Science In Action: "Any More for Moore’s Law?"

Date: April 17, 2025
Host: Ronan Pease
Featured Guests: Gordon Moore, Sri Samavedam, Scott Aaronson, Nick Harris, Stan Williams, Gareth Nelson Davis
Summary Compiled By: [Podcast Summarizer AI]

Overview

This edition of BBC’s Science in Action, marking the 60th anniversary of Moore’s Law, explores the remarkable, decades-long trend that has seen the number of components on computer chips double at steady intervals—propelling computers’ ever-increasing power. Host Ronan Pease investigates how semiconductor innovation has extended Moore’s Law, what might limit it physically and economically, and which emerging technologies—quantum, photonic, and analog memristor computing—are lining up for the post-silicon era. The show blends historical context, technical insight, and honest realism about the enduring, evolving race to build better computers.

Main Discussion Points & Insights

The Origins and Endurance of Moore’s Law

• Gordon Moore’s 1965 Prediction
  In 1965, Moore projected that the number of components on an integrated circuit would double yearly for a decade, a “naive” forecast that sowed the seeds for what became industry dogma.

  • Quote:
    “I took the rather bold and frankly naive idea and said, okay, we're going to do this for another 10 years... Instead of putting 60 components on a chip, we'll be putting 60,000...”
    — Gordon Moore (02:46)

• Becomes a Self-Fulfilling Prophecy
  The industry’s collective ambition to meet or beat this pace of progress (later slowed from doubling yearly to roughly every two years) transformed Moore’s Law from a prediction to an agenda.

  • Quote:
    “Moore's Law has gone from recording what's going to happen to driving what is happening in the industry.”
    — Gordon Moore (03:37)

• Physical and Practical Limits
  As components have shrunk to atomic dimensions, Moore himself anticipated fundamental physical barriers.

  • Quote:
    “...the atomic nature of matter starts to impact what the electrical properties... I believe we're going to find that the physics changes sufficiently... and that's going to be a real limit...”
    — Gordon Moore (05:08)

How Modern Industry Stretches Moore's Law

• Advanced Photolithography and Manufacturing
  Sri Samavedam (IMEC) explains how cutting-edge optics allow patterns just 13.5 nanometers wide—approaching the X-ray spectrum—to be etched onto silicon wafers.

  • Quote:
    "It involves shooting laser light on droplets of metal to produce these UV rays... using extremely precise lenses."
    — Sri Samavedam (10:02)
  • Chips are manufactured in vast, automated, 24/7 fabs (factories), where process throughput and cost optimization are vital.

• Stacked and Heterogeneous Packaging
  As shrinking horizontal features gets harder, companies are now stacking chips vertically and placing them side-by-side in a package, achieving chip packages with 200–300 billion transistors each, and aiming for a trillion by 2030s.

  • Quote:
    “So we're stacking chips on top of each other... we track transistors in packages... we expect in the 2030 time frame, it'll reach over a trillion transistors.”
    — Sri Samavedam (08:16)

• Future Nodes: Shrinking to Angstroms
  IMEC is already working on angstrom-scale transistor nodes (seven angstroms = 0.7 nanometers), pushing the limits of fabrication well into the early 2030s.

  • Quote:
    “Right now we are working on a 7 Angstrom technology... which we expect will go into production in the 2032-35 time frame.”
    — Sri Samavedam (11:43)

Alternative Technologies Poised for the Post-Moore Era

Quantum Computing

• Current Progress and Hype
  Scott Aaronson (University of Texas) tempers breathless claims, likening today’s quantum computers (50–100 “qubits”) to where classical computers were in the 1940s.

  • Quote:
    “Quantum computing... is about at the stage that classical computing was at in the 1940s.”
    — Scott Aaronson (14:03)
  • Major practical thresholds—especially error correction—remain unresolved, though the first error-corrected qubits are a milestone.
    • Quote:
      “There's one very, very important threshold which is called the error correction threshold... just within the last year, we have seen arguably the first demonstrations of an error corrected qubit.”
      — Scott Aaronson (16:37)

• Quantum vs. Classical: A Moving Target
  As classical computing (thanks to Moore’s Law) becomes ever more powerful, the bar for quantum computing’s practical advantage keeps rising.

  • Quote:
    “It's meant that the bar keeps getting higher and higher. Yes.”
    — Scott Aaronson (15:08)

Optical/Photonic Computing

• Why Light?
  Nick Harris (Light Matter CEO) points to electrical current’s limits (heat, speed) and highlights how photonics—light-based links and operations—could surpass them.

  • Quote:
    “We're leveraging light, which follows a completely different set of physics, to find a new scaling pathway...”
    — Nick Harris (18:14)

• Where Photonics Enters Today
  Photonic links are now being integrated into AI supercomputers, with mainstream adoption expected around 2027, enabling vast, high-speed chip clusters.

