Intel claims its 14A node is shedding defects faster than anything it has built since its old 22nm glory days.
Intel CFO David Zinsner told Deutsche Bank’s 2026 technology conference that 14A is improving at a rate the outfit has not seen since 22nm. That comparison is loaded, since 22nm gave Chipzilla its FinFET lead with Ivy Bridge in April 2012 before the manufacturing wheels started to wobble.
According to Tom’s Hardware Chipzilla says 14A will not hit high-volume manufacturing until 2028, with risk production for internal products pencilled in for the second half of 2027. That means today’s claims are about defect density curves, not proven mass-production yields.
Defect density is a useful early warning signal, but it is not the same as knowing how many usable chips fall off a production line. Modern inspection gear can spot things that might have slipped past tools in 2010, so the 22nm comparison is not quite as clean as the marketing bods would like.
After 14nm delays, the 10nm mess, the 20A cancellation and 18A’s awkward defect-density story, Chipzilla will take any good news it can get. A node that looks healthier than the recent disasters is not a victory parade, but it is less embarrassing.
14A brings second-generation RibbonFET 2 transistors, PowerDirect backside power delivery and High-NA EUV designed in from the start. Chipzilla claims that gives it 15 to 20 per cent better performance at the same power, 30 per cent higher transistor density and 25 to 35 per cent lower power use than 18A.
Intel CFO David Zinsner said customer chats had moved beyond tyre-kicking. “Engagements with customers externally, from a foundry perspective has significantly increased. Lip-Bu and the team are now meeting on weekly basis with customers. They are moving away from just looking at data to thinking about ‘how much capacity can I get? What does that supply look like?’!
Production-minded chip designs will need to be moving before 14A risk production starts in the second half of 2027.
“We are now seeing demand from our internal customers on 14A. They are actually probably the most cynical bunch out of anybody. The fact that they are now designing products on 14A was a good confidence boost for us as well,” Zinsner said.
Intel CEO Lip-Bu Tan has said PDK 0.5 is complete and PDK 0.9 remains on track for October 2026. That release gives designers the rules, models and libraries needed for more serious production-intent work.
The other leg of Chipzilla’s pitch is EMIB-T packaging, which Zinsner reckons could become a multi-billion-dollar-per-customer business each year. Revenue is expected to start ramping in the second half of 2027 and hit its stride by 2029.
