![]() There are many different alternatives I just haven't bothered looking them up yet. There are many alternatives to silicon, I recently learned of Gallium Nitride, and Techquickie just made a video on how it is being used to revolutionize phone chargers. (Not an electrical engineer, just what I learned from watching the Upscaled series on engadget, personally the only thing engadget is worth for me.)Įlectrons are small, and we are expecting just 35 atoms (7nm) to stop them, which it still does due to the wonders of modern silicon foundries, but the smallest possible is a single silicon atom, at which point electron skipping the transistor (called Quantum tunneling) is expected. Two of the main reasons we go down in lithography is to pack more transistors into a small die space and to reduce heat, because the length of a wire influences electrical resistance, and even though these wires are incredibly microscopic, there are kilometers of microscopic wires in a processor, and that can create a lot of extra heat bc one of the byproducts of electrical resistance is energy in the form of heat. However, quantum is likely not the future of consumer electronics. This can ruin the functionality of a microprocessor, and as we get smaller and smaller the chances of that happening get bigger. Basically what was supposed to be a 0 in a logic operation became a 1 bc an electron successfully phased through a transistor meant to block it. To my knowledge 2nm is probably the lowest silicon transistor length foundries are currently aiming for. As we get smaller the electrons can literally phase through transistor gates leading to false positives. After that, computing is going to plateau for a bit, I bet.Ībsolutely, I totally expect node advancements of silicon to stop at 1 or 2nm.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |