Empyrean Technology

Empyrean Technology Co., Ltd. (华大九天; SZ: 301269) was founded in Beijing in 2009 by Liu Weiping, an engineer who had worked decades earlier on Panda, one of China's very first domestic electronic design automation (EDA) efforts — the software category that translates a chip designer's circuit intent into the physical layout instructions a foundry actually manufactures from. Panda dates back to a 1990s Chinese government push to build indigenous chip-design tools, an effort that ultimately could not keep pace with Cadence Design Systems and Synopsys as they came to dominate global EDA through the 1990s and 2000s. Liu's return to found Empyrean two decades later reflects a recurring theme across China's semiconductor supply chain: early, government-seeded technology efforts that stalled commercially were kept alive institutionally long enough for their veterans to re-emerge once policy conditions turned favorable. Empyrean listed on the Shenzhen Stock Exchange's ChiNext board in July 2020, becoming the first major domestic EDA company to go public in China. EDA sits near the very top of the semiconductor design stack — before a chip design can be handed to a foundry like SMIC or TSMC for fabrication, engineers must simulate, verify, and physically lay out the circuit using specialized software that models transistor-level physics, timing, power, and manufacturability. Cadence and Synopsys, both American companies, together with Siemens EDA (formerly Mentor Graphics), have controlled roughly 70-80% of the global EDA market for decades, an entrenchment so deep that virtually every chip designed anywhere in the world, including in China, has historically depended on their tools. Empyrean's product suite has expanded to cover analog/mixed-signal and PMIC design and verification, RF IC design, digital SoC design flows, foundry-specific process design kits, advanced packaging design, and flat-panel display chip design — a broad attempt to replicate, tool by tool, the functional coverage that Cadence and Synopsys built over forty years. Empyrean's commercial rise is best understood as a direct, policy-driven consequence of the same export-control regime that has throttled other Chinese chip companies — except in Empyrean's case, the controls target its competitors rather than Empyrean itself. Beginning with 2022 BIS restrictions and tightened further in subsequent rounds, the US government limited Cadence's and Synopsys's ability to sell advanced EDA tools and support services to Chinese entities linked to advanced-node chipmaking or military end-uses, and increasingly to a broader swath of Chinese semiconductor customers generally. That created durable, policy-manufactured demand for a credible domestic alternative regardless of whether Empyrean's technology matched Cadence or Synopsys tool-for-tool — Chinese chipmakers facing restricted or uncertain access to US EDA tools had a strong incentive to qualify Empyrean's software as a hedge, even where it remained functionally behind. This dynamic makes Empyrean nearly unique among the companies profiled in this dataset: it is a beneficiary of controls imposed on someone else's export access, rather than a target of controls on its own. Empyrean today holds an estimated 6% of China's total EDA market — a small slice of the overall pie still dominated by Cadence, Synopsys, and Siemens EDA — but that 6% represents roughly half of all EDA revenue earned by Chinese-owned suppliers domestically, making Empyrean the clear leader among indigenous vendors by a wide margin. Global EDA market penetration for Empyrean remains under 2%, underscoring how far the company still has to go before it competes as a true peer internationally rather than as a domestic-substitution option. The company has pushed into memory-chip design tools, with adoption reported at CXMT (ChangXin Memory Technologies) as China's DRAM self-sufficiency drive accelerates, as well as display-panel chip design tools — segments where Empyrean can establish reference customers without needing to match Cadence and Synopsys at the most advanced logic nodes. Empyrean's financial trajectory illustrates the tension between policy tailwind and technical reality. In the first half of 2026, the company reported revenue of RMB 501-502 million (roughly $70 million), up about 13% year-over-year, but net profit attributable to shareholders collapsed roughly 92% to just over RMB 3 million, as R&D spending consumed nearly 73% of revenue. That imbalance reflects a company racing to close a decades-wide technology gap against much larger, better-resourced rivals while serving a domestic-substitution market that remains, in absolute dollar terms, still fairly small relative to Cadence's and Synopsys's global scale. Empyrean's stated ambition — helping China "fully replace foreign products" in EDA — will require years more of this heavy reinvestment, and its ultimate success or failure will likely track how durable and how far-reaching Washington's EDA export restrictions on Cadence and Synopsys turn out to be, since that policy gap is the foundation Empyrean's entire domestic position currently rests on.