Hardware Classics: Unpacking The 32X, Sega's Most Catastrophic Console Failure

How do you take half a decade’s worth of critical and commercial success and flush it down the toilet? Easy – you release a device like the Sega 32X. At the start of the 1990s Sega was arguably at the height of its powers; the dawn of the decade saw the western launch of the popular Mega Drive console (Genesis in the States), which managed to gain both commercial and critical acclaim – much to the annoyance of arch-rival Nintendo. Such success was not to last, however; in less than a decade Sega’s position would be far less dominant and the firm would be forced to limp meekly out of the hardware arena.

Why this disastrous fall from grace occurred is very much open to debate, but if you keenly followed Sega’s fortunes in the middle of the ‘90s one thing is abundantly clear – the company took far too many risks when it came to video game hardware. The forward-thinking but commercially disappointing Mega CD represented the first indication that something was amiss, but many would point to the positively disastrous retail performance of the 32X as the real straw which broke the camel’s back.

Like the Mega CD before it, the 32X was a device which augmented the capabilities of the 16-bit Mega Drive, allowing it to perform the kind of 3D graphical tricks which would later become the mainstay of 32-bit consoles. However, by the time the machine was launched it was already being out-gunned by Sega’s own Saturn, and in less than 12 months it had been dropped from the company’s plans altogether. From an outsider’s perspective, the failure of the 32X was almost a foregone conclusion, so why did Sega choose to sour its relationship with the gaming public just months before it intended to release the Saturn? Scot Bayless – a Senior Producer at Sega of America from 1990 to 1994 – is the ideal man to shed some light on this turbulent time because he was there the moment Sega’s American division got the telephone call from Sega of Japan CEO Hayao Nakayama which resulted in the birth of the machine.

"We were at CES '94 in Las Vegas and Sega of America’s head of R&D Joe Miller asked a few of us to join him in his suite for a call he was expecting from Nakayama," remembers Bayless. "There had already been some discussion about an up-gunned Mega Drive with Hideki Sato and his Sega Hardware Team, but the essence of the call was that we needed to respond to Atari’s Jaguar and we needed to do it right away. Joe said he was confident the US team could come up with a design that would do the job, so Nakayama said 'get it done' and we were off to the races. Marty Franz grabbed one of those little hotel note pads and drew a couple of Hitachi SH2 processors, each with its own frame buffer. That's pretty much where 32X started."

The 32X – which at this juncture was known by the codename Mars – was actually one of two cartridge-based projects which were in development at the time; the other was known as Jupiter (as the more observant amongst you will no doubt have noticed by now, Sega had a habit of naming its hardware projects after planets in the solar system). "Jupiter started as a ROM-based unit with theoretical specs a bit like Saturn," explains Bayless. Although it was to be a more powerful machine on paper, the emergence of Project Mars meant that Jupiter was ultimately squeezed out of Sega’s strategy. "I think Sato was really feeling the cost control heat and the CD-based Saturn was hugely attractive from a cost perspective," continues Bayless. "Therefore, Jupiter was officially put to bed and Mars was born." From an engineering standpoint the machine certainly had a lot of potential. "The design of the graphics subsystem was brilliantly simple, something of a coder's dream for the day," says Bayless. "It was built around two central processors feeding independent frame buffers with twice the depth per pixel of anything else out there. It was a wonderful platform for doing 3D in ways that nobody else was attempting outside the workstation market."

Marty Franz – then Sega's Vice President of Technology – agrees. "We pushed really hard for the dual SH2 architecture," he says. "We really liked the Hitachi SH2 CPUs that the Saturn had and felt they were the star of the show. Putting two of them in a package with a good graphics buffer was a big advance at that time; it enabled software rendering tricks that were limited only by the imagination."

