Light from the Darkness: The Making of Torchlight

PC Gamer


torchlight thumb

Diablo casts a long shadow over the action-RPG. Few games have been able to dispel Blizzard's gloomy dominance of the genre, though many have tried. Only Torchlight shines in the dark. A modest game in scale and price, made in a year by a small team, it is as accomplished a resurrection of the genre as anyone could want.

It's cheery and accessible too, doing as much as it can to delight the player as they mop up hordes of monsters in its colourful, randomised dungeons. Now its developers, Runic Games, plan to transform this success into a platform for an MMO based on the Torchlight world.

Despite its unassuming, accessible look, Torchlight's creation was a far from trivial task. Its DNA can be traced back to its makers' work on the original Diablo, after which their careers criss and cross the wiki pages of MMO and RPG history. Its long gestation survived the unmaking of Flagship Studios – and their almostmade Mythos, a Diablo-style MMO that sank along with Flagship in August 2008. It was only a month away from beta.

Yet within 24 hours of Flagship's dissolution, the Mythos team had started anew, more determined than ever. A month later they became Runic Games. Then, by October the following year, there was Torchlight.

You won't have to pay for the best gear in the game, Runic say.

“I guess it started back in 1993,” says Runic Games' CEO, Max Schaefer. “Rather than get real jobs and be responsible people, my brother Erich, David Brevik and I decided to start a game company. We were called Condor Incorporated. We cut our teeth on Gameboy and Game Gear games, and ended up doing a Sega Genesis game called the Justice League Task Force. It was a Street Fighter-style game with Batman, Wonder Woman and Superman. The publisher had another company entirely make the Super Nintendo version, but didn't put us in touch with them. We met them at one of the big game conventions and, of course, we had an instant rivalry – but they were really nice guys and we became friendly. They turned out to be the group that started Blizzard, although they had the horrible name of Silicon And Synapse at the time. Anyway, when we were looking for the next thing to do, we decided we wanted to make a PC game because we didn't know which console was going to be the next big thing and didn't want to bet on it. So we wandered round the shows with our pitch packets looking for someone to give us money. The Blizzard guys were one of the few that gave us the time of day. They signed us to make Diablo. And off we went. Then, half way through making it they acquired us and we became Blizzard North.”

Over the following decade, Blizzard North grew to a 60-man company, creating the subsequent Diablo expansions and sequel. But the Schaefers got nervous about the corporate ownership of Blizzard. They'd been passed between so many different divisions of parent company Vivendi that no one could keep track of who their bosses were.

“We would read on Yahoo news every day about who they were going to sell us to next,” says Max. “We decided we wanted to be in charge of our own fate again, so a bunch of us left Blizzard North and started Flagship Studios. We started at a good time. Publishers were throwing around money, and wanted to hear stories about how you were going to do the next great thing. So we started Hellgate: London which was going to do everything for everybody.”

Flagship, needless to say, did not meet expectations with Hellgate: London. It was meant to be the AAA breakout monster hit – an MMO that was a perfect amalgam of Half-Life and Diablo's finest features. It was, instead, a total disaster. But while Max Schaefer is frank about the mistakes that were made at Flagship, it's clear that, alongside overpromising, Flagship fell victim to circumstances outside their control.

“I don't want to throw anyone under the bus,” says Max Schaefer. “But there were a lot of logistical things that went wrong with Hellgate: London. For example, our primary PC publisher ran into some trouble. MMOs take a considerable amount of operations. You have to be able to take people's money, you have to do customer support, run servers – that was supposed to be done by other people, and yet half way through the project it was announced that they weren't going to be able to fulfil that, and so we had to figure out how to do it ourselves. We didn't really have any money for it, we had to hire up quickly and it stretched our management thin. It was just one more thing that made it too much to do in the budget and the time we had available.”

While the troubles on Hellgate edged the company ever closer to the abyss, more promising developments were underway at the studio's Seattle branch – itself something of a serendipitous creation.

“Mythos began as a test of the network infrastructure we were making for Hellgate,” says Max Schaefer. “We just needed a couple of guys doing something super basic to make sure it worked. That's when we hired Travis Baldree.”

Travis Baldree, creator of Fate, a Diablo clone much admired by Diablo's original developers, had interviewed at Flagship several months prior – but turned the job down, due to the hassle of relocation to San Francisco with a two month old baby in tow.

