Originally posted by: Arkaign
3dfx had a lot more successful products than just the V2. If the original Voodoo failed, V2 wouldn't have seen the light of day, let alone at the unheard-of prices they commanded. Voodoo3 fought evenly against TNT2 for the most part. It was STB and the too-late Voodoo4/5 that brought 3dfx down.
V3 was already dated and quickly relegated to the bargain bin because it failed to meet basic standards such as full colour and large textures. Also, it was not significantly faster but just finally integrated 2D
properly unlike their previous failures. So, they may have pushed a lot of product but it was not profitable -ergo, unsuccessful.
Indeed, the year of the V2 was the only profitable one in the company's entire history and even then they only broke-even on previous losses -which would be understandable given the nature of a start-up, however after the V2 they just continued losing money until going belly up.
It was not just too-late products but poor products -the designs were inefficient and thus too costly to make. With good product, the vertical integration would have increased margins. Heck, with good product they could have secured operating capital. Hence the notion that they had some killer product on the verge of being released which could have saved the company being utter nonsense.
So, in contrast to the pop notion, I posit that 3dfx would have failed sooner without the STB aquisition (no AIB's would have bought their poor product at any profitable price). But that, in addition to the prevalence (nay, glut) of Quake clones at the time which could run in "fast 'n' fugly" Glide mode allowed the company to linger.
Good riddance to it all, I say... and three cheers for standards and competition within them.