example. A common argument to hear on why the game always had such a low population is that it features open world PVP and full looting. I would argue that this is only part of the problem – the real problem is that the game’s quality does not meet the expectations of players and the game lacks in just about all of it’s content. There just isn’t much else to do in the game but to endlessly murder others.

Just having a free-for-all PVP system doesn’t have to turn a game into a murder simulator. I feel a game with the feature can well be designed in such a way it is still interesting to the carebear. Its also important to note  that a game of this sort could still well include quality PVE content. And lastly, balancing the economy of a sandbox title without PVP seems like a pretty tough nut to crack to say the least.

UI design: too much information

User interfaces in MMORPGs tend to be quite extensive to put it mildly. From various exact numbers like threat, damage-per-second, health and other resources to real time boss strategies and timers, you’re likely to find an option to display it on your interface in many games.

A justified question is whether this is a good thing or not. The question  was sparked by a YouTube video by Corpsealot (whose videos I very much recommend to anybody interested in MMORPG design philosophy by the way), titled Players know too much. In the video it is argued that players have too much information, and that the right way to design a truely immersive game experience would be to not convey more information to the player than can be considered to be available to the player’s character.

I don’t necessarily agree with the view all games should be designed around this principle, but then again in the video is presented a single person’s opinion. That being said, there’s a point to the commentary there. Having more information of, for example, the health state of an opponent being conveyed by visible phenomena within the game, for example by bleeding and slowed down movement rather than by numbers on the UI, might well be more immersive than the typical RPG approach that displays a lot of things as raw numerical values.

Is a minimalistic UI a prerequisite for an immersive experience then? Of course not, we all know that. Whether it was while playing a tabletop RPG or a tab-targeting based CRPG, we’ve all (I hope) at some point felt very immersed in a game world. And honestly, most people probably aren’t even looking for complete three-dimensional world simulation. The majority of gamers are perfectly happy with character health states, feelings and other things being presented to us as numbers, icons, bars and text. But I think there’s room for discussion whether we need as much information as we are currently being handed.

Take for example exact unit health values. We already have a red or green bar roughly representing the health percentage of boss X. Do we really need to know boss X has a maximum of exactly 120,063 hitpoints, of which 99853 are left? I would argue that discovering the rough amount of health a unit has by trial and error would make for intriguing gameplay in many games of the genre.

For another example, take unit levels. How do we know this cat-like beast we’re engaging in combat is exactly at a combat level of 37? And why does a unit with the exact same model and texture have a combat level or 8 instead? Or 56? Coupled with the knowledge of the exact maximum health of the unit amongst other possible bits of information conveyed to us by the UI, we can pretty much tell without ever engaging the opponent whether we’d stand a chance or not. Why does this information have to be so exact?

I’m not saying design should go back to the EQ days when you would type /con to consider every single unit you ran into. But I feel that a game needn’t tell the player exactly whether or not an opponent is suitable for them to take on or not. Even color codes for names based on relative level would in my opinion be an improvement. Not only is it ridiculous that a knight rushes headlong into battle against a massive troll whose combat level is 5, yet hides when they meet an exactly similar looking troll who’s combat level is 50 instead, but if exact level display was disabled, in PVP it would be more difficult to decide whether or not you might want to attack that lonely rider on the road as you couldn’t tell for sure if they were just a regular traveler or a master swordsman in disguise.

The overabundance of information isn’t always nonsensical. It suits EVE Online for example because, well, spaceships – complicated machines with all kinds of sensors and buttons and whatnot. But as you’re trying to become immersed in a high fantasy world, it’s a little ridiculous you have 7 boss ability timers on your screen as with a battleaxe you smash away at the Dragon Queen who you know has exactly 74302/120000 mana left, meaning that after she’s spent 24302 more mana another mechanic, shouted at you by a big block of red text in the middle of your screen, will enter the fight.

Paying not to grind is paying to win

Often when discussing online game payment models, one runs into a situation where one side considers a free-to-play title to be of the pay-to-win variety where as the other side disagrees. So what exactly can be considered paying to win? While people have differing levels of tolerance to P2W features, I don’t think the answer to the question is down to opinion alone. In fact, I feel denying certain features being pay-to-win is just being intentionally misleading. This is the issue I want to tackle today.

