Nvidia exceeded all expectations for its second fiscal quarter of 2024 with revenue of $13.51 billion, a 101% jump from the same quarter last year. Net income came in at $6.74 billion, or $2.48 per diluted share, which is up 854% from a year ago and up 202% from the previous quarter.
Analysts had expected revenue to come in at $11.04 billion with earnings per share totaling $2.07, according to data from Bloomberg.
And it’s all thanks for enterprise sales. Last quarter, enterprise sales accounted for 60% of total revenue. This quarter, $10.3 billion of the $13.5 billion in total revenue – 76% – came from data center sales.
“A new computing era has begun. Companies worldwide are transitioning from general-purpose to accelerated computing and generative AI,” said Jensen Huang, founder and CEO of Nvidia, in a statement. “Nvidia GPUs connected by our Mellanox networking and switch technologies and running our CUDA AI software stack make up the computing infrastructure of generative AI.”
Cloud service providers and large consumer internet companies fed Nvidia’s data center revenue, along with strong demand for its HGX platform, which is based on both the Hopper and Ampere GPU architectures.
Another source of Nvidia’s revenue strength is its networking business, which was up 94% from a year ago and up 85% sequentially. Generative AI requires massive amounts of data and it needs to be moved around quickly, and Nvidia’s InfiniBand network infrastructure is being used to support the HGX platform.
On the consumer side, revenue increased 22% from a year ago and 11% sequentially.
An issue that has been plaguing all of the chipmakers is the supply chain. But on the conference call with financial analysts, Nvidia CFO Colette Kress was upbeat. “Our supply partners have been exceptional in ramping capacity to support our needs,” she said on the call. “We expect supply to increase each quarter through next year.”
Nvidia projects revenue for the third quarter of fiscal 2024 to be $16.00 billion, plus or minus 2%, which is well ahead of Wall Street’s projected $12.5 billion in revenue.
This quarter’s numbers may help shoot down rumors that Nvidia was dealing with a severe supply shortage. It clearly sold a lot of chips.
In the enterprise market, processors are extremely profitable. A Series 40 GPU sells to gamers for between $300 and $700, on average. According to the financial consulting firm Raymond James, the street price for an H100 processor is around $25,000 to $30,000, while it costs an estimated $3,320 to manufacture.
That helps explains Nvidia’s astonishing 70% gross margin, which is remarkable for a semiconductor maker. AMD reported a margin of 45% in its most recent quarter, and Intel reported a margin of 35%. Nvidia is selling an awful lot of extremely profitable chips.
Meanwhile the AI train marches on. Gartner estimates that this year, AI chips will account for $53 billion in sales, and almost half of the companies surveyed by CNBC say that AI is their top priority for tech spending over the next year.
Copyright © 2023 IDG Communications, Inc.
