This presentation is great in a historical context as well as still being technically relevant.
The mid-90's is the era that saw a broad transition in game programming from at-the-metal, low-level coding towards more algorithmically driven approaches, since the hardware had become beefy enough to allow the kinds of large datasets where algorithms matter more than constant factors. Abrash notes it himself - his VGA/DOS knowledge became meaningless within the span of 2 years.
One thing that is overlooked in all the emphasis on 3D rendering is that real-time strategy also developed in the same era as the first-person shooter. This isn't just because of mice, but because it was now feasible to do fast real-time pathfinding in games. Earlier, similar attempts at the genre suffer from playing sluggishly or avoiding any pathfinding.
Another 90's genre that came about because of hardware improvements is music games: CDs and memory for sample playback made it possible to play back high-quality audio tracks.
Comments
This presentation is great in a historical context as well as still being technically relevant.
The mid-90's is the era that saw a broad transition in game programming from at-the-metal, low-level coding towards more algorithmically driven approaches, since the hardware had become beefy enough to allow the kinds of large datasets where algorithms matter more than constant factors. Abrash notes it himself - his VGA/DOS knowledge became meaningless within the span of 2 years.
One thing that is overlooked in all the emphasis on 3D rendering is that real-time strategy also developed in the same era as the first-person shooter. This isn't just because of mice, but because it was now feasible to do fast real-time pathfinding in games. Earlier, similar attempts at the genre suffer from playing sluggishly or avoiding any pathfinding.
Another 90's genre that came about because of hardware improvements is music games: CDs and memory for sample playback made it possible to play back high-quality audio tracks.