Winning in a virtual world

An argument sometimes brought up by either F2P advocates or fans of specific games is that there are no winners in MMORPGs – hence there can be no paying to win. And they’ve got a point there, to some extent: there are winners of situations within a game of this type, for example a player may win in a duel against another, but there is no definitive point where you’ve won the game.

Because of the unclear definition of winning, I feel it is appropriate to steal a term from Syncaine: pay-for-power (P4P). Essentially, all features we’re used to calling P2W in MMORPGs can be called P4P instead. From now on, I will try to use this term.

Paying not to grind

The most glaring P4P features are of course the type where real money buys the player levels, skills, equipment or another form of power not available through other means. These types of cash shop options are, surprisingly(?), fairly rare in today’s games, at least here in the West. Few people would argue against this sort of thing being P4P.

Opinions get more divided when the power being sold is also available through other means. Sometimes people will argue that if a cash shop item can be attained by just playing the game as well, it cannot be considered P4P. I would disagree here.

Character power in online RPGs tends to be a reflection of how much time the player has spent on a character. In fact, time tends to be he greatest factor when it comes to it. When one uses real money to buy powerful items or levels, even if those things can also be attained through purely in-game means, one is paying money to not have to spend as much effort for the item in-game.

A situation of this kind would be called by some a case of paying not to grind, which to them isn’t the same as paying for power. You also often hear these people saying grinding is something anybody can do, and because of this paying not to grind features should not be considered P4P.

I completely disagree. In a skillcheck situation, you either have the skill to get to an objective or you don’t (example: beating a difficult dungeon to get a powerful item). Similarly, you either have the time and determination to grind for an objective or you don’t. There is no difference. If you can skip either and claim the rewards by paying money, you are paying for power.

So what isn’t P4P?

There are some situations where it is difficult to say if a feature is P4P or not. Take for example RuneScape’s model. I have heard the model has lately changed dramatically though, so I’ll explain what I mean:

  • – As a F2P player, you are allowed a certain set of skills, all of which you can train to their maximum potential. You are also allowed access to servers that have most of the world in an accessible state.
  • – Subscribers can play on subscriber-only servers where there is more landmass to explore and more skills to train. Skills available in F2P are also given more variation through subscriber-only items. Subscriber-only items cannot be transferred to free-to-play realms.

The difficulty of defining this model as either P4P or non-P4P comes from the fact being a subscriber does sort of grant more power by giving the player access to more efficient leveling areas as well as a larger variety skills and more ways of making money.

Personally I would not consider this P4P. The F2P portion I would instead consider an extended free trial, and buying a subscription only buys you content. That content might or might not be more efficient for gaining character power, but nevertheless the player is not directly buying character power – they’re still going to have to work for it. The fact most subscriber-world-only items don’t work on free servers also helps to make the system less P4P.

World of Warcraft’s expansions are a pretty similar case to the above. One could argue they are P4P because the potential power level of characters rises as they can train to a higher level, but in the end what the expansion really buys is content, not direct increases in character power. Besides, practically everybody ends up buying the expansion anyway, and the game is sort of designed around this idea. As for the game’s instant level 90 thing though, that’s a completely different case: in it, the player is literally buying character power in the form of levels, and I don’t think there’s arguing this isn’t clearly P4P.

To give one more example, this time for a model I would consider P4P, I want to mention Eldevin, a recent browser-based MMORPG. It was looking to capitalize on RuneScape players by it’s design, but the payment model, F2P with optional subscription, was inherently different: instead of buying new areas to explore or new skills to advance, subscription increased experience gains in existing skills alongside with granting access to free potions the subscriber could use in combat. Unlike the above two examples, I would argue the system is in fact a form of paying-not-to-grind, and hence should be considered P4P.

The point

The point of all the above is not to bash F2P or or even P4P, nor to take a stance on which payment model is the best. The only thing I want to do here is to make the reader see the hypocrisy in claiming certain features offered by some games should not be considered P4P when that is clearly what they are.

