<- Back to laws

Technological and economic trend

Moore's
Law

For decades, the economically practical complexity of integrated circuits rose exponentially. The trend was not a law of nature: it was sustained by lithography, device physics, manufacturing learning, design automation, capital investment, and coordinated industry roadmaps.

TypeEmpirical industry trend
Original metricComponents at minimum cost
Scientific statusNot a physical law
Historical cadenceAnnual, then about biennial
Enabling systemTechnology + economics
Common misuseCPU speed doubles
INTERACTIVE 01 / DOUBLING ENGINE

Exponential progress hides inside ordinary time.

Choose a starting chip, elapsed time, and doubling period. The model shows why a modest-looking cadence creates enormous differences across generations.

040 years
1 year4 years
TEACHING MODELCounts are a mathematical projection, not a product forecast. Real chips differ in area, function, yield, process, packaging, and cost.
EXPONENTIAL MODEL
Doublings10.0elapsed time / period
Relative complexity1,024x2 raised to doublings
Projected count2.36Millustrative transistors
Interactive visual model for Moore's Law.

A log scale turns repeated doublings into a straight line. Each equal vertical step represents multiplication, not addition.

OBSERVATIONComplexity rose
->
EXPECTATIONRoadmaps aligned
->
INVESTMENTTools improved
->
RESULTThe trend persisted
01 / MEANING

The original claim was narrower and more interesting.

In 1965 Gordon Moore examined the number of components that could be placed on an integrated circuit at minimum cost per component. He observed an approximately annual doubling and projected the trend for roughly a decade.[1] The familiar two-year cadence came from his 1975 reassessment, not from the original article.[2]

IDEALIZED FORMN(t) = N0 x 2t/T
N0starting complexity
telapsed time
Tdoubling period

Moore did not claim that every chip, computer, workload, or user experience would double on one schedule. Component count, transistor density, cost per transistor, clock frequency, energy efficiency, and application performance are different measurements.

Original objectIntegrated-circuit complexity at an economic optimum.
Later shorthandTransistor counts roughly double every two years.
Not equivalentPerformance, frequency, and software speed.
"Moore's Law worked because an observation became a coordination device for an entire industrial system."
02 / THE ECONOMIC CORE

More components were useful only when they became economical.

Early integrated circuits faced a tradeoff. Adding components spread packaging and processing costs across more functions, but larger and more complex dies were more vulnerable to manufacturing defects. Moore's curve concerned the moving minimum between those pressures.

COST / COMPONENTCOMPLEXITY ->
ECONOMIC OPTIMUMtoo little integrationyield and complexity penalty
01Integration advantage

More functions share fabrication, packaging, and assembly.

02Yield penalty

Larger or defect-sensitive designs lose more value when a die fails.

03Learning effect

Process control and volume improve yield and reduce unit cost.

04Moving optimum

Technology shifts the cheapest useful complexity upward.

Why this mattersCounting the largest possible chip misses the original economic idea. A technically possible design does not define the minimum-cost point.
03 / SCALING ENGINE

No single invention sustained the curve.

The trend required a coupled production system. Improvements in one layer created pressure and opportunity in the others.

PRINT

Lithography

Shorter wavelengths, better optics, masks, resist, and process control patterned smaller features.

SWITCH

Device physics

New gate structures, materials, and transistor geometries controlled leakage and electrostatics.

BUILD

Manufacturing

Cleanliness, metrology, yield learning, and larger wafers turned designs into volume products.

DESIGN

EDA and architecture

Automation made it possible to specify, verify, and route designs humans could not manage manually.

FUND

Capital

Enormous fabrication and research investments were justified by expected future markets.

ALIGN

Roadmaps

Shared targets coordinated equipment, materials, foundries, designers, and customers.

THE FORMER COMPANION

Dennard scaling

The 1974 MOSFET scaling analysis described how smaller devices could improve density and speed while scaling voltage and current.[3] For years, density gains arrived with favorable power behavior.

DIMENSIONSVOLTAGECAPACITANCEENERGY / SWITCH
04 / SIX DECADES

The trend changed its mechanism while keeping its reputation.

1959-60Planar IC foundations

Planar processing and monolithic integration create a scalable manufacturing platform.

1965One-year projection

Moore charts economically favorable component complexity and projects ten years forward.

1974-75Scaling theory and revision

Dennard and colleagues formalize MOSFET scaling; Moore revises the longer-run cadence toward two years.

1980s-90sRoadmap era

CMOS, optical scaling, EDA, and global supply chains compound density and cost improvements.

2000sPower wall

Voltage scaling weakens. Clock-frequency growth slows and multicore designs become central.

2010s-nowSystem scaling

FinFETs, gate-all-around devices, accelerators, chiplets, 2.5D/3D packaging, and specialized memory broaden the path.

1971Intel 40042,300 transistors
->
1993Intel Pentium3.1 million
->
20123rd-gen Intel Core1.4 billion

The historical product figures above follow Intel's museum material; they illustrate the scale change but do not form a clean scientific dataset because product classes and functions differ.[7]

05 / WHAT STOPPED BEING FREE

Density continued, but its companions separated.

