Intel says 14A within 5% of TSMC A14, but its own numbers imply a bigger edge

Intel Foundry's chief technology and operations officer, Naga Chandrasekaran, told KeyBanc that Intel's 14A (1.4nm-class) process will deliver performance "within 5%" of TSMC's A14 node. The phrasing is ambiguous: it could mean 14A is 5% faster or 5% slower than its rival. That uncertainty matters because Intel's own published process specifications suggest a larger gap.

Intel officially projects 14A will deliver 15-20% higher performance at the same power compared with its 18A node, or 25-35% lower power at the same frequency and transistor count. TSMC, by contrast, expects A14 to be 10-15% faster than N2 at the same power, or 25-30% lower power under similar conditions. Real-world chips on these nodes show Intel's 18A reaching 5.10 GHz in the Core Ultra 9 388H, while TSMC's N2 appears in AMD's EPYC 9586F at 5.0 GHz and Apple's M6 at 4.78 GHz.

Combining those clock speeds with the companies' stated gains, Tom's Hardware calculates that 14A would hold a 2-11.3% performance advantage over A14 when using the EPYC as the N2 reference, and a 6.5-16.2% advantage when using Apple's M6. Even the most conservative combination — Intel gaining only 15% while TSMC gains the full 15% — puts 14A beyond the "within 5%" estimate. The publication notes these are illustrative calculations, not frequency predictions, and that Intel's conservative forecast may reflect factors outside simple process scaling, or a deliberate strategy of underpromising.