Can NVDA. US still achieve good results again with its AI chips when the heavyweight financial report hits?

According to the Zhitong Finance APP, NVDA. US will announce its third quarter results on the morning of November 22 Beijing time (after market hours on November 21 US Eastern time). The market expects that Nvidia Q3's earnings per share will reach $3.37, an increase of nearly 25% month on month and nearly 6-fold year-on-year. The revenue will be $15.989 billion, an increase of 18.3% month on month and about 170% year-on-year. AI chips remain the focus in the upcoming financial reports, and the sales of artificial intelligence chips are undoubtedly the most closely watched. At present, major technology companies in the United States and China have stockpiled Nvidia's A100, H100, and H800 GPUs, including Google (GOOGL. US), Amazon (AMZN. US), Meta (META. US), Alibaba (BABA. US), and others. According to Jon Peddie Research's estimate, NVIDIA's GPU shipments accounted for 87% of the entire market this quarter, followed by 10% of AMD (AMD. US) and 3% of Intel (INTC. US). Recently, NVIDIA also announced the upgrade of its H100 artificial intelligence processor and the launch of a new H200 processor, further strengthening the company's leading position in the AI computing market. This chip will be launched in the second quarter of 2024. Analysts believe that the launch of this new chip is a response to the challenges posed by competitors seeking to challenge NVIDIA's dominant position in the field of artificial intelligence. Citigroup analyst Atif Malik said that when it comes to competition, Intel's Gaudi and Google's TPU are "making progress", but NVIDIA's H100 is still "far ahead", especially in larger production. More importantly, although the US government has restricted exports to China, Chinese technology companies have invested in the latest artificial intelligence chips designed for China. For example, Tencent (00700) stated in its third quarter results that the company has already stored a large number of Nvidia H800 AI chips. To maintain the Chinese market, Nvidia has also taken other measures, including launching three AI chips targeting the Chinese market, namely HGX H20, L20 PCle, and L2 PCle, based on Nvidia's Hopper and Ada architectures. The sample delivery time is from November to December this year, and the mass production time is from December this year to January next year. At present, the market generally expects that the revenue of data center departments, including artificial intelligence services, will reach $12.73 billion, a year-on-year increase of 232%. Malik stated in a report that the buyer's market's expectations for Nvidia's data center revenue for the next two quarters are much higher than the consensus expectations on Wall Street. Investors' expectations for the third and fourth quarters were $13.5 billion to $14 billion and $15 billion to $15.5 billion, respectively, compared to previous expectations of $12.9 billion and $14.7 billion, respectively. We believe that investors in this financial report are focused on: the restrictions imposed by the United States on Chinese AI chips and their impact on NVIDIA's (fiscal years 2025 and 2026) data center sales prospects; the sustainability of overall data center sales and early data from new generation AI applications/products; and NVIDIA's new artificial intelligence hardware roadmap, "Malik wrote in a report. In addition, due to the integration of artificial intelligence, Nvidia's gaming department is expected to achieve stable sales. In May of this year, NVIDIA launched Avatar Cloud Engine (ACE) for Games, a customized AI model foundry service that enables AI driven natural language interaction to make non player characters (NPCs) smarter, thereby changing the game and enhancing the gaming experience. This may help Nvidia regain market share in the gaming industry. The market's consensus expectation for Nvidia's gaming department's revenue is $2.7 billion, a year-on-year increase of 71%.

（From IT home）