3 ms·

The article blames management failure, but I'm not convinced. There's a general bias to blame management any time a company fails. Of course management is alway

The article blames management failure, but I'm not convinced. There's a general bias to blame management any time a company fails. Of course management is always responsible, but I don’t think anyone could have saved say, Kodak. In this case, Nvidia and ATI were already making 3D accelerators for the professional 3D rendering market. All they had to do was to make stripped down cards that wouldn't cannibalize their high margin business, and sell them in volume for a price gamers could bear. Maybe 3Dfx could have made a moat out of GLide, but OpenGL already existed and Microsoft was working on Direct3D.

chrisan 9y ago
> but I don’t think anyone could have saved say, Kodak Why not? What if Kodak management had went hard on digital initially instead of clinging to film for so long?
- throwawayjava 9y ago
  Then they would've softened their blow and perhaps delayed their relative demise until the rise of smartphones. But I do think it's fair to say that the transition was going to be a bloodbath of layoffs and declining revenue in any case.
  - nwah1 9y ago
    Someone has to make the cameras in the phones. That plausibly could've been Kodak, in an alternate reality.
    - throwawayjava 9y ago
      There still would've been a bloodbath because chemical engineers aren't EEs/SEs. All the Kodak scientists/engineers/factory workers who were designing and making film would not have immediately pivoted to digital camera design and software development. The "camera design" portion of kodak isn't where the primary bloodbath happened. The bloodbath happened when film died. In other words, kodak the company may have thrived, but only by shedding huge portions of its workforce and retooling. I guess if all you care about is stock value, then sure, Kodak could've pulled it off. But for the folks on the ground, the bloodbath would've happened either way.
- TorKlingberg 9y ago
  Kodak was never an electronics company, or even a camera company. Nikon and Canon would have eaten their lunch.
  - TheOtherHobbes 9y ago
    Kodak literally invented the digital camera, and were significant in both consumer and professional digital photography in the early 00s. They missed by not innovating fast enough compared to the competition, not by ignoring the market. Unimaginative management was very definitely a problem - although to be fair, it's easy with hindsight to say they should have started developing digital much earlier, when in reality consumer digital photography only started to make commercial sense when computers and public networking became powerful enough to support it. That didn't happen until the later 90s. Until then, most computers lacked the photo quality displays we take for granted now.
    - djsumdog 9y ago
      I can remember my oldest digital camera was a Kodak my parents bought me, with a CF card in it I believe.
  - mulmen 9y ago
    What kind of company is 3M? Or IBM? Or GE? Management is responsible for the strategy and long term success of a business. Pivots are necessary and a core part of successful strategies.
- MarkMc 9y ago
  I see it the other way. What if Kodak hadn't wasted money trying to enter businesses like digital cameras and printers - ie. low-margin products where Kodak had no competitive advantage. Sure, Kodak would have still gone out of business but it wouldn't have wasted so much shareholder capital in the process.
- Zardoz84 9y ago
  interesting... my first digital camera was a Kodak. Only enough internal flash storage (no removable) that can hold around 75 JPEG photos at 640x480 . And without screen. How ever no body had a digital camera except me, for a few years. I kept it
adamroderick 9y ago
Good leaders take responsibility
pvg 9y ago
Nvidia and ATI were already making 3D accelerators for the professional 3D rendering market. Were they? I don’t see any professional cards in Nvidia’s early line-up, at least not the one shown on Wikipedia. The first Quadro doesn’t show up till 1999 and the verbiage suggests the pro cards were an offshoot of the gaming cards, not the other way round. https://en.m.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units https://en.m.wikipedia.org/wiki/List_of_Nvidia_graphics_proc...
- TorKlingberg 9y ago
  It looks like I was wrong. I was going off memory.
djsumdog 9y ago
I'm glad the article goes into STB. I really think this was what killed 3Dfx. It was a big move, but if you were following 3Dfx back the, you saw the issues. I think they tried to lay off duplicate staff. With the Voodoo3, you saw drivers that just didn't work half the time. Textures were rendered totally wrong, people would downgrade drivers to get things working, etc. I have a felling they laid off or lost people critical in the hardware itself. The website went to shit too. It got terrible. They either axed or lost a ton of web staff and focused more on the graphics used on all their box art. Finally, the cards/chips just didn't scale well. I even remember a ton of articles at the time talking about how much efficiency was lost with the multi-chip design; how many things had to be duplicated in memory for the different GPUs. The Voodoo5 even required an external plugin power source. nVidia to this day still sells its chips to everyone else. 3Dfx was trying to get into the ATi space, where you only had one ATi card and didn't have to worry about reference vs vendor drivers. It might have worked if they tried to keep both units separate for a while and not merge them into one entity.

