Sega kills 3Dfx, hands market to Sony, paves way for Xbox

Dramatic title just to have a little fun with the topic; don’t take it too seriously. I’ve been wanting to toss these little historical bits together for awhile. Please excuse my rambling :)

It was an interesting time in the gaming industry as the Dreamcast unveiling approached. 3Dfx and Sega were well down the aisle together, when at the last minute Sega got cold feet and ran back out the door. Turmoil (and lawsuits) followed.

One of the biggest stories I’d like to hear is exactly what, at Sega, led to that decision. The fallout from that event lies somewhere between tremendous and astounding.

Legal wrangling aside, the most immediate impact from that decision was shock. And it wasn’t just 3Dfx; developers too were shaking their heads. EA publicly stated (in a Next-Generation interview) that they would not be supporting Dreamcast, and that without their support they didn’t think Dreamcast would survive on the market. The reason given was the graphics decision; EA had tremendous Glide expertise, and had no interest whatsoever in investing time and money to develop tile-based rendering skills. Assuming the graphics chip aspect was a truly significant contributor to their decision, and not just a handy public excuse, the decision by EA to not support Dreamcast leads directly to the next branches in the story, as the repurcussions of the Sega boardroom decision begin to ripple through the industry.

The console world lost something that could have changed the industry (well before Microsoft set out to do it): a mainstream console could have been produced using a graphics chip that at the time was part of an architecture that already dominated the PC industry. Glide rocked gamers. Simply having that part on board would have guaranteed substantially more Dreamcast titles filtering over from the PC world. From a more impressive launch lineup, to the EA franchises (if you believe their story at the time), to countless other possible ports, the inclusion of 3Dfx architecture would have had a significant impact on console and PC gaming in general. And for Sega and 3Dfx, the effect would likely have been tremendous. For one thing they’d both probably still be in the hardware business.

But that was not to be.

Dreamcast was doomed. Sega’s software was strong, but some of their most important franchises for the US market had struggled since their inception on Saturn; their sports games were technically well-regarded, but lacked the mass appeal of Madden and the other EA franchises. Despite a reasonably successful launch, sales were disappointing. Within a few months of the Playstation 2 release, the handwriting was on the wall. That recent history is told elsewhere.

Back in the PC market, 3Dfx shifted gears. Internal teams presumably were re-assigned, but it’s possible the damage had already been done. They had not only lost development time by having a team distracted on console development for whatever period of time they must have; they also lost a residual income that could have later been key to surviving their own coming missteps. There must be people with some very good stories about the last months of 3Dfx; VS100 delays, Rampage development. Some of the key questions come down to Rampage. VS100 clearly missed it’s window of opportunity, which Nvidia capitalized on. But what about Rampage? How good was it, really? And how close? If it was truly a competitive part, well-timed, that simpy ran out of time as the coffers emptied, then the lack of a steady revenue from console sales could be considered the final damning insult as life drained out of the company with the goal in sight, leading to the announcement that shocked gamers worldwide.

Toss that console revenue stream in there, and it and it’s implications change things dramatically for 3Dfx. First for Rampage: assuming the intellectual property was valuable, the addition of a steady revenue stream either pays for continued production, or at the least serves as very good reassurance to a capital partner. And that’s ignoring the other aspect of this altered-history: had 3Dfx been included in Dreamcast, not only would there be a revenue stream, but the position of Glide in the marketplace would have been enhanced, leading most likely to increased Voodoo sales, possibly preventing the budget crises in the first place - despite the significant missteps by 3Dfx over that period. But it wasn’t to be. The company slowly bled to death, with no rescuer aside from a gleeful (and now wealthy) competitor, and quietly closed it’s doors. Gamers moved on, and the casualty list reached two with the departure of Sega from the hardware business.

Microsoft, despite the loss of a hardware partner in Sega, must have been pleased; Glide had been on life-support for some time already, and now was officially dead. It had been an important part of the PC gaming puzzle, and it wasn’t developed in Redmond. They would waste no time in filling that vacuum. And while one opportunity to create a standard that blurred the lines between consoles and PCs had passed, the obvious potential benefits of the idea were not lost on Sega’s one-time console partner. It remains to be seen whether that perceived opportunity can truly pay off, but one embodiment of that original concept exists now in XBox.

Sony, greatly aided by a weak Dreamcast showing, captured the marketplace again. You can’t help but credit Sony’s success to their own execution, but it also gives pause to think of the year before PS2 as it might have been. What would that market look like now? It’s likely that the overall market would be larger to the benefit of both companies, but PS2 would be unlikely to enjoy all of it’s current dominance in such an environment.

And Xbox might never have been. Not yet at any rate, and certainly not with the uniqueness of their current (albeit so far largely an unfulfilled promise) symbiotic relationship with the PC game market.

Finally, in perhaps the most ironically humorous aspect yet, the story turns back to EA. Sega is making games again. On most of the platforms EA is on. In many of the genres EA holds dearest. And the entire existence of Sega as a direct cross-platform competitor can be traced back to EA’s own decision to not support Dreamcast (which in turn can supposedly be traced back to the termination of the Sega / 3Dfx partnership). Given Sega’s history as a software developer, it seems hardly likely that the good folks at EA couldn’t surmise for themselves what would be the outcome of a Dreamcast failure.

But encourage it they did, and in doing so they have welcomed back one of their oldest competitors. Shareholders can assume that for every sale of a Sega game in a genre and platform EA sells to, that that money would most likely have gone to EA instead, had things turned out differently. EA management’s role in creating that outcome could reasonably draw the ire of some of those shareholders, I would think.

One of the biggest stories I’d like to hear is exactly what, at Sega, led to that decision.

The answer to a wordy post is pretty simple. Sega of Japan calls the shots, not Sega of America. Both 3dfx and NEC were competing for the graphics chip and both were designing at the same time. Sega of America thought it was a lock to get the 3dfx chip on board due to 3dfx’ success in the market and their chip’s perceived superiority.

The problem was that 3dfx was an American company and NEC is Japanese. It was along nationalistic lines that the decision was made (in Japan) and Dreamcast became an NEC PowerVR-based console.

When it comes to Japan, they’ll always choose a product designed/built in Japan over one that’s designed/built in the US. You need look no further than Xbox for the confirmation. In fact, I think you’re overestimating the effect of the 3dfx chip’s potential to make the DC a success. Most developers were unfortunately already writing off the DC before it was released. It didn’t matter what was inside. Sony had sewn up the market and PS2 was going to “win” no matter what. I heard other journalists writing the system off before it was released. No one wanted it to be out there but Sega. Fortunately for them, they didn’t alienate their Japanese user-base with an American-based console or the DC wouldn’t have survived as long as it did.

Sony’s hype is king. Never underestimate their hype.

As for 3dfx, they killed themselves when they decided to manufacture and sell their cards at retail. You simply cannot design the silicon, build the prototypes, manufacture the cards and then market and sell them at retail. It’s why NVIDIA is still here and 3dfx isn’t. There’s too much money spent for too little return and 3dfx didn’t have enough OEM deals to help them survive. They’ve only got themselves to blame for the company’s failure. Supplying for Dreamcast may not have helped. Look at Microsoft and NVIDIA wrangling over Xbox. I’m betting that’s one area NVIDIA isn’t so happy they’re involved.


In fact, I think you’re overestimating the effect of the 3dfx chip’s potential to make the DC a success.

I’m not attributing 3Dfx’s technological absence as leading directly to failure; the primary impact, assuming EA wasn’t full of it, of the lack of 3Dfx was that EA didn’t support the console. And the lack of EA, without a doubt, severely handicapped the Dreamcast in the U.S. The other aspects of it not being there are just food for interesting (to me, anyway) speculation on how things would have evolved.

I heard other journalists writing the system off before it was released.

