Just as when we talk about software we refer mainly to programs and with hardware we refer to the components that allow these applications to run, there are other terms. Specifically, the one we will deal with now is the wetwarewhich is linked to the use of hardware and software components applied to biological life forms, specifically related to ethe brain and the neural networks for AIsomething we found in a new study by water based chips.
Chip researchers are experimenting with new methods to broaden their use. Also, they are involved in new processes where they are able to desEsports Extrasprocessors that differ from the current ones, in order to be able to be better in other fields. Apparently, in the field of neural network computing, the concept of wetware, which became popular in the late 1980s, might actually be possible in reality. So, let’s leave the dystopian universes of cyberpunk and see what this new water-based chip technology is all about.
Harvard Researchers DesEsports ExtrasIonic CPUs: Wetware Made Real
In the 1980s, there were a number of novels in the “biopunk” genre that tried to combine hardware with software in the human body. Thus, this combination of technology and biology gave rise to a series of symbiotes with software implemented in their bodies. This concept reminds us of what some researchers from the Harvard School of Engineering and Applied Sciences (SEAS) have achieved. These, together with the biotechnology company DNA Scripthave developed an ionic circuit composed of hundreds of ion transistorswhich can process neural network calculations.
To carry out the desEsports Extrasof these, research has been carried out on the functioning of the brain and for this an alternative to electricity has been sought. That is why they use the electrochemistry to perform the calculations, ensuring a very low consumption. While ion transistors are not new, it is the first time that one is created circuit board containing hundreds of these transistors.
These chips capable of doing useful calculation tasks for AI
This type of ionic transistors are composed of an aqueous solution of quinone molecules, interconnected withn two concentric ring electrodes with a third center electrode. Once the voltage is passed through the transistor, the two ring electrodes are able to adjust the pH of the water changing the amount of hydrogen ions present. Thanks to this ionic current, the transistor works as a Switch on and off.
With this basic idea, the researchers put this processor to the test by placing analog transistors in a 16×16 matrix. This water-based chip was able to perform calculations of matrix multiplication, thus being useful in AI tasks that require negligible energy consumption, thanks to electrochemistry. Obviously, these chips cannot compete in power electricity and conventional processors, but the idea is that they can be useful in very specific scenarios.