A waste of resources: WildStar’s non-raid content

Its no secret: WildStar has been doing pretty poorly subscriber-wise since its release. Interest seems to be at such low levels its almost embarrassing to even write a post about the whole game. There’s multiple reasons as to why things are the way they are, but I want to address one specific reason now: the game’s target audience.

WildStar’s big selling point was, as we’ve gone through before, difficult raiding content. And the game delivered on that point, it really did. Why, then, is the game doing so poorly?

WildStar was made to cater to the modern hardcore raider. People who are used to grinding minigame after minigame, doing dungeon speedruns, playing really hardcore when new content is released, and so on. That’s all well and good, but this niche isn’t spectacularly big. Besides, the majority of this niche is already content with World of Warcraft.

That being said, WildStar’s endgame content does have it’s audience. But how about all the other content it has? Questing, housing, all that? Needless to say, the incredibly on-the-rails questing experience took a considerable amount of time and resources to develop. So did the whole level system and the content required for it, from low level dungeons, armour sets and whatnot to all the zones that have no other purpose but to act as leveling grounds, after which they are forgotten.

Despite all those resources spent on the non-raid content of WildStar, it is all very mediocre, probably because it is modeled after modern World of Warcraft. I doubt many people would deny that. Questing in the game for example is incredibly dull, and I cannot think of a reason to appreciate it over any other game’s similar system.

Developing all that mediocre filler content must have been expensive, especially because the game wasn’t even advertised to appeal to the people who typically spend the most time in that content – the casuals. No, the game was sold as a hardcore raiding game. Which begs the question: why the hell was all that development time really spent on the non-raid content?

I realize Carbine was kind of looking to make a better, more modern version of World of Warcraft. But by now we should all know nobody will beat WoW at it’s own game. This is something we’ve been saying for years, yet its what Carbine tried. Now, differentiating yourself as a hardcore raider friendly game might be a good way to specify your target audience, but if you do so and still spend  the resources on a ton of mediocre non-raiding related content, I think you are wasting resources. You should be focusing on your niche and not trying to broaden it needlessly.

Had WildStar not had a full leveling experience but a short tutorial instead, and should the only open areas have been cities that would’ve acted as lobby areas, the game, I think, would appeal to the very same playerbase it does to right now. Hell, maybe they would’ve gained a few people because several hardcore raiders were probably put off by the mind-numbingly boring leveling experience. Now that might have actually been financially viable, seeing as a lot of money would’ve been saved by focusing  development on the game’s strong points: minigames like raiding and hardmode dungeons.

Combat: complete balance isn’t necessary

Following a discussion on regarding character hitboxes in games with aim based combat, I came to think a little more about balance. The following questions popped into my mind: what exactly is the function of balance and why is it needed? And what would players consider balanced enough?

To start with the first question: combat balancing is done, of course, in the name of fairness, but also to encourage character diversity. It’s job is to ensure that players feel like they can participate in combat in a meaningful way no matter their choice of race, gender or archetype.

I mentioned in my last post regarding action combat that RPGs by their nature are not competitive games. This isn’t entirely true for the massively online subgenre of them, because there are abstract and sometimes even direct forms of competition in these games. Whether it’s fighting over getting a raid slot or a piece of gear, defeating an opponent in PVP, or maybe beating another person in damage-per-second numbers, its all a form of competition between players.

Sometimes homogenization is necessary…

Despite the fact competition exists, the acronym MMORPG still has those three letters at the end of it standing for roleplaying game. And because of this I would still argue that the goal of MMORPGs is never to solely act as competitive platforms, but rather to provide an online medium for players to take on various character roles in a virtual world.

This weird mixture of competitive and anti-competitive features makes balance in the genre a delicate topic. On the one hand a developer wants to make archetypes, races and whatnot distinct from each other, but on the other the developer also wants them to perform roughly equally or at least stand a chance against each other to not force any choices down the players’ throats.

