The Making Of: Diablo
The world of game development is littered with sour tragedies: great ideas that went awry due to mismanagement, ideas that were never that great in the first place getting painfully – publicly – euthanised. It’s a pleasure, then, to consider Diablo, a title for which the stars aligned, and a team of friends landed themselves with a sympathetic publisher that genuinely understood the concept. Even the tricky mid-project acquisition of the company couldn’t damage the final product. It helps, of course, that the end result wasn’t bad either. Diablo went on to redefine a sub-genre of the RPG, the top-down dungeon crawler, with a mixture of visceral combat and loot hoovering that played out, in the words of art lead Erich Schaefer, as a series of “simple pleasures”.
Simple, as it happens, would be the operative word for Diablo. “Back then, RPGs were so overwrought with statistics that the genre had shrunk to a tiny audience,” says Erich’s brother and Condor’s co-founder, Max Schaefer. “We wanted to do an RPG how we’d played Dungeons & Dragons as kids: hit monsters and gain loot. Our mission was that we wanted the minimum amount of time between when you started the game up to when you were clubbing a skeleton.”
Condor, the company that would create Diablo, was founded in 1993, and hit a quirky seam of good luck early on. “We were just starting out with Dave Brevik and my brother Erich,” says Max. “We were in Brevik’s house and he’d just quit his job with Iguana Entertainment. We’re having our first meeting: what’s our company going to work on, how are we going to make money? The phone rings, and it’s someone from SunSoft who heard Dave was free and had some projects. So on our first meeting we ended up getting our first job. We looked at each other and said: ‘Is this real?’”
Forging a deal with SunSoft, Condor started work on the Mega Drive version of Justice League Task Force, a Street Fighter II clone using DC Comics’ ageing cast. The team was clearly on something of a roll, as Task Force quickly lead to another piece of unlikely fortune. “SunSoft had another company doing the SNES version,” laughs Max. “It turned out they were the Blizzard boys. We finally met them at a game show, and we got to talking. We became friends straight away, and mentioned to them we were interested in PC projects. They’d just released WarCraft, and their owner said: ‘We want some more PC development’. Perfect timing all over again.”
The question of what to pitch was an easy one for Condor to answer. “Diablo was an idea that I worked on for a long time,” says lead programmer Dave Brevik. “I started working on it in high school around 1985. The design changed over time, but the biggest influence was playing Moria and Angband on Unix machines.”
“For me, the most direct influence was X-COM: UFO Defense,” suggests Erich. “The size of the characters, the camera angle and the tile-based random maps. I felt like it would make a great dungeon crawl.” Blizzard agreed, and Condor soon had a contract.
The game’s central concept – loot and monsters without the waiting – was never in question, but that doesn’t mean it emerged fully formed. Surprisingly, the ultimate action-RPG was originally turn-based. “At first we had it so that you would take a step and then the monsters would,” says Erich. “You would swing your sword and then the monsters got their chance. I think this was based on the Nethack or Rogue-style of game that Brevik liked a lot.”
Blizzard suggested switching to realtime combat, and a huge piece of Diablo fell into place. “The moment I put the change in, clicked on a skeleton and my character walked over and smacked it to pieces, the clouds parted,” says Brevik, who locked himself in his office for a week to rewrite the necessary code. “The magic was found.”
With the game starting to take shape, there were still plenty of hurdles to overcome. Like X-COM, Diablo would use randomised content, handing over certain elements of the dungeon designs to the CPU itself: the coding equivalent of Russian roulette. The system had to be able to create maps that weren’t mangled, impersonal disasters. “This was the big differentiator between what was traditional and what was different to Diablo,” says Brevik. “Randomisation was very hard to implement because we didn’t have any tools to do this. The secret was iteration. We just played and tweaked the content over and over again.”
With a fairly loose structure, team members often shared roles, but, by and large, Brevik handled programming duties while the Schaefers controlled the look and feel of the game, with Max also managing the business. Certain elements, however, like the control system which pared movement, fighting and picking through loot down to a couple of mouse clicks, was something that everyone could come together on. “We always wanted to keep your options simple: left button, right button, and a few keys,” Brevik explains.
