Microsoft and OpenGL

FS: You first gained notoriety in the gaming industry and community due to the infamous Direct3D vs. OpenGL debates, starting with John Carmack's decision to eschew D3D support for Quake-engine games. How did you see this affecting the industry, and how did it affect Direct3D development in general?

Alex: I think that one of the big things DirectX did for the industry was to stabilize the multimedia industry. There was big pressure from Microsoft for hardware manufacturers to create stable, reliable drivers, or else MS wouldn't ship them with the OS. This pressure didn't exist for OpenGL drivers which meant they were rarely as stable as DirectX ones.

FS: You're talking about WHQL?

Alex: That's correct, Windows Hardware Quality Lab certification. The customer needs something that works. If you look at it, support issues for OpenGL were much higher because there is no standardizing group to enforce OpenGL driver quality.

As a controlling force, Microsoft hated OpenGL - no matter how good it was, Microsoft wasn't going to support it. In that way, there was very little we could do to get around the issue. The company's continuing philosophy was "Oh, it's a competing API, we have to kill it!" The reason of course, was that OpenGL was not a Microsoft standard.

Well, one of the things I have to stress is that the DirectX development team did not and never did have any animosity or bias against the OpenGL API. When the DX effort started, there was a big internal competition between the Win95 and NT teams. In fact, the Windows NT design team who controlled Microsoft's implementation of OpenGL didn't allow the Windows 95 group to put OpenGL support in Windows 95! Their argument was that "Win95 will be dead next year when everyone moves to NT," so of course there would be no point in adding support because Win95 would be gone. For that reason, we (we being myself and the other folks working on the early DirectX) looked to acquire and adapt the Rendermorphics engine to become Direct3D instead.

An interesting point to note was that Michael Abrash was on the MS Windows NT team at that time. Most people don't know that id supported D3D right up to the point that Abrash left Microsoft to join them.

FS: Where do you see 3D Web media heading in the future? Will it replace standalone programs with a more seamless online experience?

Here's a relevant example of what I think. How different does the web look from the standalone apps on your PC? From a graphical and functional point of view, it's not quite there yet. 7 years ago, all CD-ROM and multimedia content was developed with Macromedia Director, and if you look at it, it really resembles how the web looks now. As technology advanced, more bandwidth became available, and DirectX took hold, multimedia applications changed drastically, and now we have "fully 3D" immersive experiences - Macromedia was in big trouble.

However, take a look at the Internet. Macromedia was able to adapt its tools for a medium that is currently at the stages of where multimedia was several years ago - the bandwidth constraints make tools like Shockwave and Flash ideal for the net. But a few years down the road, that will all change, expanding resources will allow the web to grow past these constraints.