Moore's law
From Wikipedia, the free encyclopedia
Moore's law is the observation that, over the history of computing hardware, the number of transistors in a dense integrated circuit doubles approximately every two years. The law is named after Gordon E. Moore, co-founder of Intel Corporation, who described the trend in his 1965 paper.[1][2][3] His prediction has proven to be accurate, in part because the law is now used in the semiconductor industry to guide long-term planning and to set targets for research and development.[4] The capabilities of many digital electronic devices are strongly linked to Moore's law: quality-adjusted microprocessor prices,[5] memory capacity, sensors and even the number and size of pixels in digital cameras.[6] All of these are improving at roughly exponential rates as well. This exponential improvement has dramatically enhanced the impact of digital electronics in nearly every segment of the world economy.[7] Moore's law describes a driving force of technological and social change, productivity and economic growth in the late 20th and early 21st centuries.[8][9][10][11]
The period is often quoted as 18 months because of Intel executive David House, who predicted that chip performance would double every 18 months (being a combination of the effect of more transistors and their being faster).[12]
Although this trend has continued for more than half a century, Moore's law should be considered an observation or conjecture and not a physical or natural law. Sources in 2005 expected it to continue until at least 2015 or 2020.[note 1][14] However, the 2010 update to the International Technology Roadmap for Semiconductors predicted that growth will slow at the end of 2013,[15] when transistor counts and densities are to double only every three years.
History
The term "Moore's law" was coined around 1970 by the Caltech professor, VLSI pioneer, and entrepreneur Carver Mead in reference to a statement by Gordon E. Moore.[2][16] Predictions of similar increases in computer power had existed years prior. Moore may have heard Douglas Engelbart, a co-inventor of today's mechanical computer mouse, discuss the projected downscaling of integrated circuit size in a 1960 lecture.[17] A New York Times article published August 31, 2009, credits Engelbart as having made the prediction in 1959.[18]
Moore's original statement that transistor counts had doubled every year can be found in his publication "Cramming more components onto integrated circuits", Electronics Magazine 19 April 1965. The paper noted that the number of components in integrated circuits had doubled every year from the invention of the integrated circuit in 1958 until 1965[19] and then concluded:
The complexity for minimum component costs has increased at a rate of roughly a factor of two per year. Certainly over the short term this rate can be expected to continue, if not to increase. Over the longer term, the rate of increase is a bit more uncertain, although there is no reason to believe it will not remain nearly constant for at least 10 years. That means by 1975, the number of components per integrated circuit for minimum cost will be 65,000. I believe that such a large circuit can be built on a single wafer.[1]Moore slightly altered the formulation of the law over time, in retrospect bolstering the perceived accuracy of his law.[20] Most notably, in 1975, Moore altered his projection to a doubling every two years.[21][22] Despite popular misconception, he is adamant that he did not predict a doubling "every 18 months." However, David House, an Intel colleague, had factored in the increasing performance of transistors to conclude that integrated circuits would double in performance every 18 months.[note 2]
In April 2005, Intel offered US$10,000 to purchase a copy of the original Electronics Magazine issue in which Moore's article appeared.[24] An engineer living in the United Kingdom was the first to find a copy and offer it to Intel.[25]
Other formulations and similar laws
Several measures of digital technology are improving at exponential rates related to Moore's law, including the size, cost, density and speed of components. Moore himself wrote only about the density of components, "a component being a transistor, resistor, diode or capacitor,"[26] at minimum cost.Transistors per integrated circuit. The most popular formulation is of the doubling of the number of transistors on integrated circuits every two years. At the end of the 1970s, Moore's law became known as the limit for the number of transistors on the most complex chips. The graph at the top shows this trend holds true today.