The phrase "Moore's Law is dead" is usually underspecified. Different curves slowed at different times. The end of automatic clock-frequency gains is not identical to the end of density improvement.

TRANSISTOR DENSITY
Continued, unevenly

Device structures and patterning still raise density, with growing difficulty and cost.

CLOCK FREQUENCY
Plateaued earlier

Power and thermal limits ended the simple frequency-scaling era.

COST / TRANSISTOR
No longer guaranteed

Advanced process and design costs complicate the historical economic payoff.

APPLICATION SPEED
Workload dependent

Architecture, memory, parallelism, and software dominate realized benefit.

COMPUTE->MEMORY->INTERCONNECT->POWER->COOLINGA system is limited by the whole path.
06 / MODERN CONTINUATION

Scaling moved from one dimension to many.

The 2024 IEEE IRDS roadmap treats continued scaling as a portfolio of device, interconnect, integration, and design challenges rather than a single shrinking number.[8]

DEVICE

New transistor geometry

FinFET and gate-all-around structures improve electrostatic control as dimensions shrink.

VERTICAL

3D integration

Stacking logic, cache, and memory increases functional density without relying only on planar shrink.

PACKAGE

Chiplets

Functions can use different process nodes and be combined with high-bandwidth die-to-die links.

ARCHITECTURE

Specialization

GPUs, AI accelerators, media engines, and domain-specific units trade generality for efficiency.

MEMORY

Data proximity

High-bandwidth and stacked memory attack energy and latency spent moving data.

SOFTWARE

Parallel execution

Compilers and algorithms must expose concurrency to benefit from wider and heterogeneous hardware.

PACKAGE
COMPUTE
I/O
CACHE
MEMORY
Different dies, one system

The UCIe Consortium defines an open die-to-die interconnect intended to support an interoperable chiplet ecosystem.[9] This is one example of scaling becoming a packaging and system-integration problem.

07 / LIMITS & MISUSE

Always ask: which curve?

WRONG

"Computers double in speed."

Performance depends on workload, architecture, memory, power, and software.

WRONG

"Eighteen months was Moore's original law."

The 1965 projection used an annual cadence; the 1975 revision moved toward two years.

WRONG

"It is a law of physics."

It is an empirical technological-economic trend sustained by investment and coordination.

WRONG

"One node name measures transistor size."

Modern process labels are product names, not one universal physical dimension.

WRONG

"More transistors guarantee value."

Unused complexity can raise cost, verification burden, and energy without improving the task.

WRONG

"The law has one death date."

Density, cost, frequency, energy, and performance trends diverged at different times.

09 / REFERENCES

Sources and further reading.

Original papers, technical roadmaps, official histories, and standards organizations are prioritized.

  1. Gordon E. Moore (1965) - Cramming More Components onto Integrated CircuitsThe original Electronics article on component complexity and minimum component cost.computerhistory.org/.../Moore_1965_Article.pdf
  2. Gordon E. Moore (1975) - Progress in Digital Integrated ElectronicsThe reassessment that separated early one-time factors from the longer-run doubling cadence.IEEE IEDM Technical Digest, 1975
  3. Dennard et al. (1974) - Design of Ion-Implanted MOSFETs with Very Small Physical DimensionsThe foundational MOSFET scaling paper.doi.org/10.1109/JSSC.1974.1050511
  4. Computer History Museum - Moore's Law at 50Historical account of the original projection, economic framing, and industry coordination.computerhistory.org/blog/moores-law50...
  5. Computer History Museum - Silicon Engine TimelineDocumented milestones in planar processing, MOS, integrated circuits, CAD, and manufacturing.computerhistory.org/siliconengine/timeline/
  6. Intel - Moore's Law Press KitIntel's account of the 1965 forecast, 1975 revision, and present advanced-device strategy.newsroom.intel.com/press-kit/moores-law
  7. Intel Museum - Transistors to TransformationsOfficial historical processor examples with transistor count, process, clock, and product context.intel.com/.../museum-transistors-to-transformations...
  8. IEEE IRDS (2024) - More Moore RoadmapCurrent technical roadmap for device, interconnect, integration, and density challenges.irds.ieee.org/images/files/pdf/2024/2024IRDS_MM.pdf
  9. UCIe Consortium - Specifications and ResourcesOfficial resources for the open die-to-die interconnect standard and chiplet ecosystem.uciexpress.org/ucie-resources
  10. Gordon Moore (1997) - An Update on Moore's LawMoore's retrospective discussion of economics, feature scaling, reliability, and physical limits.intel.com/pressroom/archive/speeches/GEM93097.HTM
CONTINUE EXPLORING

Related laws, with the relationship made explicit.

These are editorial connections, not claims that the laws are mathematically equivalent.

CONTINUE READING

Place this law inside the collection.

LAW 002 / 100 PUBLISHED