Warcraft’s Heel

Above: Achilles (Warcraft) drags around the bodies of his competitors!

From the graveyard of MMORPG competitors who vowed to ‘slay’ World of Warcraft, it is clear that their arrows did not hit Achilles and his heel. All these new ‘features’ do not seem to make much of a difference. With all these talented game designers, why can they not beat a seven year old video game?

The answer likely is in what game designers and consumers do not normally think about. The vulnerability of World of Warcraft isn’t obvious because people keep thinking in the context of traditional gaming. The assumption is that World of Warcraft has a good operation behind it (servers, engine, etc.) and that to ‘beat’ WoW one must make more enjoyable game features. But this assumption, adopted by every WoW competitor, is wrong.

There are two great vulnerabilities to World of Warcraft that keeps the President of Blizzard, Mike Morhaime, awake at night.

The first is the bad game engine. “But I thought the game engine was very good!” This is because Blizzard has been slick in fooling you. The World of Warcraft game engine is seven years old. “But they have made changes and improvements.” True, but no improvement can totally change the base of the game engine.

It is well known that Blizzard didn’t expect WoW to be as popular as it was when it launched. The game engine wasn’t designed to be used in the year 2012. Blizzard is unable to make any sweeping changes since millions of consumers’ data is stored and integrated into this game engine.

“But Blizzard can make sweeping changes in expansions.” This is what Cataclysm was all about. Blizzard wants the consumer to think Cataclysm (i.e. Project World Breaker) was about Deathwing or the ‘story’. It actually was about Blizzard trying to update the game engine. This is why players never got the sense of ‘epic confrontation’ with Deathwing as they did with Illidan or the Lich King. The need to update the game engine took precedence over everything and the “story” and content revolved around that effect. How does Blizzard update Vanilla content of 1-60 levels and throw in things like flying (which were in Outland and Northrend)? Have a ‘world shattering event’.

People ask, “Where is the amazing new content?” The question is asked as if Blizzard is holding back or ‘gone corporate’ and is just trying to milk the poor players. No one ever considered that Blizzard is unable to do this. “But they have tons of money and a talented staff.” Yet, their hands are tied due to the rotting game engine.

When patch 4.3 came out, Blizzard’s developers praised and hyped the people who made the Deathwing battle especially the Spine (where you ride on Deathwing’s back). Many WoW players don’t seem as enthused and aren’t that excited about the rest of the raid (which takes place mostly in a tower already in Dragonblight).

Above: Near the end you see the players on Deathwing’s back. This took a huge fight against the game’s limitations.

The best way to describe WoW’s game engine is ‘fragile’. It took considerable amount of time and programming expertise to create the Deathwing encounters. While it may not seem much compared to games current on the market, Blizzard’s developers are wrestling with a game engine that is seven years old. No matter how much money, how much talent, there are limitations. WoW’s massive popularity actually added to these limits since there is so much data in the game that Blizzard cannot afford to screw up.

The issue with the content patches and the expansions is that the mission is no longer about ‘adding content’ but about ‘fighting the game engine limitations’. This is no easy task.

A WoW competitor ought to push the limits of current game engines so that a 2012 game engine would completely blow away a 2004 game engine. Ironically, WoW competitors emulate the WoW formula and have a 2012 game engine behave like a 2004 game engine creating a “WoW in better graphics”. WoW will never be able to compete against game engine differences so a WoW competitor should strive for that.

The second is that the servers don’t talk to each other. A single World of Warcraft server costs around 100 million dollars. The issue isn’t so much the hardware but the software on the server. When World of Warcraft came out in 2004, it had the servers act as different Azeroths. The servers and their operating systems were never designed to talk to one another.

