Who Really Killed the Dreamcast?

Its a widely held belief (or at least it is now) that the PS2 was the primary executioner of the Dreamcast. This was not what was observed as it was happening, but they say history is written by the victors, and somehow people have come to believe that Sony was the vanquisher of the DC. Allow me to posit another theory on Sega’s demise — not a new one, but one that has, perhaps, been forgotten…

Known Facts

Sega’s Dreamcast was released in America on September 9, 1999 (having already been released in Japan on November 27, 1998). It bore the logo of Microsoft’s Windows CE product, Microsoft having been a key partner in the technology and development of the hardware and software (the Dreamcast ran a version of Windows CE as well as Microsoft’s graphics APIs.) The Dreamcast’s US launch garnered $98.4 Million in combined hardware and software sales, making it the most successful console launch to date. Sega had a solid product, backed by the technology and clout of Microsoft… who would soon betray them…

Sony’s PS2 was announced in March of 1999, essentially being marketed before the Dreamcast was released. The PS2 actually hit the Japan Market in March of 2000, and made it to US shores October 26, 2000.

In the period between July 23, 2000 and September 30, 2000 – leading up to the US release of the PS2 – Dreamcast sales grew 156.5%. Although the PS2 was to be a huge blow to the Dreamcast, it’s clear that Sega was still a viable contender in the race.

At the end of 1999, in an interview with Bill Gates, it was confirmed that Microsoft was developing a game console, originally called the DirectX-Box. Development was officially confirmed March 20, 2000 at Game Developers Conference 2000, and initial Demos were shown at E3 2000, on May 13. The Dreamcast was still the only “next gen” console on the market at this point, but the writing was on the wall: Sony and Microsoft were both loading their big guns.

(Also shown at E3 2000 was a prototype of a DVD-based Dreamcast. What happened to it is a matter of speculation, and an argument could be made that the device never really existed. If it did, it was certainly pulled quickly from Sega’s booth. It may be conjecture, but it’s at least interesting co-incidence that Xbox demos were first shown at this same conference.)

In September of 2000 Microsoft held an event where the launch developers for the platform were announced, EA signed on later, on December 13, 2000. These dates correspond with the marketing material we received at Future Shop. The first prototype of the Xbox was demoed Janauary 6, 2001.
Long before its actual release on November 15, 2001, the Xbox had the mindshare of gamers, retailers and the gaming press.

By this point Sony and Sega were embroiled in a bitter battle, with Sega taking significant damage from Sony’s marketing machine, and their sold-below-cost inclusion of a DVD drive. It’s important to note, however, that Sega was still in the fight and SegaNet was launched on September 10, 2000 as a somewhat effective return volley against Sony.

Click for a chart showing Dreamcast sales trends against the impacting events. What actually happened to sales may not be quite what you expect…

The timeline clearly indicates that the Xbox was a known (if not actually existing) contender by early 2000, if not sooner. The Dreamcast was not discontinued until 2001. The sales figures for the Dreamcast also clearly indicate that while it declined in the lead-up to the US release of the PS2, it actually began to recover immediately afterward — right up until the prototype of the XBox was shown…

A First-hand Retail Perspective

Christmas is the single biggest season for console sales, and by early 2001 it was apparent that Sega would not have the capital to sustain itself against two major players through until the winter. The Xbox was far from complete, but the developers were on board, the tech demos were impressive and Sega was hurting. Despite a price cut in 2000, Sony was in the lead, the GameCube and the Xbox were forth coming, and there was no room in the market for 4 players.

Sega cut its losses on hardware, and discontinued production. At the time at Future Shop we were told that the production halt was temporary – that there was plenty of hardware in the pipe and production was stopped to save costs until the demand increased. This was not reported as the death of the Dreamcast – not to the market, not yet. We continued to get hardware and software on our shelves, new (to North America) accessories arrived, including the accelerometer+motion-sensor based maraca controllers, after this date, as well as newly released software titles.

