Uploaded January 2026 | Updated September 2026, 3 hours ago
The Cartoon Guide to Physics by Larry Gonick (Author), Art Huffman (Author) Affiliate Link -- amzn.to/4gPEzxT
Are data centers about to be replaced?
Modern data centers consume tens of megawatts of power, require massive cooling infrastructure, and cost tens of millions of dollars to build and maintain. But what if much of that computing power could be delivered without constructing an entire facility?
In this video, we take a physics-first look at Omnia, a portable AI supercomputer designed to deliver massive computational power in a compact, modular form. We explore how systems like Omnia — and scalable configurations like the so-called “Infinity Cube” — could reduce the need for new data centers by allowing companies to deploy high-performance compute in existing office, warehouse, or industrial spaces.
We also dive into the physics that makes this possible:
Why data centers are so energy-hungry
How heat dissipation and electron movement set real limits on computing efficiency
Why moving data often costs more energy than computing itself
Finally, we introduce a radical new chip architecture being developed
The Cartoon Guide to Physics by Larry Gonick (Author), Art Huffman (Author) Affiliate Link -- amzn.to/4gPEzxT
Are data centers about to be replaced?
Modern data centers consume tens of megawatts of power, require massive cooling infrastructure, and cost tens of millions of dollars to build and maintain. But what if much of that computing power could be delivered without constructing an entire facility?
In this video, we take a physics-first look at Omnia, a portable AI supercomputer designed to deliver massive computational power in a compact, modular form. We explore how systems like Omnia — and scalable configurations like the so-called “Infinity Cube” — could reduce the need for new data centers by allowing companies to deploy high-performance compute in existing office, warehouse, or industrial spaces.
We also dive into the physics that makes this possible:
Why data centers are so energy-hungry
How heat dissipation and electron movement set real limits on computing efficiency
Why moving data often costs more energy than computing itself
Finally, we introduce a radical new chip architecture being developed
