Our brain is like a uniquely powerful computer. It’s in a class of technology that no typical serial or parallel processor today can replicate. Many scientists have tried to develop computer code that attempts to mimic through simulations, such as the current Blue Brain Project, but the computing power for these approaches are becoming immense.
Alternatively, a large European Union collaboration called FACETS has been working on the design, fabrication, and implementation of a new kind of transistor-based computer chip that structurally mimics the neural networks of the human brain.
The goal of the project is to create a unique computing architecture that uses what we already know of the structure of the human brain as a foundational design concept. The anticipation is that by creating this new hardware, we might gain a significant advancement in computational technology that might keep us moving upward along the classical Moore’s Law path even after traditional transistor-based architectures reach their lowest physical size limit.
The current scale of these brain-like computer chips are far from the level of the interconnectivity of the human brain. At this time, they have developed chips with around 200,000 transistor-styled “neurons” utilizing 50 million mimicking “synapses”. This is a far cry from the human brain’s nearly 100 billion neurons and countless synapses.
This project is not necessarily trying to build a silicon brain… but is wisely trying to take the structural concept of the human brain and apply it to a new hard-wired approach to develop the next-next-next generation of desktop computers.