Density at minimum cost per transistor. This is the formulation given in Moore's 1965 paper.[1] It is not just about the density of transistors that can be achieved, but about the density of transistors at which the cost per transistor is the lowest.[27] As more transistors are put on a chip, the cost to make each transistor decreases, but the chance that the chip will not work due to a defect increases. In 1965, Moore examined the density of transistors at which cost is minimized, and observed that, as transistors were made smaller through advances in photolithography, this number would increase at "a rate of roughly a factor of two per year".[1]
Dennard scaling. This suggests that power requirements are proportional to area (both voltage and current being proportional to length) for transistors. Combined with Moore's law, performance per watt would grow at roughly the same rate as transistor density, doubling every 1–2 years. According to Dennard scaling transistor dimensions are scaled by 30% (0.7x) every technology generation, thus reducing their area by 50%. This reduces the delay by 30% (0.7x) and therefore increases operating frequency by about 40% (1.4x). Finally, to keep electric field constant, voltage is reduced by 30%, reducing energy by 65% and power (at 1.4x frequency) by 50%.[note 3] Therefore, in every technology generation transistor density doubles, circuit becomes 40% faster, while power consumption (with twice the number of transistors) stays the same.[28]
The exponential processor transistor growth predicted by Moore does not always translate into exponentially greater practical CPU performance. Since around 2005–2007, Dennard scaling appears to have broken down, so even though Moore's law continued for several years after that, it has not yielded dividends in improved performance.[29][30][31] The primary reason cited for the breakdown is that at small sizes, current leakage poses greater challenges, and also causes the chip to heat up, which creates a threat of thermal runaway and therefore further increases energy costs.[29][30][31] The breakdown of Dennard scaling prompted a switch among some chip manufacturers to a greater focus on multicore processors, but the gains offered by switching to more cores are lower than the gains that would be achieved had Dennard scaling continued.[32][33] In another departure from Dennard scaling, Intel microprocessors adopted a non-planar tri-gate FinFET at 22 nm in 2012 which is faster and consumes less power than a conventional planar transistor.[34]
Quality adjusted price of IT equipment. The price of Information Technology (IT), computers and peripheral equipment, adjusted for quality and inflation, declined 16% per year on average over the five decades from 1959 to 2009.[35] However, the pace accelerated to 23% per year in 1995-1999 triggered by[11] faster IT innovation, and later slowed to 2% per year in 2010-2013.[35][36] The rate of quality-adjusted microprocessor price improvement likewise varies, and is not linear on a log scale. Microprocessor price improvement accelerated during the late 1990s, reaching 60% per year (halving every nine months) versus the typical 30% improvement rate (halving every two years) during the years earlier and later.[37][38]
The number of transistors per chip cannot explain quality-adjusted microprocessor prices fully.[37][39][40] Moore's 1995 paper does not limit Moore's law to strict linearity or to transistor count, “The definition of 'Moore's Law' has come to refer to almost anything related to the semiconductor industry that when plotted on semi-log paper approximates a straight line. I hesitate to review its origins and by doing so restrict its definition.”[26]
Moore (2003) credits chemical mechanical planarization (chip smoothing) with increasing the connectivity of microprocessors from two or three metal layers in the early 1990s to seven in 2003.[41] This has leveled off at 9-11 layers since 2007.[42][43] Connectivity improves performance, and relieves network congestion. Just as additional floors may not enlarge a building's footprint, nor is connectivity tallied in transistor count. Microprocessors rely more on communications (interconnect) than do DRAM chips, which have three or four metal layers.[44][45][46] Microprocessor prices in the late 1990s improved faster than DRAM prices.[37]
Hard disk drive areal density. A similar law (sometimes called Kryder's law) has held for hard disk drive areal density.[47] The rate of progress in disk storage over the past decades has sped up more than once, corresponding to the utilization of error correcting codes, the magnetoresistive effect and the giant magnetoresistive effect. The outlook for the rate of progress slowed in recent years, because of noise related to smaller grain size of the disk media, thermal stability and writability using available magnetic fields.[48]
