Computing Chip Trends: Memresistors


Moore’s Law is not a law but rather an astute observation that has stood the test of time – it has predicted the rate of change in computing power for the past 50 years with uncanny accuracy: Computing power will double in capacity for the same cost every 18-20 months. This “law” is now running into hard stops as chip line geometries reach below 35 nanometers – the size of a virus and at the extreme limits of the geometry drawing/etching equipment used to make chips.

This and other limitations have provided some real barriers in electronic chip-making. CPUs have now split out into dual and multicore discrete units. Memory chips are reaching limits of operational complexity and reliability. Flash memory cannot keep pace with the data bits. In sum, Moore’s Law has been in jeopardy of becoming a failed conjecture.

But Hewlett Packard may have changed that and come to the rescue of Moore’s Law. HP’s new memresistor not only changes the nature of chipmaking as 3rd dimensional layers and ever finer chip lines remove the size of chip barriers that have confronted silicon chip technology.But also memresistors provide new chip speed opportunities given new chip architecture and performance specs as well.

So memory chip capacity becomes less of an endangered electronic gadget species. But also memresistors may solve the battery drain problems because memresistors can be turned off and inherently draw less operational power. Last, but not least, memresistors means storage areas become computationally adept. So the same place you store data can also do elementary logical/computational steps. Suddenly basic chip calculations are within the realm of “memory” storage. So in effect whole new ways of designing computing architecture open up – and suddenly Moore’s law is looking pretty solid for the next decade or two. Not bad for an “observation” on the nature of innovation in computing.

Similar Posts