Despite Nakayama’s keenness to square up to the Atari Jaguar, the decision to start work on the 32X was far from straightforward and within the walls of Sega’s Japanese HQ there was much brow-furrowing over the project. This is largely due to the fact that in Japan the Mega Drive had finished in third position behind Nintendo’s Super Famicom (or SNES) and NEC’s PC Engine (known in the States as the TurboGrafx-16), and the consensus was that the company should plough all of its available resources into the 32-bit Saturn. However, Sega of Japan was savvy enough to realise that much of its current wellbeing was down to the incredible commercial performance of its 16-bit hardware in the West, and when Sega of America insisted that it was too early to pull the plug on this large market, Nakayama took notice. "The 32X was going to add additional life to the 16-bit Mega Drive market," Franz says. "This was good business for Sega, since that was where it was earning the most income." Bayless thinks timing had a lot to do with this decision, too. "There was consensus at Sega of America that making an add-on for Mega Drive was the right move," Bayless explains. "To really understand the decision, though, you need to see it in context. The 32X call was made in early January and Nakayama's mandate was to get to market by the end of the year. I think at the time he lacked confidence that Saturn would make it to market within 1994."

Although the Mars project was very much Sega of America’s baby, Bayless and his team were in constant contact with their Japanese counterparts. "The guys at Sega of Japan were great – especially Sato's team," he says. "We were all in super-double-secret-crunch-mode and nerves were wearing pretty thin. I remember one of our technical guys going completely ballistic over his dev kit losing one of its Hitachi SH2 CPU chips and then being told he'd have to wait two weeks to get a new part, but the guys in Japan were awesome. They worked their tails off to help us. We did however have a persistent problem with translations of manuals. Sega of Japan had a small localisation team in Tokyo, but those guys were completely slammed. So we started hiring translators in the Bay Area to help open up the technical translation bottleneck – with sometimes amusing results. The engineers in Sato's group were literally sending us the docs as they wrote them and then we were handing them off to contract translators in San Francisco. Technical Japanese is something of a hybrid of English and non-standard uses of Japanese language and orthodox translations can produce phraseology like: 'The cracker of remembrance receives a tickle from the command of stern ancestor accounting.' It was like a party whenever a new batch of translations came in; we'd read them over lunch and howl."

Although the 32X was meant to enhance the abilities of the Mega Drive hardware – which was half a decade old by this point, having launched in Japan in 1988 – it actually shared several similarities with the Saturn. At the heart of both machines were those aforementioned twin Hitachi SH2 processors, which were included to assist in the creation of complex 3D environments. However, despite sharing the same CPU setup the way in which the two devices utilised these chips was quite different. "Only the dual CPU architecture was lifted from the Saturn," explains Franz. "The rest was developed from scratch for the 32X. We had a short timeframe to develop the product and couldn't do much in the way of fixed function hardware development. We had to keep it simple to make the development timeline. We pushed for everything we could imagine that would enable great games in the development timeline we had."

"Saturn was essentially a 2D system with the ability to move the four corners of a sprite in a way that could simulate projection in 3D space," adds Bayless. "It had the advantage of doing the rendering in hardware, but the rendering scheme also tended to create a lot of problems and the pixel overwrite rate was very high; much of the advantage of dedicated hardware was lost to memory access stalls. 32X on the other hand did everything in software but gave two fast RISC chips tied to great big frame buffers and complete control to the programmer. To be honest, there's a part of me that wishes Saturn had adopted the 32X graphics strategy, but that ship had sailed long before the green light call from Nakayama."

When you consider the state of the market at the start of 1994 then it becomes a little easier to see why the 32X ever came to be. The 3DO and Atari’s aforementioned Jaguar were breaking through and garnering some nervous glances from established firms like Sega and Nintendo; 16-bit games were beginning to look terribly outdated and something was certainly needed to keep the momentum going. Sadly, almost from the start things didn’t go according to plan for the 32X; the aforementioned similarity between the machine and its sibling the Saturn caused numerous headaches. "Early on, the Saturn launch date was uncertain," says Bayless. "There were a number of issues bearing on launch timing in the West and, while Sega of America was busily making software for Saturn, we weren't initially fixed on a launch date. Meanwhile, the 32X had to ramp up like a rocket just to hit its timing. So what happened is the two projects basically ran decoupled from each other, which is fine if there are no dependencies between the two; unfortunately there were tons. The systems used many of the same parts, so suddenly 32X was facing shortages because chips were needed for Saturn."

