« GDC 2006: More Sony Keynote Coverage | Main | GDC 2006: "Disrupting Development" »

GDC 2006: "Designing Tabula Rasa"

Designing Tabula Rasa
Richard Garriott, NCsoft

Trials and tribulations of creating a "next generation" MMPORG

The Tabula Rasa project began slowly. We immediately had some substantial difficulties. In 2004, we made a major turn in the project -- personnel, design, and strategic changes. Since then, the project has been proceeding well. I'm going to focus on the first phase and what went wrong.

Solo versus MMP:

  • Solo player games: you are special, you are alone
  • MMPs: you are not alone, you are not special
  • MMP flaws to date: repetitive level grinding is life, farming static environment for XP and $, no real purpose, no real success, no sense of urgency or world impact
The great promise of MMP games was that you could meet people and go on missions together, so that you could have a shared experience -- we don't even like to go to the movies alone. But this comes at a big cost: in MMPs to date, your life is no longer as special as it was in a single-player game. On average, statistically, half the people you meet are going to be higher-level than you -- and that's only if you put in the time to become a high-level player.

I think that current MMP games have done a good job of refining some bad features. For example, repetitive level grinding becomes your life. That's not fun. Mindlessly farming an environment for, say, creatures -- "the Level 2 creatures always appear to the north of the town" -- isn't fun. And you no longer have the dramatic effect on the world that you did in single-player games.

DG meets NC:

  • Destination Games began hiring April 2001
  • Met NCsoft (Lineage franchise #1 world-leading MMO, merged DG and NC May 2001)
  • Goal to make worldwide bestseller
  • Forged international MMO "dream team"
  • We tried to innovate on every front (except 3D tools and technologies, which was a mistake)

Our entire team was overqualified. Even our junior designers had been lead designers on previously successful projects. This turned out to be a mistake.

The one place we didn't try to push the threshold was in tools and technologies. We felt (wrongly at the time) that we could get away with one more generation of focusing on time to market as opposed to the quality of the visuals.

Immediate trouble:

  • Too many cooks at many levels on one team (language barriers, hard to communicate design subtleties)
  • US-Asian blend (Asian art never right for Asia, Asian/US mixture compelled neither)
  • Tabula Rasa refocused as a US game first (Americans should not set out to sell Asia)
When we first merged with NCsoft, I went to a meeting in Korea where the senior staff reviewed new ideas. I listened to the ideas for Lineage 2 and offered my feedback -- what I thought was positive and well-meant. Culturally, that was a no-no. In Korean culture, that was the time to offer support, and later offer comments in private. As a result, I didn't have a great initial relationship with that development team.

Another issue was with art differences. In the US, heroic characters are often powerful and large. In Asia, though, it's common to think of big, strong guys as dumb brutes, whereas the heroes are often skinny, geeky types who succeed because of their inner spirit.

We immediately began beating our head against the wall of trying to create a unified worldwide product. We decided to refocus on one market first: the US market.

Additional errors:

  • Focus on off-the-shelf tools (theory: cut time to market; result: sub-standard capabilities and long rewrites)
  • Focus on sci-fi martial arts (plan: Warriors of Zu, Chinese Ghost Story; result: too strange for US players)
  • Focus on instances exclusively (original structure: Disneyland metaphor; result: world lacked core MMP feel)
As a result of focusing on instances, our game became almost desolate -- you never felt like you were part of a larger space.

I personally take responsibility for issuing a directive into our art department for a style that became unachievable. I wanted a futuristic version of the Art Nouveau period. We tried three or four times to generate art that manifested what I felt I could describe, and it never achieved what I wanted. Meanwhile, even our executive team was saying, "hey, your game doesn't look very good yet," and I was saying, "don't worry, it will look better when we're done." This went on way too long.

Our human clothing and architecture was uninspired. Our male clothing was so bad that everyone wanted to play as a female. Guys looked silly pretty quickly. Even our weaponry in the game was quite exotic -- we started with musical instruments that could become weapons and the like. I still have a great rationalization for why all this would be cool, but in the game, it just didn't look good.

In the fall of 2004, we rebooted the project. We all wanted to believe, and we could tell a great story. But it was clear we were failing to make our vision.

Tabula Rasa reboot:

  • Highest level of internal technology
  • Modern human foundation (architecture, clothing)
  • Focus on war elements (battlefields, shared spaces)
  • Only primary alien culture kept the Art Nouveau feel
  • Let go lots of entrenched staff
  • Hired only top artists
  • Focused on TV moments (what are the moments where people will look at this and be awe-inspired?)
  • Get it right versus time to market
  • Listen to feedback
Things that survived:
  • Intellectual property backdrop (main story, symbolic language)
  • Goals of feel of play (30-minute play cycles, story driven, no farming, no treadmill, players are pivotal to war effort)
  • Combat structure (shooter look and feel, RPG mathematics)
Challenges remain:
  • Tabula Rasa now large and moving fast (50+ in Austin, 10+ in Los Angeles, 30+ in China, 5+ in Seoul, plus outsourcing)
  • Management layers (internal and outsource)
  • Holding quality bar
  • MMP quantity (time to build, scale means no one person knows it all)
We're not outsourcing for cost savings -- we're just doing it to get the quantity of work done that we need at the right level of quality.

The big problem that all MMPs have is scale -- they're just so large to build. And the scariest thing to me is that the world is so large that no one person knows it all. How to create that volume and keep the quality consistent is a challenge.

Tabula Rasa takes place on a series of worlds, each one with its own ethical parable. We started out with an Earth-like world and then progressed to creating more alien worlds. We've finished the first world and are now working on the the second world, which is our first alien world. When we've finished the third world, Tabula Rasa will go live.

Lesson summary:

  • Too many cooks
  • High-end tools essential
  • Manage amount of innovation


TrackBack URL for this entry:

Post a comment