“Somebody struck on the idea of doing a small test game, the canary in the mineshaft, to prove Hellgate's network tech in front of the launch,” says Baldree, now president of Runic Games. “So I got the opportunity to work on that remotely. What became Mythos was just me, working out of my house. I was Flagship Seattle for about six months.”

It became quickly apparent that Mythos had potential.

“There was some talk around the office that this was the game we should have been making in the first place,” says Erich Schaefer, brother of Max and Runic Games' chief creative officer. “So we said, let's turn this into a real project – it'll still work as a network test, but let's staff it up and make it into a real game.”

But even with that decision made, the future of Mythos seemed far from certain.

“For the entire time we were making that game we were essentially putting a square peg into a round hole,” says Max Schaefer. “It was rough – it's not how you'd want to make a game.”

“It was not a normal development cycle,” agrees Baldree. “There was a lot of reassessment of game features. The game changed significantly over the course of its life as expectations changed. Initially it really was supposed to be a one person project, and the scale was based around that, and as it grew we had more people and more energy to invest.”

And then, just as the project approached beta, Flagship came apart entirely. The Mythos team lost everything they'd worked on, with the game's IP and assets promptly sold off to South Korean MMO publisher HanbitSoft. Not a single pixel or line of code could be rescued.

It was “horribly demoralising,” Max Schaefer admits. “These guys had put a couple of years of work into Mythos and, boy, it was just snatched away at the last minute. It was a really horrible time. You could see it crashing like a train in slow motion. It didn't happen overnight. It was hard to watch. In the later stages of blowing up, I was saying to my brother, we've got to take a year off after this and decompress. But then we realised that we'd lose the team unless we jumped on it and made something happen.”

The ogre, that's Hellgate's demise. The goggles are experience, and the lightning is optimism. Yes.

“The day after Flagship shut down, everyone got together in my back yard and said, hey, we're going to start up again,” says Baldree. “So we moved about five blocks, bought all our old equipment at liquidation and started Runic. We were incorporated within the month – and that was only because of the paperwork.”

“My brother, myself and Peter Hu [another Diablo II and Hellgate stalwart, now Runic's CTO] threw in some money just to keep the door open and set about looking for a publisher,” says Max. “We wanted to do something quickly – partly to get money in the door, but mostly just to get something on the shelves so the guys could clear the whole Mythos tragedy out of their heads.”

“By the end of Mythos's development we really knew what we wanted to be doing,” says Baldree. “We'd bumped into most of the walls we were going to bump into, faced most of the pitfalls. We had a pretty good idea of what would make a successful MMO action-RPG. In starting Runic, one of the big lessons from Flagship was not to attempt more than you can safely complete. So right from the beginning we talked about having an intermediate step: we'd do a singleplayer first, get our feet under us.”

“There were advantages to starting over, although it was painful,” says Erich Schaefer. “All Mythos's technology started with Hellgate and it wasn't really appropriate. The graphics engine was designed with Hellgate's over-the-shoulder perspective in mind. Even though it took a little time to ramp up with Torchlight, we could make stuff that was more appropriate for us. Better performance is one obvious result.”

“We put a lot of emphasis right at the start on good development tools,” says Max. “We had an ambitious schedule, so we set out to make tools that would make everyone as productive as possible – and because we'd already had a dry run at making this game, everyone had a wishlist of things that would make their tools that much better. Level designers could test levels and quests in the editor without having to bug a programmer. That really sped up development. Secondly we set out to make a game that would run on everyone's machines. Low minimum specs were a big, big point for us.”

“We ended up using the Ogre 3D engine, which is open source,” says Baldree. “It doesn't really cost anything, and it's got really great support for older hardware. And given the size of the development team we didn't want to spend a ton of time developing the latest pixel shader for a graphics card that only 20 or 30 people have.”

The result is something that the team refers to as an AAA-casual title – a small, polished creation that matches the production values of major developments with the economic ethos and accessibility of indie development. Baldree points to a gap in the market for quality 'impulse buy' games outside the myriad Bejeweled clones. Lower price points and easy digital distribution also make it a more appealing prospect in these days of high piracy concerns.

