NOVEMBERASIA BUSINESS OUTLOOK9NEWSROOMNVIDIA SECURES EARLY ACCESS TO TSMC'S A16 CHIP PROCESSGoldi Solar, a leading Indian solar module manufacturer, has raised US$171 million in a funding round led by Havells India.The round also saw participation from Ambit Global Private Client, Nikhil Kamath (Zerodha), Shahi Exports, SRF Transnational Holdings, and NSFO Ventures. SBI Capital Markets and Saraf and Partners acted as advisors to the deal.The funding will be directed toward expanding solar cell and module production capacity at Goldi's facilities in Gujarat, targeting 14.7 GW of modules and 4 GW of solar cells by FY27. This investment aligns with India's national goal of boosting domestic solar manufacturing under the ALMM-II framework and reducing dependency on Chinese imports.Key Highlights· Goldi Solar raised $171M led by Havells India for Gujarat expansion· Aims for 14.7 GW modules and 4 GW solar cells by FY27· Aligns with India's ALMM-II push for domestic solar manufacturingHowever, India's current 26 GW of solar cell capacity remains a bottleneck against its 100+ GW module output. Goldi's success will depend on achieving TOPCon or Heterojunction (HJT) cell efficiency standards and securing ALMM-II certification by June 2026.As India aims to reach 115 GW of cell capacity by FY27, the deal signals opportunities for equipment vendors, SaaS providers offering traceability and compliance tools, and suppliers of encapsulants, solar glass, and tinned copper interconnects.The move also supports India's localization drive amid ongoing anti-dumping probes on solar imports from Asia. Nvidia has reportedly secured exclusive early access to TSMC's upcoming A16 semiconductor process, marking a strategic leap in advanced chip development.The two tech giants are conducting joint validation tests on the A16 node, which introduces nanosheet transistors and Super Power Rail (SPR) technology -- innovations that promise tighter power routing, reduced IR drop, and improved energy efficiency.The A16 process is expected to deliver 8­10 percent performance gains and 15­20 percent lower power consumption compared to TSMC's N2P node. Risk production is set to begin in 2026, with volume manufacturing anticipated by late 2026 to 2027. Nvidia's next-generation Feynman GPU architecture, scheduled for release in 2028, is expected to be built on this process.Key Highlights· Nvidia gains early access to TSMC's A16 process with nanosheet transistors· A16 boosts performance 10 percent and cuts power by up to 20 percent over N2P· Feynman GPUs set for 2028 launch on A16, leveraging CoWoS & HBM4 techWhile OpenAI, AMD, and Nvidia are all linked to A16, Nvidia's "exclusive access" appears to represent priority entry for early development. The process leverages second-generation GAAFET nanosheets and backside power delivery, although costs and yields could pose initial challenges for large GPUs.TSMC's CoWoS (Chip-on-Wafer-on-Substrate) capac-ity remains a bottleneck, with Nvidia already reserving 70 percent of 2025 capacity. To address demand, TSMC is investing in CoPoS panel-level packaging and wafer-scale integration.Meanwhile, next-gen memory like HBM4 and HBM4E, expected to reach 16-high stacks using hybrid bonding, will complement A16-class GPUs requiring better thermal and TSV (Through Silicon Via) management. GOLDI SOLAR SECURES $171M TO BOOST CELL MANUFACTURING
< Page 8 | Page 10 >