Erich agrees: “We didn’t want anything to get in the way of what some developers dismiss as ‘the grind’. Instead we focused almost solely on the grind, trying to make it fun to kill the same monsters over and over again. What we cared about was the tactile feel of smashing skeletons and constant sense of exploration. Mouse control seemed natural, although there was a lot of iteration. We noticed that anyone could pretty much play, even people’s moms.”
“Making a game simple for the player is actually harder for the developer,” suggests Max. “One of our philosophies was to make it a reward-based rather than penalty-based game. A lot of RPGs fell into the trap of penalties: you don’t eat and you die, everything you find is a penalty. With us, it even feels good to pick up a potion in the inventory and put it back down.”
Six months from the shipping date, and even though the game was coming together, Condor was running into financial trouble. Low on operating funds and in real danger of closure, help came from Blizzard itself, which offered to buy the studio, turning it into Blizzard North in the process. “The offer was unexpected, but very welcome,” laughs Erich. “The taxman was literally at the door, threatening to shut us down.”
“We clicked with the people so well,” says Brevik. “Our ideas were exactly like theirs. Making great games was in our blood. That made it easy to agree to.”
With the sale, the development of Diablo changed considerably. “At this point the budget was very low,” admits Max. “Under half a million dollars. Once we were acquired, we sat down to work out what we could do now we were free from budgetary constraints and had a little extra time. How could we make this as big as possible? That’s where Battle.net came in.”
Blizzard South’s Battle.net system would allow for free internet matchmaking in a game which had originally been designed as a LAN-based multiplayer title. “Battle.net was coming together just as we were getting towards the end,” says Max. “We thought it was a great idea: that you could push this button and play against anyone in the world seemed like science fiction. This is an era when you were typing in IP addresses to connect with anyone. Diablo was still peer-to-peer, however, so there was loads of cheating. Hacking was something that we didn’t think people would bother doing. It was foreign to us to have a hit game, to have a million people playing. It meant that thousands of people were cheating, and we were totally unprepared for it.”
Max admits that Battle.net was incorporated in a “fairly chaotic” manner but, cheating aside, it contributed hugely to the appeal of a game that was quickly becoming more successful than anyone had imagined. “There was a moment we started to suspect Diablo might be big,” he remembers. “We were developing it in relative anonymity, and we got the chance to be on a Windows demo disc. We didn’t think anything of it and then, overnight, the phone just started ringing off the hook. At least then we knew we weren’t going in completely the wrong direction.”
Massive sales followed, and with them came clones, expansions and a sequel. From quest givers with an exclamation mark over their heads to randomised dungeons and item drops, almost every aspect of Diablo was copied in some way as rivals raced to make the next game that felt genuinely ‘Diabloesque’.
Leaving Blizzard shortly after the release of Diablo II, the Schaefers and Brevik were actually among those rivals, joining up with fellow Blizzard alumnus Bill Roper to form Flagship Studios and work on Hellgate: London, a third- and firstperson dungeon crawler with its own procedurally generated environments.
Saddled with muddled payment options and a confused release, the game was, to put it mildly, not a success. “I learned more from my experience with Flagship than I did making the first Diablo,” laughs Brevik. “We didn’t have the same company fall-back options we did at Blizzard, the design wasn’t as cohesive, we had challenges in making a whole separate networking company to support us. In the end we over-reached.”
“Flagship was a good shot at the moon that almost worked,” agrees Erich. “I blame myself more than anyone, as I had final say on the design. I really hate that Bill Roper took so much heat, because he was a constant advocate for the team and the players. But you move on to the next thing.”
For the Schaefers, that was Torchlight, a beautifully crafted singleplayer dungeon crawler with caricatured environments and a deep loot system. “We’re doing this stuff much better now, with better planning and tools,” laughs Max. “I guess we’ve found that what we do well in life is make action-RPGs.” He pauses for a second. “So we’re going to stick with that.”