  • Quote:
    “...by 2027 you'll see a lot of these AI supercomputers are going to be based on photonic links... giant million GPU clusters”
    — Nick Harris (19:18)

• Photonic Processing
  Optical modulators can be used to perform mathematical operations key to AI (like multiplications and additions), promising gains in performance and energy efficiency.

Memristors and Neuromorphic Computing

• Analog Memory and Processing
  Stan Williams (Texas A&M) advocates for memristors—devices that “remember” electrical history and store data in analog, not digital, form, echoing the function of biological synapses.

  • Quote:
    “A memristor is an electronic device that remembers its history…”
    — Stan Williams (23:20)

• Scaling by Precision, Not Just Quantity
  Moore’s Law’s spirit continues: instead of doubling component numbers, memristors allow us to double computational power by increasing analog resolution.

  • Quote:
    “...adding one equivalent bit of precision to our analog states every year or two...doubling the amount of computation that can be performed...”
    — Stan Williams (26:05)

• Real-World Deployment
  Memristors are already being used secretly in hundreds of millions of chips, mainly for specialized functions, with broader adoption likely “a lot sooner” than a decade away.

  • Quote:
    “I think it's going to be a lot sooner than that... probably on the order of hundreds of millions of chips already exist with some small number.”
    — Stan Williams (28:59)

Notable Quotes & Memorable Moments

| Timestamp | Speaker | Quote | |-----------|--------------------------|---------------------------------------------------------------------------------------------------------| | 02:46 | Gordon Moore | “I took the rather bold and frankly naive idea and said, okay, we're going to do this for another 10 years...” | | 03:37 | Gordon Moore | “Moore's Law has gone from recording what's going to happen to driving what is happening in the industry.” | | 05:08 | Gordon Moore | “...going to be a real limit in how far we can go just making things smaller.” | | 07:21 | Sri Samavedam | “It's the confidence in the collective ability of smart people to innovate continuously and solve difficult technical problems.” | | 11:43 | Sri Samavedam | “Right now we are working on a 7 Angstrom technology... into production in the 2032-35 time frame.” | | 14:03 | Scott Aaronson | “Quantum computing... is about at the stage that classical computing was at in the 1940s.” | | 16:37 | Scott Aaronson | “We've seen arguably the first demonstrations of an error corrected qubit. So that's a very key threshold...” | | 18:14 | Nick Harris | “We're leveraging light... to find a new scaling pathway to make computers better, more efficient.” | | 23:20 | Stan Williams | “A memristor is an electronic device that remembers its history, which is where the word mem comes in.” | | 26:05 | Stan Williams | “...the digital world, you know, it's either black or white. Your switch is more like a wall switch ... in the analog world, you have the entire range of gray.” | | 28:59 | Stan Williams | “I think it's going to be a lot sooner than that... probably on the order of hundreds of millions of chips already exist with some small number.” | | 31:10 | Gordon Moore | “Since then the concept of Moore's Law has been extended to a lot of things in the industry. Kind of anything that changes geometrically is called Moore's Law. And I'm perfectly willing to take credit for all of it.” |

Timeline of Key Segments

• 01:54 – First-hand recollections from Gordon Moore on the beginnings and foresight of Moore’s Law
• 06:33–11:43 – Sri Samavedam (IMEC): The present and near-future technological roadmap for silicon chips, next-gen lithography, and chip packaging
• 13:11–17:28 – Scott Aaronson: Where quantum computing stands, what makes it hard, and whether its progress mirrors Moore’s Law
• 18:14–22:44 – Nick Harris: The rise of photonics—how linking and one day processing with light could break new ground
• 23:20–30:39 – Stan Williams: The memristor’s analog computing promise (and secret adoption), neuromorphic computing, and continuation of Moore-like scaling by precision
• 31:10 – Gordon Moore reflects on the broader cultural impact and extension of 'Moore’s Law'

Tone and Takeaways

• The episode is reflective and celebratory, but also clear-eyed: Moore’s Law, once simply a hopeful forecast, became both an industry metric and imperative that’s now up against inescapable physical and economic bounds.
• The excitement around quantum and photonic computing is real, but so is the caution—the gap between compelling lab feats and industrial replacement is vast.
• Even as new approaches like memristors quietly begin seeping into practice, the host and guests consistently note how long breakthrough technologies often simmer before entering the mainstream, mirroring the slow burn of early transistor development.
• Moore’s Law’s true lesson, as the episode concludes, is to “find a technology that scales. Find something that you can continuously improve, and then ride it.”

For listeners interested in the frontiers of computing and the interplay of physics, engineering, and economics, this episode offers both inspiration from past achievements and realism about future challenges.