(ORDO NEWS) — Scientists proposed to create environmentally friendly components from honey for neuromorphic computers – systems designed to mimic neurons and synapses in the human brain.
Computing devices that mimic the human brain are just around the corner. But who would have thought that they can be made from honey!
Many scientists believe that neuromorphic systems are the future of computing. They are much faster than traditional computers and consume much less power. However, so far, researchers have not succeeded in creating a reliable nephromorphic computer.
Several companies, including Intel and IBM, have released neuromorphic chips that contain the equivalent of over 100 million “neurons,” but that’s still not as many as in the brain.
Many companies are also still using the same non-renewable and toxic materials currently used in conventional computer chips. Something must be done about this.
How to make a neuromorphic chip from honey
Many researchers, including a team at the University of Washington, are looking for biodegradable and renewable solutions for this new type of computing device.
In this capacity, scientists are testing the use of proteins and carbohydrates, but they see the greatest potential in honey. It’s all about the low water content, which prevents bacteria from multiplying, which makes honey very durable.
The physicists in the new work created memristors by converting honey into a solid form and holding it between two metal electrodes.
The result is a structure resembling a human synapse. They then tested the ability of honey memristors to mimic synapses with high turn-on and turn-off speeds of 100 and 500 nanoseconds, respectively.
Memristors also emulated synapse functions known as burst time-dependent plasticity and burst velocity-dependent plasticity.
They are responsible for the learning processes in the human brain and the storage of new information in neurons.
Now scientists plan to develop nano versions of these elements, about 1/1000 the size of a human hair, and combine many millions or even billions of memristors together to create a complete neuromorphic computing system.
Contact us: [email protected]