Nonetheless, as the Microsoft launch date approached we were instructed to move Dreamcast stuff to lower shelves, and Microsoft marketing reps began arriving in our stores with posters, stickers, pens and other collateral. Space that had been used for real Dreamcast games was now being used for cardboard representations of not-yet-existing Xbox hardware. The system didn’t even exist yet, but already Sega was being offered up as a sacrificial lamb.

The final death knell of the little white Sega box came with the announcement that NHL2K2, Sega’s strongest competitor to EA Sports and the latest in the best-selling series, would be the final software release by Sega for the Dreamcast in February of 2002. By this point, the Xbox had been on the market for months, and had a strong Christmas season, along with Sony.

By 2005 Microsoft had posted a $4 billion loss on sales of the Xbox. The hardware was sold drastically below cost in an attempt to break into the market. Sega, a company a fraction the size of both Sony and Microsoft, could not have hoped to compete with both of them.

Had the battle been between the PS2, the Dreamcast and the initially disappointing GameCube, the result would have been different. Technically the PS2 was difficult to develop for, and despite the hype was only marginally more powerful than the Dreamcast. The GameCube took years to find its niche and was generally considered a laughing stock for most of its early life.

These are facts that are not in question by anyone who was around at the time. I offer them second, only because there are no Wikipedia articles that explain what we all knew: in every round of the console wars, there has been room for at most 3 contenders. Sega had the least capital, had the most to lose (except maybe Nintendo) from holding their ground, and had the first (and therefore, the oldest) console in the fight.

Anecdotal evidence illustrating Microsoft business practices

Granted, this isn’t hard evidence, but I think I’ve already proven that the Dreamcast was able to stay in the fight only until the Xbox marketing machine (marketing vapor, but powerful nonetheless) kicked in. What follows goes to motive and a pattern of behavior that was observable in Microsoft’s practices and treatment of its partners. The story of how IBM first partnered with, and was then betrayed by, Microsoft in the development of OS/2 is the stuff of history books, but in case you’re not clear on the facts (This data is broadly available online, but much of it was plagiarized from this useful timeline):

• In 1987 Microsoft and IBM announced a partnership to develop OS/2 and released OS/2 v1.0
• In 1988 Microsoft and IBM announce the delivery of the jointly developed OS/2 1.1
• In 1989 Microsoft and IBM broaden the scope of their development agreement by agreeing to jointly develop a consistent, full range of systems software offerings for the 1990s, which would include MS-DOS and OS/2
• In 1990 Microsoft announces Windows 3.0. Both Apple and IBM, current partners with Microsoft on software development for their respective platforms are shocked. Apple sues.
• 1991 Excel ships for Windows, Macintosh and OS/2 simultaneously
  • IBM attempts to de-tangle themselves from Microsoft, shipping OS/2 1.3 without their help
• In 1992 Windows 3. 1, the first serious Microsoft UI, ships
  • Microsoft begins work using technology developed in partnership with IBM and the OS/2 project on Windows NT
• In 1993 IBM is now forced to reverse-engineer Windows APIs to maintain the arrangement with Microsoft to create “consistent software offerings.” These are included in OS/2 2.1
  • Microsoft is no longer active in the development of OS/2
• October 1994, OS/2 Warp 3.0, a rich OS including SMP support for up to 16 processors, pre-emptive multitasking, a 32-bit graphics engine running on a true 32-bit OS and compatibility with Windows 3.1 is released. By all accounts it’s the most powerful non-UNIX x86 OS on the market.
• Also in October 1994, Microsoft with no real offering available yet, announces the Windows 95 Preview, aka Chicago. Its vaporware, but the message is simple: Wait for what we have to offer, it’s going to be better.
• August 24, 1995, Windows 95 is released. Service Pack 1 follows later in the year, at which point the OS actually becomes viable. The hype leading up to the release of Windows 95 is now legendary in the marketing world, with around $300 million spent
  • Despite the promises, Windows 95 is NOT Chicago and not based on NT technology. Instead it’s a 32-bit re-implementation of Windows 3.1 with an improved user interface.
  • People buy it anyway, responding to the marketing dollars poured into the project. OS/2 marches on for awhile, but Microsoft is no longer sharing APIs, and OS/2 eventually fades into obscurity and is canceled.

