The future of Moores law?

Some people are saying that the Moores law is slowly slowing down, and the transistor count will only double every 4 years or 8 etc.. Since Intel is not releasing any 4Ghz CPUs right now, how long will it take for a commercial CPU to be marketed at 5Ghz?

I dont think there will be any 5Ghz Chips for a While, I imagine that most manufacturers will go dual core totally. Instead choosing 2x3Ghz Cores or similiar.

if you look at procesor mhz, you will see that pentium 4 at 3.06 ghz was released in 2002 or 2003 (dont remember well) and now they are at 3.8 ghz...
in three years the mhz rating go up only for around 30%. with Amd is the same story...
thats why this two companies (Ibm too) will go with multi-core cpu...
i think thats better...we will see bigger increase in speed in next years (when most aplication/games will suport multi-core cpus)

i just remember, that i buyed my old pc, a celeron 400 in year 2000, then the best procesor was a pentium 3 at 733mhz, and in next two year that speed was at 2ghz and more...
that was bad times for costumers...(because u needeed to upgrade more to keep up with advancement)

Typically you would need to upgrade on a "regular" basis, but how many applications had benefited from this development. I think back to my Celeron 500 in 98 and what it could do, and i look at my 3.4 now and its hard to imagine the huge difference. Games in 2000 can now be played on a portable device, afterall a huge push for the development does come from gaming industry.

Let me rephrase my question, when i said Ghz i used it as a performance scale. Are the CPUs 4 years from now be able to quadriple the speed of handling tasks? Since Moores law specifies the transistor count, is that going to keep up its pace as the law states?

moore's law isn't a law at all. so many people think its in-fallable, like pythagoras law. its simply the observation of moore at the time, that he supposed the transistor count would double every couple of years. that is not a mathmatical formula or anything, just some basic observations way back in 1965.

so far, it seems to be holding up (sometimes its much more than double tho, sometimes less), although who knows for how long.

......................................Year of Introduction.......Transistors
4004.......................................1971...........................2,250
8008.......................................1972...........................2,500
8080.......................................1974...........................5,000
8086.......................................1978.........................29,000
286 .......................................1982........................120,000
Intel386™ processor..................1985........................275,000
Intel486™ processor..................1989......................1,180,000
Intel® Pentium® processor.........1993......................3,100,000
Intel® Pentium® II processor......1997......................7,500,000
Intel® Pentium® III processor.....1999....................24,000,000
Intel® Pentium® 4 processor......2000....................42,000,000
Intel® Itanium® processor.........2002...................220,000,000
Intel® Itanium® 2 processor...... 2003..................410,000,000

exactly. It's an observation. And it doesn't have anything to do with clockspeed. The fact that at some point, clock speeds were doubling at approx the same time as the transistor counts were was coincidence, not necessarily by design.

hUMANbEATbOX said:

moore's law isn't a law at all. so many people think its in-fallable, like pythagoras law. its simply the observation of moore at the time, that he supposed the transistor count would double every couple of years. that is not a mathmatical formula or anything, just some basic observations way back in 1965.

so far, it seems to be holding up (sometimes its much more than double tho, sometimes less), although who knows for how long.

......................................Year of Introduction.......Transistors
4004.......................................1971...........................2,250
8008.......................................1972...........................2,500
8080.......................................1974...........................5,000
8086.......................................1978.........................29,000
286 .......................................1982........................120,000
Intel386™ processor..................1985........................275,000
Intel486™ processor..................1989......................1,180,000
Intel® Pentium® processor.........1993......................3,100,000
Intel® Pentium® II processor......1997......................7,500,000
Intel® Pentium® III processor.....1999....................24,000,000
Intel® Pentium® 4 processor......2000....................42,000,000
Intel® Itanium® processor.........2002...................220,000,000
Intel® Itanium® 2 processor...... 2003..................410,000,000

Click to expand...

