There comes a point where they become too small to function and start interfering with each other. They need 3 or 0.3 nanometers of distance apart I believe.
Moore's Law states that the number of transistors on a microchip doubles about every two years, though the cost of computers is halved.
What Jensen was referring to is the second part of this, not the first part. Specifically this "though the cost of computers is halved". Due to the current global constraints, this is not currently true and we don't really know how long this will last.
Plus, we'll get to the point where we can't make things any physically smaller, and we'll get to the point where we can't reduce the voltage any further to contain heat issues, so then it will just be cloning, which we see extensive use of already. But does that really count as Moores Law? Multiple iterations of the same thing isn't really an increase is it? Well, in the practical sense it is, but not in the scientific breakthrough sense.
I wonder if moores law will be adapted to quantum computing?
As others have mention, we are nearing the current material limitations.
There comes a point where they become too small to function and start interfering with each other. They need 3 or 0.3 nanometers of distance apart I believe.
Yup, and that's what 3D chips are supposed to at least be a temporary solution to
Ah, interesting.
What Jensen was referring to is the second part of this, not the first part. Specifically this "though the cost of computers is halved". Due to the current global constraints, this is not currently true and we don't really know how long this will last.
The cost to uphold Moore's law also goes up over time, so that's another constraint.
True, but innovation could solve this. Or rather, it has to or else Moore's law will actually die one day
Plus, we'll get to the point where we can't make things any physically smaller, and we'll get to the point where we can't reduce the voltage any further to contain heat issues, so then it will just be cloning, which we see extensive use of already. But does that really count as Moores Law? Multiple iterations of the same thing isn't really an increase is it? Well, in the practical sense it is, but not in the scientific breakthrough sense.
doesnt moores law specifically say transistor count doubles every 2 years?