Is Moore’s Law dead at 40 or is this just a mid-life crisis?

ZD Net ^ | 4/5/2005 | George Ou

[infocats]

Last week, Michael Kanellos published this FAQ on the 40th anniversary of Moore’s law, which is famously known as the phenomenon that computer processing power will double every 18 months. Actually, Gordon Moore only said that transistor count would double every 24 months and it was David House (a former executive of Intel) who extrapolated that performance would double every 18 months as a result of the increase in transistors. Ironically, it is House’s unofficial reinterpretation of Moore’s law that has become the popular definition of Moore’s law.

Over two years ago, Tomshardware released this excellent article showing the historical progress of Intel and AMD CPUs from 100 MHz to 3000 MHz from year 1994 to 2003. The results were astonishingly true to the 18-month performance doubling cycle, which reminded me when name-brand 33 MHz 486 computers were routinely sold for nearly $10,000 back in the early 1990s. Because of this, I had been conditioned to the point that if someone had told me two years ago that we would still be stuck with 3 GHz class CPUs today, I would have told them that they were crazy. It’s a good thing I never made a wager on that theory since what has happened since February 2003 seems to signal that Moore’s law is dead at the age of 40, or at least going through a serious mid-life crisis.

[Army Air Corps]

Have a look at this http://www.texastech.edu/news/CurrentNews/display_article.php?id=1685

[RockinRight]

I think processor speed will continue to increase but more slowly. We may have reached a point where the usefulness of faster CPU's is less and less of a factor.

[ShadowAce]

[infocats]

"We have shown that Peirce's theory of signs can be applied directly to elementary particle interactions," said Beil, the senior author of the patent. "An aspect of our patent is that elementary particles such as photons and electrons can be used as carriers and processors of information. This is also proposed in previous designs; however, those designs involve multiparticle or parallel states with two (binary) values. Our designs involve single particles or sequential states with possibly more than two values each."

Thanks for the heads up. Moore might have been on to something...even with his mid-life crisis.

[SampleMan]

Nothing can continuing doubling forever.

In the case of computer speed, I'm surprised that the market supported the growth curve of speed for as long as it has.

Follow the money. No one (99.9% of computer users) needs a 1 GHz processor. After the initial introduction of the automobile, it was all about speed. Every decade saw an increase of speed and the development of better roads to handle the speed. However, most driving is done in the city at speeds below 50mph, and extremely few people would drive faster than 85 mph on the highway, even if they could. For the case of argument, no one is willing to pay twice as much for a car that will go 300 mph, to haul the kids to school in.

Generalizations, which I'm sure some of you will find necessary to shoot at with specific examples, but I'd say the exceptions prove the rule.

When there is no longer a payoff for pushing faster processing, people stop getting paid to develop it.

[Mr. Jeeves]

We may have reached a point where the usefulness of faster CPU's is less and less of a factor.

It think it's because 3GHz CPU's are so fast they've outrun the ability of software vendors to put out bloated code that will bog them down. Once the software vendors catch up, CPU speeds will trend up again. ;)

[marvlus]

I think I would drive drive a computer at 10 times normal speed than I would a car.

[marvlus]

The problem isn't so much the speed of the CPU as it is the slowness of all the peripherals the CPU has to manage and work with.

[ExtremeUnction]

When there is no longer a payoff for pushing faster processing, people stop getting paid to develop it.

We've used the faster speeds to plow through the massive amount of code found in modern applications..

[OldEagle]

With the limited functions of PC's, 3 Gigs is a good juncture to wait and see where it is going. Right now the attention is on cell phones and PDAish items.

[infocats]

I think processor speed will continue to increase but more slowly. We may have reached a point where the usefulness of faster CPU's is less and less of a factor.

Everything seems to reach a point of diminshing returns, even the human brain. I seem to recall reading that at one point in humankind's evolution (pardon me while I don my flame retardant jumpsuit), the head had to stop increasing in size to accomodate an ever larger brain because of difficulty in passing through the uterine tract, and the brain itself had to start folding in more complex patterns to accomodate more complex thought processes (note that this in no way precludes intelligent design.

[Uncle Fud]

Nobody but gamers are buying the bleeding edge PC's now, and even some of them are starting to be deterred by the cooling requirements.

When the only way to keep my PC from melting is to have fans like 767 engines or alternatively a water cooling system, I decide not to upgrade.

I currently have a year-old 2 ghz AMD chip and see it meeting all my needs for some time to come.

[AFPhys]

Once processors reached the speed that they could support graphical interfaces easily, necessary so the bulk of people find computer technology accessible, the major reason for increased speed was fulfilled.

Far and away, the bulk of our present day processors' cycles are spent waiting for the next character typed, or the next couple of bits from the 'net. Very few people (outside the server business, or motion video processing) utilize even a tenth of the processing power of their 'puters - even in computation intensive environments. I was surprised the other day when I actually discovered one of our 'puters was running steadily at about 80% computation speed --- very, very rare case. Not even text-talking even creates great demands on these GHz machines.

We've finally reached a good machine speed for most people's use. There is little reason to do much more here (other than make the programs more affordable).

[doc30]

Who knows, the day may come when you can cook roadkill on your motherboard like certain people do on their car engines.

[fireforeffect]

The human eye and ear can only process information at certain rate. Fast enough is however fast the chip needs to be to just exceed this rate.

For generic data processing (spreadsheets, accounting, engineering) the current CPU's with the current applications are fast enough. For now.

The bottleneck is on the motherboard chip sets. Current processors are faster than the bus on the motherboard. Thus the rise in Front/Back/Memory/Graphic/Audio Bus speeds. For gamers the current bottleneck is graphic rewrite. Graphics cards are getting faster, but the path between the CPU and the graphics card has only recently speeded up.

Another problem with motherboards has been the trend to make them backward compatible. People like to continue to use old modem, sound, and other cards that are paid for. The newest hot boards do not have this capability.

Another bottle neck is heat management. The faster you cram electrons through the chips the hotter they get.

[MediaMole]

Computer-fried egg sandwich...

http://www.phys.ncku.edu.tw/~htsu/humor/fry_egg.html

[Army Air Corps]

No joke. Did you read the part that described this switch operating in a TRIANRY fashion? That blew me away.

[Army Air Corps]

trianry=trinary