Lol interesting how you utilise the Itanium processor (from an entirely different segment and pricing structure) in your table to prove that Moores law still hold true. You were obviously using a later Moore's Law to make it fit!

This completely foregoes Moore original formulation back in 1965 which actually counted components not only transistors and also costs.

In his 1965 article, "Cramming more components onto integrated circuits," Moore discussed the future of integrated electronics. He observed that the object of miniaturization had been in putting increasingly complex electronic functions in limited space with minimum weight. In his 1965 paper, Moore mentioned the Apollo manned moon flight program as a demonstration that integrated circuits can be as reliable as individual transistors.

Moore went on to describe reduced costs as one main attraction of integrated electronics. According to Moore, manufacturing costs were dominated by two factors. First, the cost to manufacture simple chips was nearly independent of the number of components on the chip. In other words, the cost per component was nearly inversely proportional to the number of components. Second, Moore noted that the increased complexity of chips rapidly lead to decreasing yields in manufacturing. This quickly overwhelmed the cost advantages of cramming more components on the chip. As a result of these two cost drivers, integrated electronics had an optimal complexity of circuitry that led to most cost-efficient chips. As improvements in manufacturing led to better yields, the optimum complexity was growing accordingly.

In his 1965 paper, Moore makes two different arguments, which however become integrated. The first argument was based on the costs of making circuits with different complexities. Moore drew production cost curves for two years, 1962 and 1965, which showed that the cost per component was high if there were too few components on the chip or if there were too many. In 1962 the optimum number of components was about 10 whereas in 1965 it was about 50. Extrapolating this trend, Moore noted that during the next five years the manufacturing cost per component could be expected to drop to one tenth of the present cost and that a minimum-cost chip could have 1,000 components.

Moore then switched to existing chips, presenting them as minimum cost chips. Starting from Jack Kilby's 1959 one-transistor chip, and adding data from a circuit that was under development, Moore observed that the number of components on minimum-cost integrated circuits had increased roughly by a factor of two per year. He further noted that this would mean that by 1975 the number of components per integrated circuit would be 65,000. According to Moore, this would be a minimum-cost chip.

So whilst I2 and Itanium might fit the transistor count they are not minimum cost chips!

What is also often forgotten is that Moores' Law was subsequently reformulated time and again. In 1975 he revisited his original "Law" whereby he stated that the number of components on the most complex chips would double about every two years. This was when it first became known as Moores Law and the method of counting was changed to transistors.

However bearing in mind his original prediction of 65,000 components by 1975 when that time came around his prediction was already wrong!

Moore's 1979 paper subsequently makes an important amendment to the 1975 Moore's Law. Moore still claims that the complexity of the most complex commercially available chips had - for the first 15 years - been doubling roughly every year and, after that, doubling every two years. Now he however explicitly notes that the curve represents the limit of achievable complexity and that most chips fall far beyond the curve. (Hence how you can now fit in the I2 etc!)

Since the 1980s, Moore's Law has been used often and in many different ways. For example, Intel once described Moore's Law on its Web site in the following way:

"Gordon Moore made his famous observation in 1965, just four years after the first integrated circuit was discovered. The press called it "Moore's Law" and the name has stuck. In his original paper, Moore predicted that the number of transistors per integrated circuit would double every 18 months. He forecast that this trend would continue through 1975 ... Moore's Law has been maintained for far longer, and still holds true ..."
This statement was on Intels website back in 2002 and has now been subsequently amended to every 24 months!
However Moore himself has noted:

"I never said 18 months. I said one year, and then two years ... Moore's Law has been the name given to everything that changes exponentially".

Whereas Moore's observation was originally about the number of components on a lowest-cost chip, it was, however, quickly extended outside this well-defined area. There have been three main extensions, which all transform Moore's Law qualitatively. First, Moore's Law has been defined as the "doubling of processing power on a chip every 18 months". Second, it has been defined as "doubling of computing power every 18 months". Third, it has been defined as "price of computing power falling by half every 18 months".