Journalist opinions have very little real impact on momentum in that market. They do well just to keep up with what’s going on, and predictions and assumptions are wrong as often as right. And many are completely full of it :p to absolutely nobody’s surprise. People take credit when they’re right, and make excuses when they’re wrong. As for developers, they tend to develop where the market goes… prior speculation doesn’t matter alot unless they get the chance to back up desires due to market realities. Plenty were pining for the Xbox and looking forward to ditching PS2 development back before Xbox came out. That hasn’t worked out that way either. But going back to the DC, I seem to remember alot more excitement around the product when it was to be 3Dfx-based.

As for 3dfx, they killed themselves when they decided to manufacture and sell their cards at retail.
Sort of missing the point. 3Dfx did alot wrong, buying STB being one of the biggest. The point I was making is that they had no backup, where they could have. A steady income can buy you out of alot of failures, and buy you plenty of time to keep trying - just ask Microsoft. It’s not that they got themselves into a bind; it’s that, once there, they had no way out, and not even enough hope for future revenue to encourage any financial friends. Of course, again, this matters alot more for the survival of 3Dfx in the PC biz if they’re sitting on a winning hand with Rampage, which I just don’t know. But even then the other aspects make for some fun speculation :)

Nvidia makes a nice bit of change off those Xbox parts. Yes, they had to write off 10 million (?) due to a design change, and they’re probably getting annoyed with MS trying to hardball them into lowering their cost on the chips, but I doubt they’re ready to ditch the deal.



Hey Ciparis, why don’t you talk about sh*t that coo.

I actually find this discussion interesting, though I’m not really qualified to add much to it.

Yes, this is interesting. My first Voodoo card was the best purchase I had ever made for my computer at that time, it was so revolutionary in my eyes. As soon as I heard 3dfx’s plans to buy STB and make their own cards I knew they were doomed though, I still don’t understand their thinking. They had Diamond, Creative and others doing have their marketing for them and putting their product out in the field. That decision doomed them. :(

“Sort of missing the point. 3Dfx did alot wrong, buying STB being one of the biggest. The point I was making is that they had no backup, where they could have.”

Thats assuming if the DC had gone with a 3dfx chip, that EA would have supported the DC. I don’t know if thats the sole reason they didn’t do it. Then even with EA’s support I don’t think the DC would have survived the PS2’s onslaught. COnsidering how much money 3dfx was loosing at the end I doubt a deal with Sega would have saved them. They lost something like $700 million in their last year which was huge for a company of 3dfx’s size(they never were that big). Their incompetent management did them in.

As Jason says…the DC would not have been 3dfx’ savior. They were already going down the wrong road with the STB purchase and moving into retail. That was foolishness of the first degree. If they had gone into software or something at the same time, maybe they could have done it.

There was no guarantee that EA would support the DC. They didn’t like Sega hardware after Genesis because they felt they were competing with them on everything. Now they’ve got it even tougher…sort of… I’m not convinced that NFL2K3 is stealing all that much thunder from EA’s Madden. People would buy the new Madden if you opened it up and found the turkey bones inside from Madden’s last Thanksgiving…

Anyway, I think you’re overstating some things. EA didn’t have as huge an effect on the DC as you’d like to think. The lack of money in Sega’s warchest was their biggest problem. When they spent 50 million to market, Sony was spending hundreds of millions. The games spoke for themselves. On the average, games for Sega’s now-dead consoles are worth more than games on any competing device from the same generation. They’re sought after because they were and still are great. Sure there’s less of them due to premature system death, but they wouldn’t be as collectible if they sucked.


Good points, all :)

As Jason says…the DC would not have been 3dfx’ savior.

My speculation is that it (direct revenue and reduced Glide attrition) might have been enough to let them get Rampage out the door, either through directly bringing them enough revenue, or at least showing revenue so that they could attract an investor to help get Rampage to market, all of course assuming Rampage was actually close. And assuming Rampage was going to be successful - I’m actually asking the questions as much as speculating here.

It’s the weakest link (as far as being speculative) in the entire chain of events, and the most whispered about. Anyone know the answers?

Then and only then does the chain of events link to an interesting outcome for 3Dfx, regaining relevance and actually using some of that manufacturing ability that was, to that point, doing little for them.