Gamasutra caught up with Bill Roper at GDC China for a very interesting interview about his time at Blizzard Entertainment, the rise and fall of Flagship Studios, what went right (and wrong) with Hellgate: London and Mythos, and more.
If you were to start a company, do you think that you would go the Flagship route again of a big studio? Or do you think that's less of a feasible model these days?
BR: It can be a feasible model. I think there's a lot less support for it on the financial side right now. It's just harder to start a studio at that size.
Flagship actually got a lot bigger than we ever intended it to. In our heads, we wanted to have 25 people. Like, that was how big we wanted our company to be. We had to grow to a larger size within Flagship to support everything we tried to do with that game.
The biggest failure with Hellgate is we just tried to do too much. We were a single-player game, or you could go online and play for free, and there was also this hybrid subscription model that you could get into, and the game was coming out on the new Windows platform, and we were part of the Games for Windows program, we shipped in 17 languages, we had a very high-end graphics engine that we had built but at the same time we did low-poly versions of the game. I mean, the list just went on and on and on.
Whereas Torchlight was an example of honing in on something very specific.
BR: I honestly think that as much as people hated me, as much as people hated Hellgate and hated Flagship, people would have loved Mythos. And that was the second game from the studio. And there was a lot of crossover, you know. And I think that Torchlight proved that that next game idea was...
But the thing is they're kind of doing it in this three-stage process, where they released a single-player version. Runic did the game like "Here's a single-player version." The next is going to be a peer-to-peer type thing. And then it's going to be, "Okay, now everybody goes online and plays."
The advantage we had with Mythos is we would have been right at stage three because all that tech was already there. It's a real drag. There was a version of Mythos... We did internal, and then we pushed from the internal server to beta server. And people who were playing Mythos when we closed the company, the version that was next going to get pushed to beta, which was internal, had all these changes to it based on everybody's feedback that was playing.
It wasn't hub-instanced anymore. It was a big open world that then you would go into all the instanced content. You could actually run around with people. We were like, "Oh my God, this is it. This is going to be so great." But you don't get there. And I think that's the difficulty, that we were in an unsustainable business model.
I mean, the game sold, actually, a good number of units, not a failure number of units at all. We never released our box money because we never cracked [our royalty numbers]... Because development cost was so high. We were like, "God, we'll never make that."
We even at one point just realized, "We're never going to make money off box sales. Even if this game sells multiple millions of copies, we might never make our money back on the box sales. We're going to have to make our money on the back end, on the online." Because that was a much lower nut to crack every month. But we just didn't get the number of players.