Nvidia Profits Soar 1300%: Earnings Frenzy

Since 2023, Nvidia's stock price has soared by 240%, not only significantly outpacing the growth rate of the NASDAQ index but also exceeding the combined stock price increase ratio of several U.S. stock giants that have benefited from AI. Its current market value has reached 1.2 trillion dollars.

The fervent embrace by the capital market is attributed to Nvidia's skyrocketing performance, with the "weapons" (chips) for generative AI computing power reserves by various tech giants filling Nvidia's coffers. Indeed, Nvidia's latest quarterly report data has once again shocked the world.

One might know it's formidable, but it's hard to imagine just how impressive it is!

Revenue of $18.1 billion brings a net profit of $9.2 billion.

Let's clarify Nvidia's fiscal year timeline, which is quite unique. It does not align with the calendar year, starting on February 1st of each year and ending on January 31st of the following year. Therefore, since February 2023, Nvidia has entered its fiscal year 2024, with August to October 2023 being the third quarter of Nvidia's fiscal year 2024. Please do not confuse this point.

Advertisement

In the third quarter, Nvidia achieved a revenue of $18.12 billion, a sequential increase of 34% and a year-over-year increase of 206%; it achieved a net profit of $9.243 billion, a sequential increase of 49%, and a year-over-year increase of 1259%.

Friends who do not have a clear concept of corporate revenue and profit may not understand the fervor behind these cold numbers. Let's interpret them simply.

Before the AI revolution, Nvidia was globally renowned for selling gaming chips and was a well-known GPU chip company in the industry, but its quarterly revenue was short of the $10 billion mark; after the AI revolution, Nvidia's computing power chips became the coveted darlings of the market, with a continuous stream of orders, and quarterly revenue easily surpassed $10 billion, already aiming for $20 billion.

Compared to revenue, Nvidia's profitability is even more exaggerated, with the third quarter approaching the $10 billion mark, and the operating net margin close to 50%. It should be noted that very few manufacturing companies can achieve a gross margin of 50%, let alone a net margin. The irreplaceability of computing power chips gives Nvidia the confidence to "charge exorbitant prices," and even with high premiums, there are still people scrambling to buy.

Data center business becomes the key to Nvidia's "revival".Using the term "revival" to describe NVIDIA's performance this year might not be entirely appropriate, as the company has never really experienced a decline. It is a dominant force in both gaming and cryptocurrency mining. However, the global craze for ChatGPT has catapulted NVIDIA's products to new heights, with growth rates increasing almost exponentially, as reflected in the financial report data.

NVIDIA's products are primarily used in four areas: data centers, gaming, professional visualization, and automotive. The revenues for the third quarter were $14.514 billion, $2.856 billion, $0.416 billion, and $0.261 billion, respectively.

In other words, the revenue from the data center sector accounts for 80% of the total revenue. This figure was 76.4% last quarter and 64.6% in the third quarter of last year. It is almost certain that this percentage will continue to rise in the future, as long as the demand for AI remains strong, the computing power chips will continue to support NVIDIA in achieving better performance.

The fourth quarter is expected to bring in $20 billion in revenue, with export controls affecting the sequential growth rate.

NVIDIA's financial report was released after the U.S. stock market closed.理论上, such good performance in the third quarter should have led to a significant increase in stock prices. However, NVIDIA's stock price fell after hours. The reason for this may be related to the future outlook.

In this quarter's financial report, NVIDIA预计d that the fourth quarter's operating revenue would be $20 billion, with a fluctuation of no more than 2%.

The expected revenue of $20 billion seems quite good, but we notice that the sequential growth rate has significantly weakened. The revenue in the third quarter increased by 34% compared to the second quarter. According to the forecast, the sequential growth rate for the fourth quarter would be 10.4%, which is less than one-third of the third quarter's growth.

The main reason for NVIDIA's slower pace is related to the Chinese market. In the past few quarters, revenue from China accounted for 20% to 25% of NVIDIA's data center revenue. A large number of technology companies in China have ordered computing power chips from NVIDIA to enhance their big data models.

However, in October of this year, the U.S. government strengthened export controls on cutting-edge computing power chips, including several models such as A100, A800, H100, and H800, which are not allowed to be exported to China.

NVIDIA's CFO mentioned in the financial report meeting that they are developing products that meet the export control requirements and are expected to be launched in the next few months. However, whether chips that have been repeatedly downgraded and significantly reduced in computing power can stimulate the interest of Chinese buyers is uncertain, even to NVIDIA itself.Huawei's Ascend series seizes a favorable opportunity, potentially becoming Nvidia's biggest "threat" in the future.

The Kirin 9000s chip equipped in the MATE 60 PRO series demonstrates Huawei's strength in the field of cutting-edge consumer electronics chips. Is Huawei also involved in the layout of computing chips? The answer is yes. Although it is not yet comparable to Nvidia's products, we can look forward to the gradual realization of domestic substitution.

Huawei's Ascend series processors adopt advanced architectures and designs, with both computing power and energy efficiency ratios being quite good, and have been widely applied in fields such as artificial intelligence, cloud computing, and big data.

Previously, the A100 chip was banned from being sold to China, and Nvidia designed the A800 chip specifically for the Chinese market, which has not much difference in functionality. Now, the A800 has also been banned, and the new product that Nvidia has introduced may be significantly reduced in computing power, which may not meet the needs of Chinese buyers. Domestic substitution is imperative. This undoubtedly gives Huawei an opportunity to narrow the gap between Huawei's Ascend and Nvidia's products. Under the situation of this rise and fall, more buyers will choose domestic substitution solutions.

With the surge in order quantity, the positive feedback from users to chip designers and the experience accumulated in the manufacturing process can help enterprises achieve greater technological breakthroughs as soon as possible. Perhaps it won't take too long for Nvidia to find that there is a strong local competitor in the Chinese computing power market, becoming the biggest "threat" to its solid market share share, which is what we are happy to see.