Network capacity. According to Gerry/Gerald Butters,[49][50] the former head of Lucent's Optical Networking Group at Bell Labs, there is another version, called Butters' Law of Photonics,[51] a formulation which deliberately parallels Moore's law. Butter's law[52] says that the amount of data coming out of an optical fiber is doubling every nine months. Thus, the cost of transmitting a bit over an optical network decreases by half every nine months. The availability of wavelength-division multiplexing (sometimes called WDM) increased the capacity that could be placed on a single fiber by as much as a factor of 100. Optical networking and dense wavelength-division multiplexing (DWDM) is rapidly bringing down the cost of networking, and further progress seems assured. As a result, the wholesale price of data traffic collapsed in the dot-com bubble. Nielsen's Law says that the bandwidth available to users increases by 50% annually.[53]
Pixels per dollar. Similarly, Barry Hendy of Kodak Australia has plotted pixels per dollar as a basic measure of value for a digital camera, demonstrating the historical linearity (on a log scale) of this market and the opportunity to predict the future trend of digital camera price, LCD and LED screens and resolution.[54][55][56]
The great Moore's law compensator (TGMLC), generally referred to as bloat, and also known as Wirth's law, is the principle that successive generations of computer software acquire enough bloat to offset the performance gains predicted by Moore's law. In a 2008 article in InfoWorld, Randall C. Kennedy,[57] formerly of Intel, introduces this term using successive versions of Microsoft Office between the year 2000 and 2007 as his premise. Despite the gains in computational performance during this time period according to Moore's law, Office 2007 performed the same task at half the speed on a prototypical year 2007 computer as compared to Office 2000 on a year 2000 computer.
Library expansion was calculated in 1945 by Fremont Rider to double in capacity every 16 years, if sufficient space were made available.[58] He advocated replacing bulky, decaying printed works with miniaturized microform analog photographs, which could be duplicated on-demand for library patrons or other institutions. He did not foresee the digital technology that would follow decades later to replace analog microform with digital imaging, storage, and transmission mediums. Automated, potentially lossless digital technologies allowed vast increases in the rapidity of information growth in an era that is now sometimes called an Information Age.
The Carlson Curve is a term coined by The Economist [59] to describe the biotechnological equivalent of Moore's law, and is named after author Rob Carlson.[60] Carlson accurately predicted that the doubling time of DNA sequencing technologies (measured by cost and performance) would be at least as fast as Moore's law.[61] Carlson Curves illustrate the rapid (in some cases hyperexponential) decreases in cost, and increases in performance, of a variety of technologies, including DNA sequencing, DNA synthesis and a range of physical and computational tools used in protein expression and in determining protein structures.
As a target for industry and a self-fulfilling prophecy
Although Moore's law was initially made in the form of an observation and forecast, the more widely it became accepted, the more it served as a goal for an entire industry. This drove both marketing and engineering departments of semiconductor manufacturers to focus enormous energy aiming for the specified increase in processing power that it was presumed one or more of their competitors would soon actually attain. In this regard, it can be viewed as a self-fulfilling prophecy.[4][62]Moore's second law
As the cost of computer power to the consumer falls, the cost for producers to fulfill Moore's law follows an opposite trend: R&D, manufacturing, and test costs have increased steadily with each new generation of chips. Rising manufacturing costs are an important consideration for the sustaining of Moore's law.[63] This had led to the formulation of Moore's second law, also called Rock's law, which is that the capital cost of a semiconductor fab also increases exponentially over time.[64][65]Major enabling factors and future trends
Numerous innovations by a large number of scientists and engineers have helped significantly to sustain Moore's law since the beginning of the integrated circuit (IC) era. Whereas assembling a detailed list of such significant contributions would be as desirable as it would be difficult, below just a few innovations are listed as examples of breakthroughs that have played a critical role in the advancement of integrated circuit technology by more than seven orders of magnitude in less than five decades:- The foremost contribution, which is the raison d’etre for Moore's law, is the invention of the integrated circuit itself, credited contemporaneously to Jack Kilby at Texas Instruments[66] and Robert Noyce at Fairchild Semiconductor.[67]
- The invention of the complementary metal–oxide–semiconductor (CMOS) process by Frank Wanlass in 1963.[68] A number of advances in CMOS technology by many workers in the semiconductor field since the work of Wanlass have enabled the extremely dense and high-performance ICs that the industry makes today.
