Personal tools
You are here: Home Research Trends & Opportunities New Media and New Digital Economy Future Compute and Microelectronics Moore's Law and Technologies to Extend It

Moore'S Law and Technologies to Extend It

Moore's Law_032923A
[Wikipedia: Moore's Law: A semi-log plot of transistor counts for microprocessors against dates of introduction, nearly doubling every two years (1970-2020).]

 

- Overview

Moore's Law is one of the best technology predictions of the past 50 years. Gordon E. Moore predicted that the number of components on an integrated circuit (IC) would double every year. His hypothesis became known as Moore's Law and was confirmed in 1975. The increase in chip density is mainly attributed to four main factors: chip size, circuit size, technological sophistication, and technological innovation. 

According to Moore, one of the main attractions of integrated electronics is their low cost. This advantage becomes more and more significant as technology advances; a single semiconductor substrate can implement more complex circuit functions. The cost per component in a simple circuit is almost inversely proportional to the number of components. 

However, as the number of components increases, the cost per component tends to rise, and lower yields compensate for the added complexity. Therefore, at any given period in the development of technology, costs are at their lowest. For example, the projected manufacturing cost per component in 1970 was only one-tenth of that in 1965.

In Moore's own words, at first it was just an observation, an attempt to predict that this would be a way to reduce the cost of electronics. However, these industries are moving at a constant rate of improvement, and various technology nodes appear periodically to keep up. Therefore, all business players realize that if they don't move fast, they will fall behind in technology, driving growth faster and faster.

Please refer to the following for more information:

 

- Moore's Law

Moore's Law predicts that the number of transistors on a microchip doubles approximately every two years, leading to increased computer performance and functionality while simultaneously reducing costs. 

This trend, observed by Gordon Moore in 1965, is not a natural law but a description of the progress driven by innovation in semiconductor technology. It's not a fixed law of nature, but rather a projection based on the historical trend of exponential growth in electronics.

  • Transistor Count Doubling: Moore's Law specifically states that the number of transistors that can be placed on an integrated circuit doubles roughly every two years, according to the Intel Newsroom.
  • Impact on Performance and Cost: As transistor density increases, computers become more powerful (faster processing, more memory) while becoming more affordable due to reduced manufacturing costs per transistor, says Robinhood.
  • Not a Natural Law: It's crucial to understand that Moore's Law isn't a physical law like gravity. It's a description of a trend based on ongoing technological advancements in the semiconductor industry, according to Wikipedia.
  • Driven by Innovation: For this trend to continue, continuous innovation is required to overcome the challenges of shrinking transistors and increasing density.
  • Exponential Growth: The law reflects the exponential growth observed in electronics over time, with no apparent end to this trend.

 

- Extending Moore's Law

Moore's Law, the observation that the number of transistors on a microchip doubles approximately every two years, is slowing down, but the tech industry is actively exploring technologies to extend its benefits. 

These include advanced transistor designs, lithography breakthroughs, and advanced packaging techniques like 3D chip stacking and chiplets. Additionally, innovations in materials, software, and computing architectures are crucial for continued performance gains. 

Extending Moore's Law:

  • Advanced Transistor Designs: Moving beyond traditional planar transistors to designs like gate-all-around (GAA) transistors (RibbonFET) that wrap the gate around the channel, improving performance and efficiency.
  • Lithography Breakthroughs: Continued advancements in lithography techniques, enabling the creation of smaller and more densely packed transistors, are essential.
  • Advanced Packaging: Techniques like EMIB (Embedded Multi-Die Interconnect Bridge) for side-to-side connections and Foveros for 3D stacking of chiplets allow for greater transistor density and performance within a smaller footprint.
  • Chiplets: The use of chiplets, smaller, independently manufactured chip components that are interconnected, allows for greater flexibility and customization in chip design and manufacturing.
  • Material Innovations: Developing new materials with superior properties, such as those just three atoms thick, can help overcome the limitations of silicon and enable further miniaturization.
  • Software and Architectural Innovations: Optimizing software for parallel processing, developing heterogeneous compute architectures (combining different types of processors), and exploring new computing paradigms like neuromorphic computing are crucial for maximizing performance gains.


