Does Moore's Law still apply ?

chunter

Just wondering if Gordon E Moore's 1965 paper still roughly applies

http://en.wikipedia.org/wiki/Moore's_law

I haven't heard it mentioned in a few years.
Wikipedia has a graph but it stops in 2004

isofa

I can see no reason why it shouldn't still apply, especially if you look at the current processors with their predecessors.

BillScarab

I've read articles recently that suggested it still applies, although now multi-core CPUs are common it is slightly different as the number of transistors is split over more than one CPU.

Intel certainly seem to think they're keeping up http://www.intel.com/technology/mooreslaw/

esuhl

Everyone always says "Moore's Law can't apply forever", but so far it seems to have. Every time they hit a supposed "brick wall" in development, a new solution to allow the transistor count to increase has been found.

TonyLisaP

On the flipside of the question, does Murphy's law still apply?
If that does then so does Moores law.

fwor

The wiki that the OP points to includes some fairly strong indicators that it can continue for a while, but there are several factors that will make it very hard to continue for more than perhaps 20 years, some of which are:

the size of the structures is limited by the wavelength of light used in photolithography - if you go to shorter wavelengths, the light is harder to focus

the size of the structures is limited by the size of the atoms of material they are made of

the heat dissipation of a CPU has turned out to be higher as clock speed increases than expected.

While the wiki quotes several techniques (multiple cores, nanothechnology and quantum effects, perhaps), it seems there are some very difficult problems to be solved.

The wiki does also highlight another effect of increasing processor power: as the hardware gets more powerful, the software writers can get away with putting more "bloat" in their programs, which therefore need more processor power to run....