I don’t want to belittle Windows 95 too much. It wasn’t what they promised, it was a while before it was actually stable, and it wasn’t nearly as ambitious as we’d all hoped. But it was an important milestone. Nonetheless, it came at the expense of one (or more) of their partners. IBM’s software team, at the time, was not the behemoth you think of when you think of IBM. They were, for a long time, belittled within IBM’s organization. Personal Computers were thought of as a toy by Big Blue. The OS/2 team was, for all intents and purpose, a dynamic little company within a company, with great ideas and a stellar product. Unfortunately, they foolishly signed on with a treacherous partner…

Comparison of the OS/2 experience and the Dreamcast:

Sure, these are conclusions, so you can take them with a grain of salt. But if you look at it objectively, even the biggest Microsoft fan boy has to at least be aware of the similarities between these two “partnerships:”

• Microsoft and Sega partnered to release a product that could run a Microsoft OS, and used Microsoft APIs. I’m not familiar with all the touch points, but its apparent to anyone who looks that the original Xbox was, at the very least, a spiritual successor to the Dreamcast – with similar controllers, similar focus on online play, and even similarly enforced UI rules (for example: the B button is always back, the A button is always accept – such rules are not present on the other two big consoles and were written in the Sega developers guide.) Similarly, much of the initial Windows experience was lifted from OS/2, and in fact, co-developed with IBM.
• During the life-cycle of that partnership, Microsoft announced their intention to formally join the market on their own.
• While the heart of their partner’s product was still beating, Microsoft began developing a competing product based on the same root technologies they’d signed on to share.
• Before they had a product to sell, Microsoft began marketing the snot out of it, filling the market with uncertainty and forcing consumers to ‘wait and see’ if Microsoft’s offering would be a better investment.
• Microsoft spent millions, accepting massive operating losses to basically out-spend the competition – not out-innovate, or out-perform: it was simply a matter of who bled-out sooner. The smaller company, of course, had to fold.
• As OS/2, arguably a superior OS, certainly a valid competitor, had to give up, so did Sega.

Now, I’m not saying any of this activity is illegal. Certainly Microsoft has a right to back out of a partnership; certainly they have a right to use their own technology to develop their own products. And certainly they’re entitled to use their money to market their products.

What I’m saying is that Microsoft has a pattern of using a partner as a way of researching a market, then using what they’ve learned from that partnership, and their immense operating capital, to crush their former partner. In some cases, the market benefits from this. In some cases, the partner survives, learns from the experience, and adapts their future partnerships to be equally treacherous (see Apple and their partnership with Microsoft on, then later entry into, the browser wars.)
Sadly, in the case of the Dreamcast, the market suffered. Granted the Xbox was a fantastic console, and Microsoft’s below-cost sale of the hardware allowed the living room to make an exponential leap-forward in technology. But what we lost was the little company with some wacky, innovative ideas that made this space – one usually filled with copies of copies of game types invented in the early 90s – a little bit more creative and fun.

The facts show clearly that Sega was still in the running against Sony – despite the hype, despite the DVD player sold below-cost. They had fresh ideas, and a take on on-line play that Sony didn’t (still doesn’t) get. Sega’s hardware, save for the DVD player, was technically competitive with the PS2, and their sales put them in a strong second position. It wasn’t until Microsoft announced the Xbox, talked it up, then began marketing their not-yet-existent product, that Sega was forced to give up. That their first real defeat came before the Xbox was actually for sale is only an indicator of Microsoft’s standard operating procedure when it comes to entering a market: using former partners, and marketing vapor…