When a player transfers his character to another server, he will notice the loss of certain reputation (such as the Oracles/Frenzyheart or the Steamwheedle Cartel). What is going on is that servers record this reputation information separately. When a character is moved over, it doesn’t get transferred over with it (or not the more recent information)

“But what about when I faction change?” A faction change is the equivalent of changing servers. Blizzard had to get another server just to allow faction changes. Your character actually gets transferred to this other server first, then all the changes are made (e.g. your mounts switch to the other faction’s mounts), and then the character is placed back onto the server. But because the faction change is, technically, a server change, the reputation losses can still occur.

When you understand the technical difficulties Blizzard has with the servers talking to each other, you can see why Blizzard thinks Looking for Dungeon or Looking for Raid to be ingenious. It is because they are with the issues they are facing.

There are still many things Blizzard would like for the servers to do. They want heirloom gear to be able to be sent across servers. They want account-bound pets, mounts, and achievements (achievements are the most likely and easiest of these to be implemented).

Blizzard doesn’t see progress as how much content they are delivering to the player but rather how far they can go against the limits of the game engine (Deathwing battle) or how they can get the servers to talk to each other more (Looking for Raid). Blizzard sees progress in World of Warcraft as in breaking down these limitations. In other words, considerable money has been spent in trying to make the servers talk to each other better.

However, allowing server transfers and faction changes has opened a pandora’s box. There are low populated servers and servers that are greatly faction imbalanced. I, myself, am currently on a greatly faction imbalanced server where the faction side has low population (which I blame Reddit for screwing over my server). Since MMORPGs are a game about people, the lack of people significantly effects other players’ game experience. In other words, it can take hours to get a decent PUG for raid night. This is a very, very bad situation which Blizzard sees as a high priority problem to be fixed.

The only ‘fix’ is to have the servers talk to each other better. We may be seeing hints of where Blizzard is going with allowing cross realm raids to take place with real ID. Why stop there? Why not cross-realm guilds?

But look at what Blizzard said just days ago:

Having said all that, yesterday we discussed low-population and faction-imbalanced realms with our developers. They have some pretty bold and spectacular plans for addressing this in anticipation of implementing some of the features we plan to in Mists. I just don’t have a lot of information to share with you at this stage of programming and development.

They have something planned to address this. I’m curious to know what it is. All I know is that it certainly involves the servers to talk to one another.

WoW competitors keep making the mistake of accepting WoW’s current server situation as the standard. They should, instead, focus on making the servers talk to each other very well at the beginning and create gameplay based on this as WoW cannot do it. The idea of a ‘server world’ is pretty much outdated anyway. If the servers are talking to each other correctly, there should be no such thing as a server population isolated from another server population.

“But I like the idea of a server world.” And I liked the idea of playing video games over a modem via the telephone line. But things change and evolve. Since the Internet always trends to more interconnectedness, isolated server worlds should be a thing of the past. When a player makes a new character in a MMORPG, they shouldn’t have to worry if they picked the right server.

When you hear ‘analysts’ discuss (eating their wine and crackers since they look like that is what they would eat) MMORPGs or when you hear gamers discuss, you never ever hear discussion revolve around the game engine or the server communication. The discussion from the analysts tend to be more concerned about the marketing strategy. The discussion from the gamers tend to be more concerned about the content quality.

Pretend someone was going to make a new RTS game to compete against the original Starcraft that was released nearly a decade ago. Imagine that RTS game using a near identical game engine while using the exact same Battle.Net system. The result would be instant death to the RTS game no matter how ‘innovative’ and ‘cool’ it is. Everyone knows you need to apply a current game engine and be willing to use it. You need to adopt the modern ways of multiplayer and Internet. Imagine a RTS that came out today with no Internet play but just modem multiplayer?

As absurd as that RTS example is, this is exactly what these “WoW killers” are doing. Instead of leveraging a modern game engine, they have that modern game engine behave like one from 2004 in WoW mimicry. Instead of having the servers talk to each other well, they adopt the server limitations from 2004 and have the servers ‘isolated’ and ‘not talking to each other’.

If any company follows the above, how much money have they saved from not throwing out yet another ‘WoW killer’ onto the graveyard? Who knows? This blog post may have saved a billion dollars.