Introduction

In 1959, physicist Richard Feynman delivered a landmark lecture at Caltech titled There's Plenty of Room at the Bottom. He argued that the laws of physics allow manipulating matter at extremely small scales, envisioning a world where the entire Encyclopedia Britannica fits on a pinhead and all human knowledge fits in a dust-speck-sized device. Decades later, that vision has driven extraordinary progress in computing, yet the easy gains have largely stalled.

What Happened

The second half of the 20th century was defined by relentless scaling. Transistors shrank repeatedly, doubling chip density and enabling the exponential growth captured by Moore's Law. Supporting laws like Dennard's, which linked smaller transistors to lower power per gate, and Koomey's, which tracked computation per joule, reinforced the trend. However, around 2004, Dennard scaling broke down as power density stopped improving, forcing the industry to add cores rather than boost single-thread speed. Other benchmarks have likewise flagged: Kryder's law for magnetic storage has slowed, and Koomey's trajectory has flattened. Meanwhile, the six largest hyperscalers are on track to spend more than $700 billion this year on computing infrastructure, buying performance through brute-force scaling rather than fundamental improvements.

Why This Matters

The gap between today's computing technology and fundamental physical limits is striking. Quantum mechanics, general relativity, and information theory set ultimate ceilings for speed, energy, memory density, and communication. Today's chips operate far above the Landauer minimum energy per bit, and optical interconnects still deliver far fewer bits per second than the Holevo limit allows. Bridging these gaps isn't just about faster hardware; it drives new materials, reversible logic, and deeper scientific insight. Proof-of-concept systems that approach these limits often spawn entire industries, as the transistor, laser, and giant magnetoresistance all began as physics experiments before scaling into general-purpose technologies.

Key Takeaways

The descent into smaller, faster, more efficient computing has slowed, but the room at the bottom remains vast. Four key limits computational speed energy per operation memory density and communication bandwidth define the true boundaries. Reversible computing quantum architectures and bio-inspired storage like DNA each offer pathways closer to those limits. The industry's current strategy of building bigger systems masks the fact that individual device improvements are increasingly costly. Progress now requires both scaling up and pushing toward the physical floor, where each computation bit and photon is optimized at the fundamental level.

Conclusion

History shows that the most transformative technologies emerge when a physics experiment outpaces engineering then gets scaled up. The transistor the laser and modern data storage all followed that path. Today, the fundamental limits of computing stand as both a challenge and a roadmap. Building machines that operate closer to those limits and then expanding outward may be the only way to sustain progress as the ceiling of conventional scaling comes into view. There is still plenty of room at the bottom and the next wave of innovation will likely come from building upward from that floor.