When Ion Storm was formed in 1996 it seemed to have everything going for it. It was the dream of John Romero, fresh from working on Quake at id Software and promising that "design is law" would be the motto of his new development studio. Co-founders included Tom Hall, the man behind the Dopefish and co-designer on cult titles such as Commander Keen and Rise of the Triad during his time at id Software and 3D Realms.
Five years on it looks like the dream is over, with publisher Eidos dropping their strongest hints yet that their relationship with Romero's Dallas-based developer may be at an end. And they're not alone - the last year has seen a number of celebrity studios kicking the bucket.
It all started with Cavedog, founded in 1995 by Ron Gilbert (of Monkey Island fame) and Shelley Day (a veteran of Lucas Arts and EA). In January 1996 they were joined by another EA escapee, Chris Taylor, who spent the next eighteen months developing the stunning real-time strategy game Total Annihilation. It was all downhill from there.
In March 1998 Taylor left the company to form his own celebrity studio, Gas Powered Games, and three years later is still working on their debut title Dungeon Siege. After that Cavedog only managed to release a couple of Total Annihilation add-on packs and TA : Kingdoms, a fantasy spin-off. One by one their other projects were cancelled as it became obvious that the games were simply far too ambitious, until in February 2000 the company was closed down entirely and its remaining staff absorbed into parent company Humongous.
There had been hopes that work on grown-up games would continue at Humongous, as Cavedog had apparently been little more than a cover name to seperate the more hardcore games from the kids' titles that had made Humongous famous. But within weeks both Ron Gilbert and Shelley Day had left the company they had founded, disillusioned with the way the gaming industry was developing, and just over a year later new owners Infogrames have effectively neutered what was left of Humongous.
Digital Anvil Hammered
The next high profile collapse was to be Digital Anvil, co-founded in 1997 by the brothers behind the Wing Commander and Privateer series, Chris and Erin Roberts. The company soon had four seperate titles under development, not to mention the truly awful Wing Commander movie helmed by Chris Roberts.
Four years later their sole release to date is a rather unremarkable space combat sim called Starlancer. In fact, most of the actual work on the game had been done by a British developer called Warthog, so arguably Digital Anvil still haven't released anything of their own after four years. With the company still soaking up money and no end in sight for many of its more ambitious titles, something had to give.
In November publisher Microsoft announced that they had pulled the plug on space strategy game Conquest, apparently because it didn't meet their high expectations. It was already two years late. This proved to be the end for Digital Anvil, as just days later Microsoft announced that they had bought out the ailing company. Loose Cannon, another long overdue title, was also canned, with the designer leaving the company and eventually striking up a deal with UbiSoft to complete work on it. Chris Roberts will be leaving the company as soon as he has finished his work on Freelancer.
Going Down A Storm
And so we come back to Ion Storm. Most people have probably forgotten that their debut release was way back in 1998. Dominion : Storm Over Gift 3 was a mediocre real-time strategy game that Ion Storm made the mistake of acquiring from 7th Level when Todd Porter joined the team. Apparently intended to get something on the shelves in time for Christmas 1997 and reduce the number of games Ion Storm were contracted to produce for Eidos, it was eventually released five months later than planned and promptly tanked at retail.
After years of hype their first fully in-house production was Daikatana, a disappointing first person shooter than also flopped after getting diabolical reviews from most magazines and websites. Meanwhile Eidos continued pouring money into the Dallas studio in exchange for more control over the company, and what with leaked e-mails appearing in a local newspaper, rumours of in-fighting, mass walk-outs by staff, and the threat of a legal challenge by a former employee, things were looking dire. The only real success to emerge from Ion Storm so far has been Deus Ex, which topped the charts across much of Europe and won rave reviews, but this was produced by Warren Spector's Austin studio.
Finally last month things came to a head. Reports began to circulate that Ion Storm Dallas would be closed down once Tom Hall's long-awaited role-playing game Anachronox was finished, and although there were denials from Ion Storm at the time, Eidos remained strangely quiet. Until this week, when Mike McGarvey told trade magazine MCV that "the truth is a lot of money went in and not much came out", adding that "I'm sure those guys will be making games in the future, but whether or not it's with us remains to be seen". Meanwhile life continues at Ion Storm Austin, with Warren Spector's team working on two new Unreal-engined games - Deus Ex 2 and Thief 3.
What Went Wrong?
Three high profile studios run by celebrity designers, all of them formed in the mid 1990s, and all of them bankrupt within five years. What went wrong? Well, the answer seems to be that developers just don't make good managers. They leave another studio or publisher to strike out on their own, create an over-staffed designer's paradise to create ridiculously ambitious new games, and then run the company into the ground.
As Ron Gilbert told GameSpot in an interview soon after his departure from Humongous, "don't try to design a game and run a company at the same time - that's a huge mistake". His own game Good & Evil went nowhere for almost three years as he struggled with the business and management side of the company, acting as producer on Total Annihilation while trying to juggle his commitments to two other Cavedog titles and a whole string of successful children's games from Humongous.
Which brings us to the second mistake that most celebrity start-ups make - they are far too ambitious. Instead of starting small and working their way back up the gaming industry ladder, they want to run an empire with multiple big budget games under development. As Chris Roberts admitted to GameSpot, "I wanted to develop not only hugely ambitious games, but too many hugely ambitious games". Digital Anvil had four games in development (Starlancer, Freelancer, Loose Cannon and Conquest), as did Cavedog at one stage (Total Annihilation, Good & Evil, Amen and Elysium). Both of them gradually cancelled or delayed their titles, often shedding staff in the process.
Meanwhile Ion Storm Dallas went down from three teams (Dominion, Daikatana and Anachronox) to just two, with Todd Porter having his Doppelganger project put on indefinite hold and getting side-lined into a management role after the fiasco that was Dominion. Interestingly, Ion Storm Austin has gone the other way. They expanded after the success of Deus Ex and are now working on at least two new games for Eidos. A good example of how to do things right?
Over-ambitious business plans and protracted development cycles took their toll on all three companies, and in the end they have all been bailed out by their publishers. GT Interactive bought Cavedog / Humongous as far back as 1997, Microsoft bought out Digital Anvil at the end of last year, and Eidos essentially owns Ion Storm.
While the day of the celebrity development studio might not be over just yet, it should be obvious by now that anybody going this route in future will have to think things out more carefully. Start small and work your way up instead of starting big and then slowly imploding. Bring in an experienced management team so that you can get on with what you do best - developing games. And make sure you have the financial backing to last at least three years under your own steam, because it's going to take you that long to build your team and develop your first game.