Final Fantasy 14 killed by Square Enix’s stubbornness, reborn by a new approach

Final Fantasy 14: A Realm Reborn director Naoki Yoshida gave a presentation at the Game Developers Conference today to dig into the behind-the-scenes story of why the initial version of Final Fantasy 14 failed and how Yoshida and his team managed to pick up the pieces.

What went wrong

Beginning with a history of Final Fantasy 14, Yoshida noted that the game originally launched in 2010 to an extremely negative reaction from both fans and press. It had a Metacritic average of 48 and suffered from unstable servers — up to 400 crashes per day! a lack of content and minimal story in a franchise known for having plot as one of its core parts.

Yoshida pointed to three main reasons for Final Fantasy 14's failure in that first form. First, the team had an "unhealthy obsession with graphical quality." In a series of videos comparing the original version of Final Fantasy 14 to A Realm Reborn, he showed that the core game may have looked slightly better, but its areas were also empty and it took up a lot of processing power.

For a specific example, Yoshida showed a screenshot of a flowerpot outside of an inn in the first form of Final Fantasy 14. He called it "the loveliest flowerpot in an MMO," but then revealed its heavy cost: That single flowerpot contained over 1,000 polygons and 150 lines of shader code, meaning it took up as many resources as a single player character.

In order to accommodate such heavily detailed graphics, the original team made compromises that went against the reasons why people play MMOs. For example, they limited the number of player characters on screen at any one time to 20. While this allowed the game to look beautiful while still running, it lost sight of how great it is to log on to an online game and see dozens of players running around at once.

The second reason for Final Fantasy 14's failure according to Yoshida was the lack of MMO knowledge and experience amongst the game's developers. Yoshida noted that because MMOs are so time- and resource-intensive to create, not many Japanese teams make them, which means not many Japanese people — developers or otherwise play the genre.

Square Enix was able to create a huge success in Final Fantasy 11 in spite of this problem, but they ignored massive changes in the MMO market in the eight years between the launch of 11 and the launch of 14. What Yoshida called "Square Enix's unique culture" allowed the 14 team to fixate on graphics while ignoring the fact that ever since World of Warcraft MMOs had been all about content volume.

Yoshida said Square was also stubbornly adhering to policies that had proven successful for them in the PlayStation 2 era. At that time, Square's development process was solidified as employing teams of highly-talented "masters" to "craft" each game. But in the PlayStation 3 era — and with titles like Final Fantasy 14 going to PC as well the size of team needed ballooned. They didn't have enough "masters" and didn't know how to incorporate new technologies into their process, which led to longer development periods.

Yoshida said this issue has impacted Japanese development as a whole, but he believes A Realm Reborn managed to avoid it.

Yoshida's final reason for Final Fantasy 14's failure was Square's mindset that every problem could be patched. The game had major flaws in its design from the start, but they believed that once it was up and running, they would be able to figure out a solution within that framework one way or another. It was a lack of planning, according to Yoshida.

Finalfantasy14_2_1080

The fix

Yoshida was pulled in to the Final Fantasy 14 team in December 2010 and given the unenviable task of fixing this incredibly broken game. He decided that the game's problems were such a core part of the product that their best bet was to totally rebuild the game from the ground up, but they also decided to continue creating content for the original version while they rebuilt.

In order to turn Final Fantasy 14 around, Yoshida had to completely upend the team and the method for development they had employed previously. He created a workflow where the server technology, graphics engine, level design and content could all be built at the same time.

He also streamlined the design process considerably. Yoshida estimated that he made around 400 core design decisions for A Realm Reborn, eliminating time lost waiting for approvals. He only gave lead positions on the project to the few developers who had experience with other MMOs, ensuring that they would be able to implement a long list of mandatory "standard" features.