- The invention of the dynamic random access memory (DRAM) technology by Robert Dennard at I.B.M. in 1967.[69] that made it possible to fabricate single-transistor memory cells, and the invention of flash memory by Fujio Masuoka at Toshiba in the 1980s,[70][71][72] leading to low-cost, high-capacity memory in diverse electronic products.
- The invention of chemically amplified photoresist by C. Grant Willson, Hiroshi Ito and J.M.J. Fréchet at IBM c.1980,[73][74][75] that was 10-100 times more sensitive to ultraviolet light.[76] IBM introduced chemically amplified photoresist for DRAM production in the mid-1980s.[77][78]
- The invention of deep UV excimer laser photolithography by Kanti Jain[79] at IBM c.1980,[80][81][82] that has enabled the smallest features in ICs to shrink from 800 nanometers in 1990 to as low as 22 nanometers in 2012.[83] This built on the invention of the excimer laser in 1970[84] by Nikolai Basov, V. A. Danilychev and Yu. M. Popov, at the Lebedev Physical Institute. From a broader scientific perspective, the invention of excimer laser lithography has been highlighted as one of the major milestones in the 50-year history of the laser.[85][86]
- The interconnect innovations of the late 1990s. IBM developed CMP or chemical mechanical planarization c.1980, based on the centuries-old polishing process for making telescope lenses.[87] CMP smooths the chip surface. Intel used chemical-mechanical polishing to enable additional layers of metal wires in 1990; higher transistor density (tighter spacing) via trench isolation, local polysilicon (wires connecting nearby transistors) and improved wafer yield (all in 1995).[88][41] Higher yield, the fraction of working chips on a wafer, reduces manufacturing cost. IBM with assistance from Motorola used CMP for lower electrical resistance copper interconnect instead of aluminum in 1997.[89]
Some of the new directions in research that may allow Moore's law to continue are:
- Researchers from IBM and Georgia Tech created a new speed record when they ran a supercooled silicon-germanium transistor above 500 GHz at a temperature of 4.5 K (−269 °C; −452 °F).[92][93]
- In April 2008, researchers at HP Labs announced the creation of a working memristor, a fourth basic passive circuit element whose existence had previously only been theorized. The memristor's unique properties permit the creation of smaller and better-performing electronic devices.[94]
- In February 2010, Researchers at the Tyndall National Institute in Cork, Ireland announced a breakthrough in transistors with the design and fabrication of the world's first junctionless transistor. The research led by Professor Jean-Pierre Colinge was published in Nature Nanotechnology and describes a control gate around a silicon nanowire that can tighten around the wire to the point of closing down the passage of electrons without the use of junctions or doping. The researchers claim that the new junctionless transistors can be produced at 10-nanometer scale using existing fabrication techniques.[95]
- In April 2011, a research team at the University of Pittsburgh announced the development of a single-electron transistor 1.5 nanometers in diameter made out of oxide based materials. According to the researchers, three "wires" converge on a central "island" which can house one or two electrons. Electrons tunnel from one wire to another through the island. Conditions on the third wire result in distinct conductive properties including the ability of the transistor to act as a solid state memory.[96]
- In February 2012, a research team at the University of New South Wales announced the development of the first working transistor consisting of a single atom placed precisely in a silicon crystal (not just picked from a large sample of random transistors).[97] Moore's law predicted this milestone to be reached in the lab by 2020.
- In April 2014, bioengineers at Stanford University developed a new circuit board modeled on the human brain. 16 custom designed "Neurocore" chips simulate 1 million neurons and billions of synaptic connections. This Neurogrid is claimed to be 9,000 times faster and more energy efficient than a typical PC. The cost of the prototype was $40,000; however with current technology a similar Neurogrid could be made for $400.[98]
- The advancement of nanotechnology could spur the creation of microscopic computers and restore Moore's Law to its original rate of growth.[99][100][101]
The trend of scaling for NAND flash memory allows doubling of components
manufactured in the same wafer area in less than 18 months.