Bayless and Franz – along with the rest of the hardware team at Sega of America – were essentially attempting the impossible; they were trying to bring a hardware blueprint to life in less than a year and had to fight within Sega to get the resources to accomplish the task. To make matters significantly worse, Sega of Japan dropped a bombshell which essentially wrecked the 32X’s chances of any kind of success. "Saturn got its launch date: November of 1994 in Japan," remembers Bayless with a pained grimace. Rather than being the forerunner of the Saturn, the 32X now had to face the prospect of sharing the same release window as its 32-bit big brother. "Not surprisingly, word got out quickly in the West," continues Bayless. "US and EU consumers immediately started asking the obvious question: 'Why should I buy 32X when Saturn is only a few months away?' Sadly, the best answer Sega could come up with was that 32X was a 'transitional device' – that it would form a bridge from Mega Drive to Saturn. Frankly it just made us look greedy and dumb to consumers, something that a year earlier I couldn't have imagined people thinking about us. We were the cool kids."

The earlier-than-expected launch of the Saturn had thrown all of Sega of America’s already flimsy plans into complete and utter disarray, transforming the 32X from a life-saving blood transfusion for the Mega Drive into a cancerous tumour which would further erode the company’s standing in the global marketplace. As Bayless is keen to point out, the timing was horrendous. "Sega of America had a devil of a time convincing anyone that 32X made sense when Saturn was just around the corner. Imagine how much harder that would be to sell to the public if Saturn was already on store shelves?"

As well as boosting the performance of the Mega Drive, the 32X also had the ability to connect with the Mega CD to create more impressive CD-ROM titles – that was the hype communicated at the time, at least. In reality this hellish amalgamation of hardware didn’t live up to the propaganda and those games which were specially coded to take advantage of the setup were mainly poor FMV titles like Corpse Killer and Supreme Warrior. The controversial Night Trap also got a re-release which featured better quality video, but the gameplay itself remained identical to the Mega CD original. If the lack of decent games wasn’t enough to convince you to disassemble the towering Mega Drive/Mega CD/32X fusion then the fact that it required three separate power supplies in order to function – one for each component - most certainly was. Looking back, one can understand why Sega was so keen to create all-in-one consoles like the never-released Sega Neptune.

Despite the obvious setbacks, the 32X experienced a reasonably successful launch in the West. It hit American stores shelves in November 1994, retailing for the substantial sum of $159.99. Regardless of this hefty price tag the machine shifted its initial shipment of 600,000 units with ease; it was even reported at the time that demand had far outstripped supply. A similar story can be told of the European release, which is unsurprising when you consider how much power Sega still held in PAL territories at the time. However, despite this initial interest for the device on both sides of the Atlantic, demand quickly cratered thanks largely to a distinct lack of compelling software. Indeed, it could be argued that the 32X never really saw any games which could truly demonstrate its potential and give it a solid reason for existing. "Not to be too harsh, but the launch mix for 32X was horrible," laments Bayless. "Actually, it was non-existent. Some of the games were pretty good, but in context they needed to be amazing. Unfortunately for Sega, by the end of 1994 that context had become a whole lot more demanding. When PlayStation launched in Japan, any argument in favour of 32X just sounded ridiculous."