These expected features are one of the reasons MMOs are so difficult and risky, Yoshida said. He showed a slide with a comically long list, including login server stability, highly customizable characters, an endless variety of gear, fully developed quest lines, public quests and much more. Laughing, he said that the MMO audience expects each of these features in a game at launch, and that's why it's such a difficult undertaking.

Going over the full development process for A Realm Reborn, Yoshida said they worked for about two months to nail down design and then had programmers begin building the game. He refused to let employees start coding until the systems were completely designed, which freed up programmers to keep updating the original, again streamlining their process.

He also forced lesser-experienced staff to play the original game over and over, to make sure they understood decisions and were focused on the gameplay experience above all else. While some questioned the need to continue updating the original Final Fantasy 14 while working on its replacement, Yoshida felt it was important to earn back the trust of players who had bought into the first game. Those updates also allowed them a testing ground for new features and ideas as they worked to solidify the standards in A Realm Reborn.

Another important part of Yoshida's process was reaching out to the community. He said that for as much as we might not believe him, he and his team actually read the Final Fantasy 14 forums obsessively and actively discuss and implement fan feedback. In addition to internal updates, Yoshida began a regular livestream during the development of A Realm Reborn, which helped connect fans to the game and also fostered motivation for the team. Square Enix management thought it was risky to be so open about development to fans, but the players loved it so much that Yoshida and his crew are still doing livestreams now.

Finalfantasy14_3_1080

The future

Yoshida said Final Fantasy 14: A Realm Reborn was an industry first — "the first MMO to ever be rebuilt and relaunched." All told, his team managed this feat in an astoundingly short two years and eight months while continuing service and updates for the original. By comparison, Yoshida said it's common for a massively multiplayer game to be in development for as long as five years.

Speaking to the future of the MMO genre and providing advice for other developers, Yoshida warned that no matter what, you're going to be compared to World of Warcraft. "Comparing A Realm Reborn at launch to World of Warcraft right now is like comparing a child entering elementary school to a college student," he joked. But he's not scared by the comparison anymore.

He urged that proper planning is of key importance for an MMO, especially in consideration of long-term updates to the game. "Running an MMO is like running a country," Yoshida said. "Dissatisfied players will leave."

However, he said, as long as players are complaining, they still care. The trick is to learn to listen and adapt based off of those complaints. He also said it was important for creators to play their own games. Returning to the country analogy: "If the leaders don't live there, why would their subjects?"

Wrapping up, Yoshida said that Square Enix itself has learned many lessons from the struggle with Final Fantasy 14. While the initial game was a massive failure, it was an important opportunity for the publisher to reassess its methods and grow. And this is just the start.

"A Realm Reborn is still a baby chocobo in the MMO field," Yoshida said. "I look forward to continuing to nurture it."

In This StoryStream

GDC 2014: Panels, announcements and more, live from San Francisco
  1. Apr 4 21 comments
    Exploring consciousness, suicide and the quiet horror of SOMA
  2. Mar 20 48 comments
    Final Fantasy 14 killed by Square Enix’s stubbornness, reborn by a new approach
  3. Mar 20 18 comments
    The Last of Us cleans up at the 2014 Game Developer Choice Awards

Top Stories

More from Polygon

The horror of Five Nights at Freddy's

  • Dark Souls 2: Crown of the Old Iron King Overview

  • Spacecom: a fast 4X built for multiplayer

  • Pillars of Eternity builds on role-playing classics

  • Tour the 1 KB hard drive built inside Minecraft

Latest Discussions

X
Log In Sign Up

forgot?
Log In Sign Up

Please choose a new Polygon username and password

As part of the new Polygon launch, prior users will need to choose a permanent username, along with a new password.

Your username will be used to login to Polygon going forward.

I already have a Vox Media account!

Verify Vox Media account

Please login to your Vox Media account. This account will be linked to your previously existing Eater account.

Please choose a new Polygon username and password

As part of the new Polygon launch, prior MT authors will need to choose a new username and password.

Your username will be used to login to Polygon going forward.

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.
Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_5353_tracker