The easiest way to achieve balance between classes is to make them function very similarly to each other but change the visuals. The problem here is that doing so makes gameplay feel less varied and bland. In a tabletop game, a wizard is a wizard. With a massive fireball, if available to them during a specific encounter, they can be the most destructive thing you’ve ever seen – yet in a fight where they don’t have that fireball available to them, it might be the wizard can suddenly be considered a nearly useless character. For comparison, an archer can provide a good amount of well sustained damage all the time, but is incapable of delivering similar amounts of massive destruction to the fireball-weaving wizard in their prime.

Now compare the tabletop versions of the wizard and archer to similar archetypes in just about any modern MMORPG. Both the wizard and the archer are ranged classes, they both use a limited resource such as mana or stamina or focus, and their abilities, while having different visuals, do roughly the same things: grant sustained damage, slow targets, and so on.  Frankly, they are differently skinned versions of each other – there may be some small differentiating features such as reagent usage or amount of AoE damage in addition to the visuals being different, but nevertheless the archetypes remain very similar in fuction.

In the specific case of the wizard versus archer we can justify the homogenization with the fact that playing a traditional tabletop wizard in a 3D, massively multiplayer online setting could not be made to work very easily. Their  downtime would make them boring to play most of the time, and when they would be able to use their spells, other players would probably deem them overpowered. So while unfortunate, it makes sense to make wizards more similar to archers in the case of MMORPGs.

…but sometimes, homogenization just ruins the game

While the former is a case of balancing where homogenization is almost inevitable, often there are other ways to work around balancing a feature.

Take character hitboxes in aim based games from which this discussion started. In a typical fantasy setting you have races of various sizes: humans, dwarves, ogres, elves… This creates a problem: smaller races are inevitably going to be more difficult to aim at. How would you solve the problem?

If you were a company called Aventurine working on a title called Darkfall: Unholy Wars, you would take the easiest and cheapest route: take out all the small races. See, in the original Darkfall game the competitive PVP folk used to be upset about certain races having smaller hitboxes than others. Some of them felt it was mandatory to play those races to remain competitive.

Aventurine decided it was a good idea to change this for their re-branding of the game so that all races would use the same 3D model but a different texture. This required some lore changes, such as having dwarves go practically extinct by crossbreeding with humans. The end result was that all the game’s races ended up looking like humans. For a game classifying itself as an MMORPG, I would say this is horrible design.

Aventurine’s way of solving the issue is by no means the only way to achieve balance. They did it because it was easy and cheap, although to be fair it probably didn’t affect their playerbase much as the game had already become a sort of a giant battle arena instead of an Ultima Online-esque sandbox by this point – a topic for another time.

There’s a couple of other ways to solve the issue of hitboxes, actually. Imagine yourself being a game developer. If your ogres have giant hitboxes making them easy to hit and everybody rolls a gnome with the smallest hitbox as a result, what can you do? One of the following, for example:

1. Treat hitboxes as just another stat. If your ogres are easy to hit, give them other statistical bonuses such as higher strength or longer reach.

2. Restrict classes, archetypes or skills based on race. This requires that all the classes are useful and wanted.

3. Normalize hitboxes but not the visible character models.

Out of the three, the third one is in my opinion on about the same level as normalizing character models – a lazy and bad solution. It isn’t as big a negative for the roleplaying aspects of a game, but it is going to create confusion when an arrow passes a gnome a meter over their head and they still take damage. Additionally, if collision detection exists as well, this becomes even more confusing.

My favourite solution would be the combination of options one and two. Including reach and hitbox size as statistics and adding some race restrictions to skills or classes gives the developer a good amount of parameters to play around with when it comes to balancing a character’s other traits in relation to the advantages or disadvantages gained through various hitbox sizes. This model would, as far as I can see, include all the positives of the RPG part of the genre, yet would also suit the needs of the more competition-driven player.

Balance doesn’t necessarily have to be perfect

There are difficult and easy, good and bad ways to balance an MMORPG when it comes to combat. Despite this, it is important to keep in mind one fact: this is a subgenre of the RPG, and as such balance needn’t be perfect. There’s no reason to spend too many resources in finding the perfect balance in a genre that wasn’t intended to be competitive in the first place – as long as classes and races are varied and they have at least some job to do, there’s a good chance most of your players are going to be pretty happy.