Ultimate limits of the law
On 13 April 2005, Gordon Moore stated in an interview that the law cannot be sustained indefinitely: "It can't continue forever. The nature of exponentials is that you push them out and eventually disaster happens". He also noted that transistors would eventually reach the limits of miniaturization at atomic levels:
In terms of size [of transistors] you can see that we're approaching the size of atoms which is a fundamental barrier, but it'll be two or three generations before we get that far—but that's as far out as we've ever been able to see. We have another 10 to 20 years before we reach a fundamental limit. By then they'll be able to make bigger chips and have transistor budgets in the billions.[102]In January 1995, the Digital Alpha 21164 microprocessor had 9.3 million transistors. This 64-bit processor was a technological spearhead at the time, even if the circuit's market share remained average. Six years later, a state of the art microprocessor contained more than 40 million transistors. It is theorised that with further miniaturisation, by 2015 these processors should contain more than 15 billion transistors, and by 2020 will be in molecular scale production, where each molecule can be individually positioned.[103]
In 2003, Intel predicted the end would come between 2013 and 2018 with 16 nanometer manufacturing processes and 5 nanometer gates, due to quantum tunnelling, although others suggested chips could just get bigger, or become layered.[104] In 2008 it was noted that for the last 30 years it has been predicted that Moore's law would last at least another decade.[91]
Some see the limits of the law as being in the distant future. Lawrence Krauss and Glenn D. Starkman announced an ultimate limit of around 600 years in their paper,[105] based on rigorous estimation of total information-processing capacity of any system in the Universe, which is limited by the Bekenstein bound. On the other hand, based on first principles, there are predictions that Moore's law will collapse in the next few decades [20–40 years]".[106][107]
One could also limit the theoretical performance of a rather practical "ultimate laptop" with a mass of one kilogram and a volume of one litre. This is done by considering the speed of light, the quantum scale, the gravitational constant and the Boltzmann constant, giving a performance of 5.4258 ⋅ 1050 logical operations per second on approximately 1031 bits.[108]
Then again, the law has often met obstacles that first appeared insurmountable but were indeed surmounted before long. In that sense, Moore says he now sees his law as more beautiful than he had realized: "Moore's law is a violation of Murphy's law. Everything gets better and better."[109]
Futurists and Moore's law
Futurists such as Ray Kurzweil, Bruce Sterling, and Vernor Vinge believe that the exponential improvement described by Moore's law will ultimately lead to a technological singularity: a period where progress in technology occurs almost instantly.[110]
Although Kurzweil agrees that by 2019 the current strategy of ever-finer photolithography will have run its course, he speculates that this does not mean the end of Moore's law:
Moore's law of Integrated Circuits was not the first, but the fifth paradigm to forecast accelerating price-performance ratios. Computing devices have been consistently multiplying in power (per unit of time) from the mechanical calculating devices used in the 1890 U.S. Census, to [Newman] relay-based "[Heath] Robinson" machine that cracked the Lorenz cipher, to the CBS vacuum tube computer that predicted the election of Eisenhower, to the transistor-based machines used in the first space launches, to the integrated-circuit-based personal computer.[111]Kurzweil speculates that it is likely that some new type of technology (e.g. optical, quantum computers, DNA computing) will replace current integrated-circuit technology, and that Moore's Law will hold true long after 2020.[111]
Seth Lloyd shows how the potential computing capacity of a kilogram of matter equals pi times energy divided by Planck's constant. Since the energy is such a large number and Planck's constant is so small, this equation generates an extremely large number: about 5.0 * 1050 operations per second.[110]He believes that the exponential growth of Moore's law will continue beyond the use of integrated circuits into technologies that will lead to the technological singularity. The Law of Accelerating Returns described by Ray Kurzweil has in many ways altered the public's perception of Moore's law. It is a common (but mistaken) belief that Moore's law makes predictions regarding all forms of technology, when it was originally intended to apply only to semiconductor circuits. Many futurists still use the term Moore's law in this broader sense to describe ideas like those put forth by Kurzweil. Kurzweil has hypothesised that Moore's law will apply – at least by inference – to any problem that can be attacked by digital computers as is in its essence also a digital problem. Therefore, because of the digital coding of DNA, progress in genetics may also advance at a Moore's law rate. Moore himself, who never intended his law to be interpreted so broadly, has quipped:
