Why Nvidia can’t simply make more GeForce cards — even if it wants to - PCWorld
Despite high prices for GeForce graphics cards, Nvidia cannot simply shift manufacturing capacity from AI chips to gaming cards due to complex supply chain constraints.
Nvidia is investing billions in AI chip production while gamers face high GeForce prices. However, even if the company wanted to redirect manufacturing capacity toward gaming cards, the process isn't as simple as reallocating factory space. The complexity lies in the specialized nature of semiconductor manufacturing and the distinct requirements for different product lines.
Manufacturing Complexity
While Nvidia designs its chips, TSMC handles most manufacturing, with memory supplied by companies like SK Hynix, Samsung, and Micron. AI accelerators use high-bandwidth memory (HBM) and specialized packaging, while GeForce cards rely on GDDR memory and different packaging techniques. Though there's some overlap in manufacturing resources, these aren't interchangeable production lines.
Typically vendors have to make wafer orders from people like TSMC and wait six months from start of manufacturing to end.
Memory Priorities
Memory manufacturers are currently focused exclusively on HBM and DDR technologies, with little attention paid to GDDR memory that powers GeForce cards. This shift in priorities reflects where the industry sees the greatest demand and profitability. Even if Nvidia wanted to prioritize GeForce production, coordinating the entire supply chain would take months of planning.
Nvidia's financial commitments underscore its focus on AI. The company disclosed $279 billion in supply and manufacturing-capacity commitments in August, primarily supporting its data center business. In October, Nvidia announced another AI manufacturing expansion, including a $1.1 billion advance for semiconductor testing facilities.
Supply Chain Challenges
Manufacturing decisions made during Jensen Huang's tenure continue to influence Nvidia's production capabilities. While his successor could theoretically prioritize GeForce cards, getting more graphics cards to store shelves requires convincing the entire supply chain to realign resources—a difficult proposition when AI hardware generates significantly greater returns.
Related