Beyond Transistor Scaling:

  • 3D Integration: Stacking multiple layers of transistors on top of each other, increasing transistor density without shrinking individual transistors.
  • Silicon Photonics: Utilizing light (photons) for data transmission on chips, offering potential for higher bandwidth and lower power consumption.
  • New Materials and Devices: Exploring alternative materials and devices, such as carbon nanotubes or quantum computing, as potential replacements or complements to traditional transistors.


Impact and Future: 

  • Performance Gains: These innovations aim to maintain and even accelerate improvements in processing speed, energy efficiency, and density of integrated circuits.
  • Cost Reduction: While initial costs may be higher, the long-term goal is to make computing more powerful and affordable.
  • New Applications: These advancements are crucial for supporting emerging technologies like AI, IoT, and advanced driver-assistance systems (ADAS).


Challenges and Considerations: 

  • Economic and Physical Limits: Reaching the limits of miniaturization and the increasing complexity of manufacturing pose significant challenges.
  • Energy Consumption: Managing the power consumption of increasingly complex and dense chips is a critical concern.
  • Software and Architecture: Adapting software and hardware architectures to fully utilize the potential of these new technologies is essential.

 

Stanford University_121322B
[Stanford University]

 - 12 Materials that Could Extend Moore's Law

Moore's Law states that as the number of transistors on a silicon chip doubles approximately every two years, the performance and functionality of computers will continue to increase while prices will decrease. This was a prediction made by American engineer Gordon Moore in 1965.

The law describes the trend in semiconductor production. It is not a natural process, but is driven by technological progress and requires constant innovation to continue. The law is based on the observation that electronic products have increased exponentially over time and there is no sign of technological stagnation.

Here are 12 materials or technologies that could keep our hardware performance improving for years to come. 

  • Extreme Ultraviolet (EUV) Lithography: Current photolithography has reached its limits. Extreme ultraviolet (EUV) lithography uses smaller light waves, allowing for higher density chips. 
  • Vacuum tubes: Some researchers are considering reviving the long-obsolete vacuum tube technology. The nanofabrication group at Caltech is developing micropipes to avoid the unpredictable behavior of silicon once it starts reaching low nanometer measurements. 
  • Graphene: The most widely known super material, graphene is a two-dimensional material composed of a single layer of carbon atoms. It is harder than steel, harder than diamond, flexible and transparent, and has excellent electrical conductivity.  
  • Carbon nanotubes: Graphene, but rolled up like newspaper, making it very strong and conductive. Encounters the same difficulties as graphene in terms of mass production.  
  • Stanene and 2D friends: Graphene was the first 2D material, but many more have been discovered. Stanene, Silicene, Germanene, White Graphene, Phosphorene, Molybdenum Disulfide, and Tin Monoxide all have their own unique "supermaterial" qualities.  
  • Diamond: Replacing silicon with synthetic diamond-based transistors, capacitors, and resistors has the potential to eliminate many overheating issues, enabling better performance and removing heat sinks from devices.  
  • Quantum computing: It is unlikely to replace chips in mobile phones forever, but it may have important applications in high-performance computing scenarios. It's all very complicated, but swapping binary bits that are 1 or 0 for qubits that are all 1 or 0 could dramatically increase our current computing power and could make AI smarter, encryption better, and predictive analytics more precise. 
  • Perovskite: A material first discovered in the Ural Mountains of Russia in 1839, perovskite allows electronic devices to work in the terahertz band. Using light instead of electricity to move data can increase computing and internet speeds by up to 1,000 times. 
  • Electronic blood: What if you could cool electronics directly with liquid? IBM's 5D electronic blood aims to do just that.  
  • Neuroelectronics: The brain is very good at learning things in a fast, energy-efficient manner. IBM is working on artificial neurons that can fire and carry electrical impulses in a similar way to our own organic neurons, meaning machines will be able to think like us.  
  • Biocomputers: In addition to mimicking biology, some companies are trying to use biology in machines. Companies such as Microsoft are using artificial DNA for data storage, while researchers are working on writing code into bacteria or using proteins found in the human body in microchips.  
  • Biodegradable microchips: With the e-waste problem growing, electronics that can degrade in a harmless way are promising. The University of Wisconsin-Madison has been working on a way to replace harmful materials in semiconductors, such as gallium arsenide, with a thin layer of wood crystals bonded together with epoxy. The resulting electronics can be dissolved in a glass of water and are more drinkable than regular tap water.  

 

 

[More to come ...]


Document Actions