Where do you think the bottlenecks (limitations or barriers) to the growth of the computer, in terms of its computational power and its abilities/application, lie?

Question

What will be an ideal response?

Answer 1

This is partially answered by other questions on, for example, Moore’s Law.
Limitations depend to some extent on the nature of the user. This answer is largely aimed at the home or office computer user, rather than the scientific or government user, because scientific/government users can spend their way out of bottlenecks (e.g., by means of massive parallelism).

In terms of data storage (capacity) we are doing well. The greatest storage requirement comes from high definition moving images. At the moment personal computers and tablets can hold a reasonable amount of data; for example, hundreds of hours of video. As the resolution of displays increases, there will be a continued need for more storage capacity (and some will wish to carry their entire video storage with them).

Processing power is not usually an issue in desktop computing. However, those working on very high resolution still and moving images continue to require more power. The same is true for high?resolution dynamic games. High?performance games computing will probably be a driving force for years to come.

AI applications will continue to soak up computing power as fast as it can be generated for the foreseeable future. For example, face recognition could be used to compile databases of all scenes and actions across many movies, a gigantic task.

A practical limitation to computing is bandwidth, either for a fixed computer (e.g., via cable modems) or for mobile computers (via WiFi). It may well be that the lack of bandwidth will prove to be the single most limiting factor for the average user.

Another limitation is power. This manifests itself by limiting computing speed (the need for more energy) and displays (the need to provide backlighting). Another power limitation is dissipation in terms of the need to emove heat from a computer. This is a particular problem in mobile computing. Finally, when the public electricity supply is not available (e.g., mobile computing), the principal limitation is battery life.

In non?home and office computing (commercial, military, government, medical) it is hard to see any limit to the demand for both storage capacity and computing power. Medical imaging alone has increasingly larger and larger storage requirements. Simulation also has endless requirements from the simulation of nuclear explosions, to the simulation of weather systems, to the simulation of molecular processes in physics. There is simply no foreseeable limit to computational requirements.