I can't think of anyone more qualified to answer, so I'd like to ask John Carmack:
How do you anticipate 3D computer graphics techniques and hardware will change in the future? How would you like them to change?
If I may elaborate:
Are you happy with continued evolution of current technology (faster fill-rates, more effects, faster triangle rates, more acceleration of the graphics pipeline), or should there be a change of focus? Are other approaches than texture-mapped polygons (eg. voxels) valuable, or potentially valuable? Should they be hardware accelerated?
Could (eg.) parametric representations, fractals, procedural textures, etc. ever replace static model data in polygon-based engines, or only complement it?
Looking at current games (and the space they take up), I'd guess the work involved in building the virtual worlds in 3D games is becoming more intensive. Do you anticipate this becoming a serious problem, or are current modelling techniques scalable?
Will there be a use in games for mathematically modelled worlds (generated in real-time), rather than sampling the real world and/or using artists to create world data? (Tom Barbalet's work comes to mind.)
Slashdot Mirror
I can't think of anyone more qualified to answer, so I'd like to ask John Carmack:
How do you anticipate 3D computer graphics techniques and hardware will change in the future? How would you like them to change?
If I may elaborate:
Are you happy with continued evolution of current technology (faster fill-rates, more effects, faster triangle rates, more acceleration of the graphics pipeline), or should there be a change of focus? Are other approaches than texture-mapped polygons (eg. voxels) valuable, or potentially valuable? Should they be hardware accelerated?
Could (eg.) parametric representations, fractals, procedural textures, etc. ever replace static model data in polygon-based engines, or only complement it?
Looking at current games (and the space they take up), I'd guess the work involved in building the virtual worlds in 3D games is becoming more intensive. Do you anticipate this becoming a serious problem, or are current modelling techniques scalable?
Will there be a use in games for mathematically modelled worlds (generated in real-time), rather than sampling the real world and/or using artists to create world data? (Tom Barbalet's work comes to mind.)