“We were rebelling against the whole five-year 30-million-dollar project,” says Max Schaefer. “Partly because it had just blown up our company, obviously – but also I don't think it serves gamers all that well. You just don't get that many games coming out. I mean, what's it been? Nearly ten years since the last Diablo? And part of it is because the giant scope and budget of each project. I don't think it really results in substantially better games. Obviously WoW is an exception, and there are others. But as game makers we kind of feel that more ideas and faster projects are better for our psyches. So we thought there was a gap in the market in between casual games and these bloated Hollywood-style projects, which would give people good gaming experiences, and new ideas without taking five years and costing 50 bucks for the consumer.”

It's an ethos Runic are applying to their future developments – principle among which is a massively multiplayer Torchlight game. In keeping with its affordable, accessible forebear, the as-yet-untitled MMO is intended for release as free-to-play, supported by microtransactions.

“The general vision is to have an MMO that plays as much like the singleplayer as we can get it,” Baldree says. “It'll have the same focus on relatively fast action: carving your way through hordes of monsters with a large number of hugely devastating skills. The MMO will obviously be different in that there'll be large static shared spaces that will not be randomised. That's one of the things we learnt from Mythos: you lose a sense of community and place if everything is randomised. So we decided to do a mix of the two.”

Maintaining the tactile feel and relentless pace of combat will also be challenging when player numbers increase, and Runic have been busy experimenting, using their own Torchlight multiplayer hack.

“The game has to be solo-able,” says Baldree, “but players still have to have skills that make them useful in a party and encourage them to work together. It's a bit of a balancing act.”

There are other challenges too: defining longer term goals, cultivating trade skills and commerce, building guild systems and PvP support, overhauling the character classes and introducing extensive character customisation.

Runic have also been keeping a close eye on Torchlight's inventive modding community. “It's been useful to look at what people think are the most important things to change in the game,” says Baldree. “The class mod stuff is the most fun, seeing people take the skills apart and put them back in new ways. Somebody made this steampunk robot class mod that fires missiles and had a machinegun. A lot of people also don't like item identification. If we keep it we are going to make it less onerous. Some of the things we're talking about are making identification automatic for items that are lower than your level. Or a passive lore skill that could automatically identify stuff for you within a certain range of levels.”

One of the biggest questions that remains is how Runic will make money out of their MMO. While the exact breakdown of purchasable items is still nebulous, the team have taken a principled stance. “I will never, ever buy a microtransaction item,” says Baldree. “I am that kind of player. And the game has to be enjoyable for me too. We don't want our monetisation stuff to offer ways to skip the game because the game is boring. We don't want it to feel like an enormous grind, or that you want to get past a bit because it sucks. It should be fun to play and you should want to do it; what you should be paying for is variance – additions that make it more enjoyable. Mounts are a great example. Everybody should be able to get mounts but you should be able to buy super special mounts that look cool or travel a slight bit faster or give you a secondary benefit.”

Runic have some other interesting ideas: leftovers from Mythos's proposed 'cartography' scheme, whereby a player could pay a tiny amount to have a random dungeon rolled especially for them.

“You could control how many bosses you might fight or the number of champions that spawn, or the luck factor in the dungeon,” says Baldree. “So if you were in a party situation and you just wanted something quick to do together you could buy an item that'd allow you to do that.”

“We don't want to sell anyone the most powerful sword,” says Erich Schaefer. “Maybe you can spend more money to increase your chances to getting these things, or speed up your levelling by a reasonable factor. Although it may be that the successful guilds need to spend a little more money, but we'll look into that.”

“We talk a lot about guild houses and outfitting them with trainers and quest givers and stashes,” explains Baldree. “But mostly we want the game to be fun.”

“I think our overall business philosophy is that we just want to have the widest base possible,” says Erich. “We want as many free players as possible. We'll worry about converting a few into payers as a secondary thought.”

With the accessible, cheap Torchlight filling the gulf between Diablos, Runic have certainly generated goodwill and an eager audience for an MMO sequel.

There's no doubt that the company feel their current fortunes have been hard won. But if the lessons from Flagship were painful, they've ultimately made for a better game.

“It's all water under the bridge,” says Baldree of the traumatic last days of Flagship. And given the success of Torchlight and bright prospects for its MMO incarnation, it's not hard to see the silver lining. “In the end, we had a really extended prototyping period on an action-RPG MMO and we got a really great team out of it. It was a net win.”

Martin Davies

Around the web