All of which are deviations from his original predictions.
OK enough of the history lessons!

EDIT: lol just did a search of Intels site and sure enough the 18 month claim still exists - here is one where they try and show it is roughly 18 months from the 4004 in 1971 and its 2250 transistors to the P4 with less than 100 million transistors in 2000 - someone in Intel obviously cant do Maths! By 1999 if it was a 18 month doubling period they would need to have over a billion transistors!
http://www.intel.com/technology/itj/2002/volume06issue02/art07_emergingdirections/p03_delivery.htm
It would seem since 2002 they have changed their message so that it still fits Moores Law lol

hah! that was a great post sir!

its funny you brought up what intel puts on their site with regards to their take on moore's law, because guess what...that's exactly where i got that table from! (notice, only intel chips on there )

i have actually learned more about moore's law from your one post than i ever knew before, so i have to say, thank you very much for the time on the very well thought out and presented post!

basically, my main point was that it is not a "law" in the sense that i normally think of a "law", like laws of physics, that are pretty much infallable. my post does go to show that this "law" can be basically bent and shaped anyway one wants, as demonstrated by intel.

so, is there a real corretation between moore's "law" as HE wrote it and the chips we have today?

Well with the move to dual cores and (eventually) quad cores cpu manufacturers are essentially putting about twice the transistors in a CPU so the transistor count aspect of Moore's law still has (at least the potential) to continue for a few more years. Not sure if it exactly fits the timeline or how the pricing will go but it will at least be a decent approximation I suspect (which is all it should've been taken to be in the first place, not a "law").

hUMANbEATbOX said:

hah! that was a great post sir!

its funny you brought up what intel puts on their site with regards to their take on moore's law, because guess what...that's exactly where i got that table from! (notice, only intel chips on there )

i have actually learned more about moore's law from your one post than i ever knew before, so i have to say, thank you very much for the time on the very well thought out and presented post!

basically, my main point was that it is not a "law" in the sense that i normally think of a "law", like laws of physics, that are pretty much infallable. my post does go to show that this "law" can be basically bent and shaped anyway one wants, as demonstrated by intel.

so, is there a real corretation between moore's "law" as HE wrote it and the chips we have today?

Click to expand...

As you rightly point out there is always a correlation and the reason why there is a correlation is precisely as you pointed out - that his "law" can be bent and shaped anyway one wants so that it fits!
In terms of as he originally wrote it then no, in terms of components (as he wrote in '65) the hypothesis failed by the mid '70s and never caught up. (even with just a single component in '65 a cpu today would need 550 billion components using his originally doubling yearly theory).
Actually you can find Moores original paper on the matter on Intel's website (but you probably know that).
The quote I gave about Moore stating he never said 18 months was in 2003 so that would explain why later Intel documentation shows 24 months or 18 - 24 months. (That quote is also on Intels website).

Albigger said:

Well with the move to dual cores and (eventually) quad cores cpu manufacturers are essentially putting about twice the transistors in a CPU so the transistor count aspect of Moore's law still has (at least the potential) to continue for a few more years. Not sure if it exactly fits the timeline or how the pricing will go but it will at least be a decent approximation I suspect (which is all it should've been taken to be in the first place, not a "law").

Click to expand...

I understand what you are trying to say - but think about the terminology you are using - you are not getting twice the transistors on a "cpu". Remember what the term "cpu" means

We even have a thread about that:
http://www.ocforums.com/showthread.php?t=386443

yeah, yeah....ok. I saw that thread too. Guess I should've thought more before I posted about that.

What if something else is invented so that CPU's don't rely on transistors?

delorean700 said:

What if something else is invented so that CPU's don't rely on transistors?

Click to expand...

Somewhere I read about this. Using like micro organisms and stuff or something to that effect.... see if i can dig up a link....

http://computer.howstuffworks.com/dna-computer.htm

there is also work being done on using light pulses, and laserz and such...

give it another 10 years or so...