gumby_trucker said:
This is simply not true. Transistors have indeed continued to shrink more or less as Moore predicted, but that in no way means performance gains at the same pace. I'm no expert but I think there is a problem with increasing the switching speed of a transistor too much at these tiny scales as they become prone to "leaking" current simply due to Brownian motion. This I think is one of the reasons why so called "processing speeds" (amount of MHz/GHz) have barely gone up in the last decade. Check Wikipedia and you'll see Pentium 4 processors at speeds above 3 GHz have been on the market as far back as 2002. With the move to multi-threaded and later on multi core CPUs it becomes possible to perform more tasks in parallel so as far as the consumer is concerned they are getting more "bang" per MHz due to parallelization, but this scales much worse then you think it does. ie: twice the cores is in most practical cases much less than twice the performance on any given application. Some of this can be improved by fundamentally re-writing programs which were meant to run serially, but even this process takes many years and it's not always clear how much it benefits performance. |
http://www.intel.com/technology/mooreslaw/
That's Moore's law still aplying right there. Here's more:
"Moore’s law isn’t tracking exactly, but the spirit of the law is still alive in that the dies are still shrinking, and CPUs become more and more capable every 12-18 months or so," said Joel Santo Domingo, lead analyst, desktops at PCMag.com. His former boss agrees.
"I did the math, and while it’s not exactly doubling every two years, it’s pretty close," agreed Michael Miller, the award-winning math geek and former editor in chief of PCMag.com.
Read more: http://www.foxnews.com/scitech/2011/01/04/years-later-does-moores-law-hold-true/#ixzz1Mq8tUOy2
EDIT: So if that graph up there is straight then the growth is exponential then the law pretty much holds. It looks close enough to me.