Moore's law has been the name given to everything that changes exponentially. I say, if Gore invented the Internet, I invented the exponential.[112][note 4]
Consequences and limitations
Technological change is a combination of more and of better technology. A 2011 study in the journal Science showed that the peak of the rate of change of the world's capacity to compute information was in the year 1998, when the world's technological capacity to compute information on general-purpose computers grew at 88% per year.[114] Since then, technological change has clearly slowed. In recent times, every new year allowed mankind to carry out roughly 60% of the computations that could have possibly been executed by all existing general-purpose computers before that year.[114] This is still exponential, but shows the varying nature of technological change.[115]The primary driving force of economic growth is the growth of productivity,[116] and Moore's law factors into productivity. Moore (1995) expected that “the rate of technological progress is going to be controlled from financial realities.”[26] However, the reverse could and did occur around the late-1990s, with economists reporting that "Productivity growth is the key economic indicator of innovation."[11] An acceleration in the rate of semiconductor progress contributed to a surge in US productivity growth[117][118][119] which reached 3.4% per year in 1997-2004, outpacing the 1.6% per year during both 1972-1996 and 2005-2013.[120] As economist Richard G. Anderson notes, “Numerous studies have traced the cause of the productivity acceleration to technological innovations in the production of semiconductors that sharply reduced the prices of such components and of the products that contain them (as well as expanding the capabilities of such products).”[121]
While physical limits to transistor scaling such as source-to-drain leakage, limited gate metals, and limited options for channel material have been reached, new avenues for continued scaling are open. The most promising of these approaches rely on using the spin state of electron spintronics, tunnel junctions, and advanced confinement of channel materials via nano-wire geometry. A comprehensive list of available device choices shows[122] that a wide range of device options is open for continuing Moore's law into the next few decades. Spin-based logic and memory options are actively being developed in industrial labs[123] as well as academic labs.[124]
Another source of improved performance is in microarchitecture techniques exploiting the growth of available transistor count. Out-of-order execution and on-chip caching and prefetching reduce the memory latency bottleneck at the expense of using more transistors and increasing the processor complexity. These increases are empirically described by Pollack's Rule which states that performance increases due to microarchitecture techniques are square root of the number of transistors or the area of a processor.
For years, processor makers delivered increases in clock rates and instruction-level parallelism, so that single-threaded code executed faster on newer processors with no modification.[125] Now, to manage CPU power dissipation, processor makers favor multi-core chip designs, and software has to be written in a multi-threaded manner to take full advantage of the hardware. Many multi-threaded development paradigms introduce overhead, and will not see a linear increase in speed vs number of processors. This is particularly true while accessing shared or dependent resources, due to lock contention. This effect becomes more noticeable as the number of processors increases. There are cases where a roughly 45% increase in processor transistors have translated to roughly 10–20% increase in processing power.[126]
On the other hand, processor manufactures are taking advantage of the 'extra space' that the transistor shrinkage provides to add specialized processing units to deal with features such as graphics, video and cryptography. For one example, Intel's Parallel JavaScript extension not only adds support for multiple cores, but also for the other non-general processing features of their chips, as part of the migration in client side scripting towards HTML5.[127]
A negative implication of Moore's law is obsolescence, that is, as technologies continue to rapidly "improve", these improvements can be significant enough to rapidly render predecessor technologies obsolete. In situations in which security and survivability of hardware or data are paramount, or in which resources are limited, rapid obsolescence can pose obstacles to smooth or continued operations.[128] Because of the toxic materials used in the production of modern computers, obsolescence if not properly managed can lead to harmful environmental impacts.[129]
Moore's law has significantly impacted the performance of other technologies: Michael S. Malone wrote of a Moore's War following the apparent success of shock and awe in the early days of the Iraq War. Progress in the development of guided weapons depends on electronic technology.[130] Improvements in circuit density and low-power operation associated with Moore's law have also contributed to the development of Star Trek-like technologies including mobile phones[131] and replicator-like 3D printing.[132]