One possible argument is that developers struggled to co-ordinate the internal architecture of the Mega Drive and 32X correctly, which prevented them from truly pushing the console to its limits. "I don't think complexity was the problem," retorts Bayless. "By then, experience with Mega CD had taught us orderly ways of spreading the workload across the various buses and chips in the combined system. I think the real issue was timing; the games in the queue were effectively jammed into a box as fast as possible which meant massive cutting of corners in every conceivable way. Even from the outset, designs of those games were deliberately conservative because of the time crunch. By the time they shipped they were even more conservative; they did nothing to show off what the hardware was capable of."

Taking this chain of events into account, it begs one fairly obvious question: did Sega's technical staff ever have any faith in the project at all? "I think 32X was a great hypothesis," he replies. "But in execution it was disastrous. Aside from the obviously murky marketing message that crippled it before it even launched, everything about the device was rushed. Nine months from a cold start is a ridiculous timeline for launching a new platform; everything about it was slammed together at breakneck speed and the result was exactly what you'd expect. The hardware was flaky, the industrial design was questionable and the games were either late or buggy – or both."

Bayless is quick to point out that it wasn’t for lack of trying, and he himself dedicated a considerable portion of his time trying to make the 32X a viable product. "I spent weeks working with id Software’s John Carmack, who literally camped out at the Sega of America building in Redwood City trying to get Doom ported. That guy worked his ass off – he sweated blood – and he still had to cut a third of the levels to get the game done in time. What amazes me now is that with all that going on nobody at Sega was willing to say: 'Wait a minute, what are we doing? Why don't we just stop?' Sega should have killed 32X in the spring of 1994, but we didn't. We stormed the hill and when we got to the top we realised it was the wrong damned hill. Looking back now I'd say that really was the beginning of the end for Sega's credibility as a hardware company."

The 32X’s dismal reception killed off another piece of hardware which Sega had on the table at the time: Project Neptune. Clearly of the opinion that you can never have too many consoles on the market at once, the plan was that following the 32X’s release Sega would launch an all-in-one machine which pulled together the internal architecture of the Mega Drive and 32X to create a console which could play both software formats and win a straight fight with the Jaguar. This bold strategy came to nothing. "By the time Neptune got into serious discussion Sega CEO Hayao Nakayama was betting the company on Saturn," reveals Scot Bayless. "That, plus the fact that 32X was clearly going to fail made Neptune pointless."

By the time 1995 arrived the writing was on the wall for the 32X. Sega of America CEO Tom Kalinske remained bullish, insisting that better titles were on their way, but Bayless feels he was making hollow claims. "We knew it was DOA. Everybody knew it, but nobody would say it. It's a phenomenon that's all too familiar in big companies; people are afraid to speak out against the company's public posture. They're afraid of hurting their colleagues. They want to believe in what they're doing, so they remain silent – but we all knew. I've never spoken to Tom about what was in his head at that point, but I suspect he knew as well. But what was he going to do? The chance to stop 32X had long since come and gone. He had to make the best of the situation he was facing and admitting publicly that 32X was a mistake just wasn't an option."

Kalinske didn’t have to make much of this bad situation for long; in 1996 he left the company he had so brilliantly taken to the top of the video game arena in North America just a few years before. Bayless had already beaten him to it, handing in his notice at the end of 1994. He admits losing faith in Sega as a whole, something which is hard for him to disclose, even after all this time. "Even now I feel bad admitting it because I genuinely liked and respected some of the people making those decisions," he explains. "When you look back at the hardware choices the company made between 1992 and 1995 it was like watching the death of the Hindenburg in slow motion. Just about every call the company made turned out to be the wrong one. Using cheap consumer drives in Mega CD, FMV games, positioning 32X as an orphan system, designing Saturn as a modified last-generation 2D system when clearly 3D was going to be the next big thing... even Sega's peripherals were stupid. Remember Activator? Sega VR? The company poured insane amounts of money and time into projects that simply didn't make sense, and consumers did what they always do. They voted with their wallets and stayed away."

