From a One-Day Fab to Spruce Pine: What Actually Limits Chip Capacity

The gap is not a story about engineers getting worse at their jobs. It is a story about what happens when a manufacturing process grows more complex faster than the automation around it improves. IEEE Spectrum's history of Project SWIFT, published in its December 2024 issue, lays out the numbers plainly. SWIFT averaged five hours to complete each layer of its fabrication process. The fastest modern fabs take nineteen hours per processing layer. The industry average is thirty-six.

SWIFT was the work of Bill Harding, who managed IBM's Manufacturing Research group in East Fishkill, New York. In 1970 he proposed a fully automated wafer line that would turn out integrated circuits in less than a day. At the time, memory chips typically moved through dozens of manual work stations in a monthlong stop-and-go march. IEEE Spectrum reports that SWIFT's turnaround time has never been equaled.

The modern constraint is not the fab floor

Today's chips use far more layers, on wafers the size of small pizzas, and the processing is more intricate. Those factors do not fully account for the difference, according to IEEE Spectrum. But they point at something the SWIFT team never had to manage: a supply chain that runs through a single small town in North Carolina.

Spruce Pine is home to the only facilities producing ultra-pure quartz in the world, according to Tom's Hardware. Two companies mine there, Sibelco North America and The Quartz Corp. The quartz goes into crucibles that hold molten silicon while it is grown into ingots. Purity matters because impurities migrate into the melt at temperatures above 1400 degrees Celsius.

When Hurricane Helene hit in late September 2024, it dropped more than two feet of rain on the area, cut off roads and left residents without power. Sibelco confirmed a suspension of operations, Tom's Hardware reported, while The Quartz Corp said it was too early to determine whether its production facility had been affected. Readers pushed back in the comments, noting the mines sit in the hills rather than the town, and that groundwater flooding underground could not be ruled out.

"The modern economy rests on a single road in Spruce Pine, North Carolina," the Wharton professor Ethan Mollick wrote, as quoted by Construction Physics.

Brian Potter, writing on Construction Physics in November 2024, checked that claim and found it partly right. Quartz crucibles are necessary for most semiconductor manufacturing, and Spruce Pine is where most of the quartz comes from. But the town is not an irreplaceable linchpin. Alternatives exist, he wrote, and they are all some combination of not yet developed, not quite as good and not quite as cheap. Cutting off Spruce Pine would not choke off semiconductors entirely. It would mean yields falling and costs rising.

Capacity promises, measured against the past

That is the backdrop for the current wave of capacity announcements, most of which are about geography rather than speed. In Malaysia, second finance minister Amir Hamzah Azizan told CNBC at the IMF spring meetings in Washington that the country holds 13% of the global market for chip packaging, assembly and testing, per a February report from the Malaysian Investment Development Authority. Malaysia is trying to move up the value chain rather than compete head-on at the leading edge.

India is making a larger bet. TechCrunch reported on 15 July 2026 that New Delhi unveiled a ₹625 billion (about $6.5 billion) Mobile Phone Manufacturing Scheme running five years, plus ₹1.28 trillion (around $13.3 billion) for domestic semiconductor manufacturing. China accounted for 63% of global smartphone production in 2025, against India's 18%, according to Counterpoint Research.

Texas Instruments is taking a different route, expanding internal manufacturing for gallium nitride power semiconductors. The company said on 24 October 2024 that production had begun at its Aizu, Japan factory, quadrupling its internal GaN capacity alongside Dallas. TI's Mohammad Yunus said the company aims to grow internal manufacturing to more than 95% by 2030.