Deutsch English
← back to archive
AI & Technology · 577 words · MN-260624-0020

IBM's NanoStack: The Skyscraper Under the Chips

MuckyNews reading

IBM has pulled the wraps off a new chip design that dips below the one-nanometer barrier, landing at around 0.7 nanometers. The promise: cramming roughly 100 billion transistors onto a surface the size of a fingernail. The performance numbers IBM is throwing around are bold—50 percent faster than their 2nm chips, 70 percent more energy-efficient.

The catch? None of this is coming to your phone or data center anytime soon. Production is still years away. And this isn’t the first time IBM has staged this exact play; back in 2021, they debuted their 2nm tech with almost identical percentage leaps in performance and efficiency. It’s part of the lab game: announce a landmark, even if the factory lines aren't running yet.

The real story here is the architecture. The old playbook of just squeezing transistors closer together on a flat surface is stalling out. Moore’s Law is gasping for air. IBM’s answer is to build upward. Their "NanoStack" tech layers transistors on top of each other. University of Surrey’s Alan Woodward compares it to a 100-story skyscraper, noting that rivals like Samsung and Intel are currently working on the equivalent of 30-to-50-story buildings. But building upward brings its own headaches: heat rises, and if the layers get too thin, the transistors fail to switch off properly. A 100-story chip is incredibly ambitious, and the physics could still push back before this ever reaches a production line.

Short version

  • What happened? IBM unveiled a 0.7nm chip prototype promising 100 billion

transistors on a fingernail-sized surface.

  • Who is involved? IBM as the developer, rivals like Samsung and Intel for

comparison, and the AI/data center industry as the ultimate customer.

  • Why does it matter? Because flat chip scaling is hitting a wall, and 3D

stacking is the new battlefield.

  • What is still open? Whether production can actually be achieved in a few

years and if the heat and switching problems in ultra-thin layers can be solved.

  • What is the real pressure point? The shift from horizontal to vertical

building—and the gap between IBM’s lab announcements and actual market-ready products.

Perspective delta

western_media (BBC): The coverage celebrates the breakthrough as a "landmark moment" and readily adopts IBM’s architectural imagery ("block of flats"). The focus is on the potential benefits for AI and everyday devices. The massive gap between lab prototype and industrial production is noted, as are the persistent heat and switching issues with 3D chips. However, it doesn't emphasize that IBM pulled almost the exact same PR maneuver with similar percentage gains in 2021—that detail is mentioned, but almost in passing. The framing is: big technological leap, with a few realistic caveats attached.

Source base / limits

The source base is extremely narrow. The entire write-up relies on a single BBC article. There are no dissenting voices from the industry, no direct statements from competitors, and no independent hardware expert opinions beyond Alan Woodward. The performance claims rely entirely on IBM’s own data. The resulting picture is therefore one-sided and heavily reliant on corporate PR—the critical aspects are raised, but not independently verified.

Sources / basis

  • MuckyNews container ID: MN-20260624-020
  • Source outlets: BBC
  • Perspective groups: western_media
  • Article count: 1
  • Source count: 1
  • Perspective count: 1
  • Approximate raw text size: 3814 characters