Ultimately, the 32X is a mere footnote in the history of our beloved industry, but in the eyes of Bayless it represents an important lesson in how not to produce and position an item of video game hardware. "32X is a great case study in two things," he explains. "First, messaging; your number one job in marketing is to establish the value proposition. Even with all the rushed hardware and late software, if Sega had been able to convince people that 32X was really worth having it might have had a chance to succeed. But we never did that; we never managed to explain to anyone in any credible way what was so unique and worthy about 32X. The result is exactly what you'd expect in that situation: Sony ate our lunch. Second, honesty; not in the legal sense – nor in the public sense – but internally. I remember when I arrived at Microsoft in 1998 I attended an executive orientation briefing on my first day. The VP who met with us said, 'The one thing we demand of every one of you guys is to say what you think.' That attitude was what kept Microsoft vibrant, healthy and successful for more than 20 years. Sega, by contrast, lacked that ruthless cultural honesty. Nobody wanted to hurt anyone's feelings. Even when everybody knew 32X and Saturn were way behind the power curve, nobody was willing to stand up and say so. And it wasn't just the hardware; during the same period, Sega published some of the oddest games it ever released, games that were deeply flawed. Games that completely failed to connect to their imagined constituency. And all the while everyone was smiling and saying 'Gosh aren't we great?' I wasn't able to articulate all this at the time, but I know I felt it intuitively. I knew there was something wrong, that we were losing our way."

Today, the 32X is experiencing something of a renaissance as collectors dig into gaming's history in search of new experiences and challenges. Despite its tiny library of just 34 games (40 if you include those which also required the Mega CD), the 32X isn't a cheap machine to collect for, especially if you're after boxed and complete games. For some strange reason, Sega decided against employing the plastic cases seen during the Mega Drive era and instead used flimsy cardboard packaging; needless to say, this packaging hasn't stood the test of time and finding pristine, fully-boxed games at a reasonable price is getting harder and harder. The 32X unit itself is also rising in value as the years march on; not so long ago it could be obtained for relatively little on the second-hand market, but those days are gone. Another concern – especially if you intend to use the 32X along with the Mega CD – is that all three machines require their own power supply. Using the original Sega PSUs, this creates an insane amount of cables and, depending on the available power sockets in your setup, may not even be possible. Thank goodness then that modern alternatives are available.

Still, if you don't mind making use of flash carts like the Mega Everdrive, you can avoid paying through the nose for cartridges – while this might seem like a morally dubious means of experiencing the 32X's minuscule selection of titles, it's unlikely that many of its most notable releases will ever see any form of re-release; the surprisingly accurate ports of Space Harrier and After Burner have been rendered superfluous by the arrival of better versions on systems like the Nintendo 3DS, while primitive 3D games like Star Wars Arcade, Virtua Fighter and Virtua Racing Deluxe – which were best described as mildly impressive back in 1994 – are underwhelming when viewed with modern eyes. Shovel-ware like Spider-Man: Web of Fire, Cosmic Carnage and Primal Rage was embarrassing even by the standards of the time, but there are some gems to consider, such as Darxide (by Elite Dangerous developer Frontier) Knuckles' Chaotix (a Sonic title in all but name) and Kolibri (a dream-like 2D shooter from the team behind Ecco the Dolphin). The only way to experience many of these forgotten titles is to own a 32X, or resort to emulation.

Ultimately, the 32X deserves its less-than-stellar reputation; it was a compromised product which, as Bayless says, should really have been put to bed before it launched – something Sega was brave enough to do with its abortive VR headset for the Mega Drive. Still, there's something fascinating about exploring the failures of a once-dominant company; it's why the Virtual Boy, which launched not long afterwards and is Nintendo's most notable hardware flop, remains equally compelling. The 32X may not have been Sega's last throw of the dice in the domestic hardware market – it struggled on with the Saturn and Dreamcast before finally throwing in the towel in 2001 and becoming a third-party publisher – but it was arguably one of its most costly and mortifying ventures.

Damien's feature previously appeared in a slightly different form in Retro Gamer magazine, and is reproduced here with kind permission.