Is Moore's law broken again?

Hannover Fist

I bought a Dell PC last fall with about a 3.0 GHz processor. I see that Dell is still offering 3.2 GHz as their fastest processor, and memory seems really expensive? Does anyone have any insight into what's going on with processor development? I don't follow it anymore.

dacris

I don't know but I got my P4 2.6 GHz in December, 2002, and IIRC the highest CPU speed back then was 3.0 GHz. So if we've only advanced about 0.4 GHz in 1.5 years, then Moore's law has indeed been broken. Intel has some explaining to do... I have a symbiotic relationship with my computer.

Heath Stewart

There was a discussion on /. the other day about increasing memory prices. I honestly don't remember too much about it (mostly because it's still relatively low compared to several years ago), but if you're interested in the many conspiracy theories (which may be true), you chould check it out on /.[^]. There's also been several articles cross-posted on /. about Moore's broken law. There seems to be a general agreement that Moore's law is indeed in need of a revamp ( guess that means it's bumped down to a hypothesis :rolleyes: )

Microsoft MVP, Visual C# My Articles

Farhan Noor Qureshi

How about Montecito[^] with 1.7 billion transistors? Farhan Noor Qureshi

Chris Losinger

IIRC, Moore's law only applies to the number of transistors on a chip, not to the speed of a processor. Cleek | Jewelry | ClickPic | ThumbNailer

Joe Woodbury

From my observation, Moore's law [edit: at least how's it's commonly perceived--Chris above has a point] hasn't actually been accurate for the past twelve years; it appears that performance triples everything three years, rather than quadruples. Having said that, even Moore strenuously pointed out that his observation is a generalization over over time. In other words, saying transistor count doubles every 18 months, doesn't mean it increases by [about] 4.5% every month. So even my observation isn't necesarily accurate since during the late 80s and early 90s, technology was arguably changing at a faster rate than Moore had predicted. As for right now, you can get a 3.4 GHz extreme processor and the 3.6 and 3.8 should be out in the next six months. Intel's also releasing new MBs in June and PCI-X, or whatever it's called, is just around the corner. Whether systems released in spring 2005 will be twice as fast (or have twice as many transistors) as your Dell 3.0 GHz remains to be seen. Anyone who thinks he has a better idea of what's good for people than people do is a swine. - P.J. O'Rourke

Marc Clifton

Intel is shifting to a dual core processor, because something like 6Ghz it the top end you'll see in a single core. A dual core consumes less power, but each half runs at a lower frequency, thus to get the full benefits of say, a 4Ghz dual core, you'll need an OS/software that takes advantage of multithreading. There are other processes that I read about recently for chip manufacturing that seem promising, but I can't remember where or what. Memory expensive? Blame the weak US dollar and the rising cost of raw materials and energy. Even with that, the trade deficit was the highest ever last month. Marc Microsoft MVP, Visual C# MyXaml MyXaml Blog

David Cunningham

Buy a dual proc box and you're covered :-D David

ColinDavies

Yeah, I too have noticed that processor speeds have not been increasing a lot. But then with the current memory bottlenecks being experienced it's pointless for Intel etc to release faster CPUs. I think we are going to start seeing a big boost in cache sizes to compensate. Regardz Colin J Davies

*** WARNING *
This could be addictive
**The minion's version of "Catch :bob: "

It's a real shame that people as stupid as you can work out how to use a computer. said by Christian Graus in the Soapbox