Now Playing
Intel 90% Yield, Meta's GPU Cloud & Nvidia China Curbs
Audio is available on Spreaker — see link below.
Intel's 18A-P hits 90% yield two quarters early while Meta launches a 600K-GPU cloud to rival AWS and Azure on price. Plus Samsung's 1.4nm delay decoded, ASML raises guidance, and US export controls on Nvidia China sales tighten further.
Audio is available on Spreaker — see link below.
Samsung just confirmed it won't hit one-point-four nanometer production until twenty twenty-nine. That's the headline.
Intel is running a different kind of story right now, and it's moving faster than most expected. The eighteen A-P process variant entered risk production on June sixteenth, with first-month yields tracking at ninety percent.
Meta is building a dedicated AI cloud service using a fleet of over six hundred thousand H100-equivalent GPUs. Private beta is targeted for August twenty twenty-six, with general availability in November.
ASML raised its revenue guidance, citing stronger-than-expected demand from AI chip production. The stock is up over twenty-three percent in the past month.
New US export restrictions have tightened controls on Nvidia's AI chip sales to China. This continues a sustained pattern of escalating technology controls targeting the most capable AI accelerators.
Intel broke ground on a hundred and seven thousand square foot foundry manufacturing center in Santa Clara focused on mask operations. Masks are the precision templates used to print circuit patterns onto silicon wafers.
Pulling back to the wider picture: the foundry race through twenty twenty-eight and twenty twenty-nine is no longer just about who hits which node first. Yield stability, design ecosystem support, and geopolitical supply chain positioning are now the real differentiators.
Chapter summary auto-generated from the verified script. Listen to the full episode for the complete content.