~Q Group Chat [AMP Laboratory Backstage: 679837318]~Really?Two days ago, NVIDIA announced its Q2 financial report for fiscal year 2026. Although it did not achieve the expected growth, as an industry giant, NVIDIA still maintained its consistent high growth. During the conference call, the company also discussed the recent boost in production of its Blackwell Ultra GB300 AI accelerator.
According to analyst twi@dnystedt, sources have revealed that the fully functional Rubin chip is expected to tape out at TSMC by the end of this year, ready for customer deployment.
Rubin is NVIDIA’s next-generation GPU architecture, which will create top-tier products in the industry, integrating components and elements that set new industry standards. The Vera Rubin platform is said to be a combination of the latest CPU and GPU, reportedly utilizing TSMC’s N3P process and CoWoS-L packaging. More importantly, this indicates NVIDIA’s transition towards smaller chip designs in its AI architecture, beginning to compete with AMD on this technological front.
Rubin’s I/O chip is reportedly using TSMC’s N5B (5nm) process and connects to 12-Hi HBM4 memory through CoWoS-L packaging. The Vera CPU will utilize both TSMC’s N3P and N3B processes, making it NVIDIA’s first small chip ARM CPU. This helps explain why the NVIDIA Rubin architecture is a complete redesign from the ground up.NVIDIA CEO Jensen Huang claims that the AI computing market is expanding to a $3 trillion to $4 trillion addressable market, and according to his plans, the Rubin product ecosystem will play a crucial role.At this moment, TSMC is busier than ever serving NVIDIA, as this semiconductor giant is handling every element, knowing that NVIDIA’s AI frenzy has a bright outlook, and the upcoming Vera Rubin product ecosystem will drive this trend.Previous Reading: NVIDIA Announces Q2 Financial Report for 2026, Gaming Business Achieves Record High Revenue, 60 Series Graphics Cards to Begin Mass Production in the Second Half of Next Year