Just wanted to jot down a quick note concerning my impressions and predictions about the recent announcement of the merger of 3DFX (the makers of the Voodoo2 chipset) and STB (videocard designers). Jumping right in, I think we all can see that there are two primary reasons behind this merger:

• Entry into the OEM business. 3DFX has been hinting for several months that they consider the OEM marketplace a vast, untapped resource, especially after they had essentially stagnated the retail market with their Voodoo2. Unfortunately for them, making headway into this market proved all but impossible for them. STB is a big OEM player, so they think they can use the name to quickly and easily step into the ballpark.



• Increased profitability from the gamers/upgraders market. Not card makers themselves, 3DFX has been at the mercy of third-party card designers (primarily Diamond and Creative Labs) for their sales. While these two companies definitely had the power and pull to make 3DFX practically a household name, there were also downsides to this partnership. Diamond and CL aren't buddies, but are competitors in a fierce marketplace. With identical Voodoo products, the battlefield was solely based on price. And, of course, 3DFX only made chipset profit, not whole-card profit.

What I see here are two conflicting goals. On one hand, 3DFX wants to make 1 brand of 3DFX card and sell it for $300, making mucho profit. On the other hand, they desperately want those Dell and Gateway OEM contracts, and we both know that OEM is decided almost entirely on low cost. So, what does 3DFX do? Do they strive to make very good (and very expensive) upgrader's cards in which hardcore gamers will gladly pay $300? Or do they focus on maintaining low unit cost (and inevitably, lower quality) to make the OEM crowd happy? IMO, these two paths are contradictory, and 3DFX is destined to fail at one or the other. I also think that in attempting to have it both ways, 3DFX will quite likely end up failing at both.

Despite what actions 3DFX takes and does/does not manage to accomplish, there are repercussions of this merger outside of themselves. First, GLide as a 3D programming language is dead and buried as of the merger date. What game developer in their right mind would release a future game that will only run on STB cards? 3DFX will now be forced to focus on quality openGL and D3D implementation, and we all know how spotty that has been, historically. Second, its obvious that CL and Diamond will start courting nvidia in order to fill the 3D chipset vacuum they were left with, and I'm sure nvidia couldn't be happier. The monstrous advertising departments of these two companies will launch a campaign blitz, letting the gamers know that there is a new king of the hill - the TNT (and TNT2). Overnight, 3DFX lost two major supporters and gained two major competitors.