chrisan 9y ago

> but I don’t think anyone could have saved say, Kodak Why not? What if Kodak management had went hard on digital initially instead of clinging to film for so long?

throwawayjava 9y ago
Then they would've softened their blow and perhaps delayed their relative demise until the rise of smartphones. But I do think it's fair to say that the transition was going to be a bloodbath of layoffs and declining revenue in any case.
- nwah1 9y ago
  Someone has to make the cameras in the phones. That plausibly could've been Kodak, in an alternate reality.
  - throwawayjava 9y ago
    There still would've been a bloodbath because chemical engineers aren't EEs/SEs. All the Kodak scientists/engineers/factory workers who were designing and making film would not have immediately pivoted to digital camera design and software development. The "camera design" portion of kodak isn't where the primary bloodbath happened. The bloodbath happened when film died. In other words, kodak the company may have thrived, but only by shedding huge portions of its workforce and retooling. I guess if all you care about is stock value, then sure, Kodak could've pulled it off. But for the folks on the ground, the bloodbath would've happened either way.
TorKlingberg 9y ago
Kodak was never an electronics company, or even a camera company. Nikon and Canon would have eaten their lunch.
- TheOtherHobbes 9y ago
  Kodak literally invented the digital camera, and were significant in both consumer and professional digital photography in the early 00s. They missed by not innovating fast enough compared to the competition, not by ignoring the market. Unimaginative management was very definitely a problem - although to be fair, it's easy with hindsight to say they should have started developing digital much earlier, when in reality consumer digital photography only started to make commercial sense when computers and public networking became powerful enough to support it. That didn't happen until the later 90s. Until then, most computers lacked the photo quality displays we take for granted now.
  - djsumdog 9y ago
    I can remember my oldest digital camera was a Kodak my parents bought me, with a CF card in it I believe.
- mulmen 9y ago
  What kind of company is 3M? Or IBM? Or GE? Management is responsible for the strategy and long term success of a business. Pivots are necessary and a core part of successful strategies.
MarkMc 9y ago
I see it the other way. What if Kodak hadn't wasted money trying to enter businesses like digital cameras and printers - ie. low-margin products where Kodak had no competitive advantage. Sure, Kodak would have still gone out of business but it wouldn't have wasted so much shareholder capital in the process.
Zardoz84 9y ago
interesting... my first digital camera was a Kodak. Only enough internal flash storage (no removable) that can hold around 75 JPEG photos at 640x480 . And without screen. How ever no body had a digital camera except me, for a few years. I kept it

adamroderick 9y ago

Good leaders take responsibility

pvg 9y ago

Nvidia and ATI were already making 3D accelerators for the professional 3D rendering market. Were they? I don’t see any professional cards in Nvidia’s early line-up, at least not the one shown on Wikipedia. The first Quadro doesn’t show up till 1999 and the verbiage suggests the pro cards were an offshoot of the gaming cards, not the other way round. https://en.m.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units https://en.m.wikipedia.org/wiki/List_of_Nvidia_graphics_proc...

TorKlingberg 9y ago
It looks like I was wrong. I was going off memory.

djsumdog 9y ago

I'm glad the article goes into STB. I really think this was what killed 3Dfx. It was a big move, but if you were following 3Dfx back the, you saw the issues. I think they tried to lay off duplicate staff. With the Voodoo3, you saw drivers that just didn't work half the time. Textures were rendered totally wrong, people would downgrade drivers to get things working, etc. I have a felling they laid off or lost people critical in the hardware itself. The website went to shit too. It got terrible. They either axed or lost a ton of web staff and focused more on the graphics used on all their box art. Finally, the cards/chips just didn't scale well. I even remember a ton of articles at the time talking about how much efficiency was lost with the multi-chip design; how many things had to be duplicated in memory for the different GPUs. The Voodoo5 even required an external plugin power source. nVidia to this day still sells its chips to everyone else. 3Dfx was trying to get into the ATi space, where you only had one ATi card and didn't have to worry about reference vs vendor drivers. It might have worked if they tried to keep both units separate for a while and not merge them into one entity.