SemiAnalysis Deep Dive into Huawei HiSilicon 9030 Pro: Architecture and Performance Significantly Improved, Advancement of Chinese Advanced Process Nodes Unhindered

BlockBeats News, June 15, SemiAnalysis STEEL Lab released a report on the deep teardown of the Huawei Mate 80 Pro equipped with the Kirin 9030 Pro, confirming its use of the SMIC N+3 process, with a minimum metal pitch of only 32.5nm, approximately 10% tighter than Intel's 18A at 36nm. Through aggressive DUV multi-patterning + DTCO optimization, N+3 achieved a logic density equivalent to TSMC N6 (about 113.4 MTr/mm², slightly higher than N6's 107.7), but at the cost of higher process complexity, increased cost, and more challenging yield control.

The Kirin 9030 Pro chip architecture and performance improvement make it an evolution of the Kirin 9020, with a similar die size but higher utilization: adding 1 big core, increasing GPU CUs from 4 to 6, expanding the NPU Tiny core, and enlarging the cache. The core areas of CPU/GPU/NPU have significantly reduced (big core reduced by 22%, GPU CU by 28%), leading to a significant performance improvement compared to the previous generation (GPU 3DMark improvement by 70-79%). However, it still lags behind current flagships (Apple, Snapdragon 8 Elite, etc.), with a much larger energy efficiency gap, approaching the Android flagship level of nearly 3 years ago.

The SemiAnalysis report pointed out that export controls have not prevented China from advancing in advanced processes but have forced SMIC to take a more complex DUV route, which will rely on tighter design rules, backside power delivery, and Huawei's LogicFolding stacking technology in the future. N+3 demonstrates the resilience of the Chinese semiconductor industry, but it still lags behind leading nodes, with maturity and cost being the main constraints.

---------------------------------

Click the original text link below to join the BlockBeats · Feishu AI News Channel and stay tuned for 24/7 monitoring of global AI hot topics and news.