Intel's 18A node crosses the 85% commercial yield threshold with Nvidia, Google, and Apple on its design-win list — while EMIB-T packaging hits 98% yield and cracks TSMC's CoWoS monopoly. Today's briefing covers every layer of the AI hardware stack, from memory architecture to TSMC's upcoming earnings.
Audio is available on Spreaker — see link below.
Intel's foundry just crossed eighty-five percent yield on its eighteen-A node, and that number matters more than almost anything else in the foundry business right now. Eighty-five percent is the threshold where merchant foundry economics actually work.
Design wins and production ramps are two different things. The confirmed wins lack explicit volume commitments or ramp timelines.
The packaging story may actually be more immediately significant. Intel's EMIB-T advanced packaging has hit ninety-eight percent yield, matching TSMC's CoWoS technology.
Stepping back to the equipment layer, ASML raised its full-year twenty twenty-six guidance sharply, lifting sales to between forty-three and forty-five billion euros from a prior range of thirty-six to forty billion. Gross margin guidance moved to fifty-four to fifty-six percent.
On memory architecture, two research teams, one from UNIST in Korea and one from the University of Tokyo, have proposed vertical-DRAM designs that go beyond what HBM four and HBM five can deliver. The UNIST V-Die and Tokyo's MOSAIC architecture both stack dies vertically rather than horizontally, achieving four times more interconnects than HBM four, thirty-seven percent lower latency, and triple the thermal conductivity.
Nvidia's own automotive chief disclosed that Jensen Huang personally arbitrates weekly disputes over internal GPU compute allocation. That single detail tells you more about supply constraints than any quarterly filing.
TSMC reports on July sixteenth. The market is pricing in a gross margin around sixty-seven percent, potentially touching seventy.
Chapter summary auto-generated from the verified script. Listen to the full episode for the complete content.