Analysis: Nvidia Vs. Intel Vs. AMD Q1 Earnings Face-Off
Nvidia finished the first quarter of its current fiscal year with more than double the revenue of what Intel and AMD earned combined for roughly similar periods, but the latter two companies ended up having substantially different quarters.
Whereas Nvidia made $44.1 billion in the first quarter, Intel generated $12.7 billion and AMD collected $7.4 billion for a period that ended roughly a month before Nvidia’s. While Nvidia’s first quarter ended on April 27, Intel and AMD capped theirs on March 29.
[Related: Intel Struggles To Reverse AMD’s Share Gains In x86 CPU Market]
Only Nvidia was able to increase revenue both sequentially (12 percent) and year over year (69 percent). Revenue for Intel, on the other hand, was down 11.1 percent from the previous quarter and slightly down from where it was over a year ago. AMD’s revenue was down 3 percent sequentially but up 36 percent year-over-year.
As for net income, Nvidia delivered $18.8 billion in the first quarter, down 10 percent sequentially but up 26 percent year over year. By contrast, Intel suffered a net loss of $800 million, down 115 percent from the same period last year, and AMD earned a net income of $709 million, up 47 percent sequentially and up 476 percent year over year.
The three semiconductor companies reported their first-quarter earnings results over the past five weeks. Their results highlighted the continued success of Nvidia’s fast-growing AI data center business and a big bump in its gaming business, the significant but slightly waning momentum of AMD’s data center business and its upward progress with PCs as well as Intel ceding some data center momentum and struggling on the PC side.
While Nvidia, Intel and AMD face growing competition from large and small companies embracing new or alternative chip architectures, they are nevertheless the biggest independent suppliers of processors for PCs and servers in the IT channel.
What follows is an analysis of how Nvidia, Intel and AMD compared when it came to the revenues they generated from their data center and PC businesses in the first quarter as well as the underlying customer and industry trends for those segments.
Nvidia’s Data Center Reign Continues Unmatched
It was in the data center market where Nvidia’s reign continued unmatched in the first quarter, with AMD making strides with its established CPU business and somewhat nascent GPU business and Intel ceding some momentum it gained in CPUs.
Nvidia’s data center business—which consists of GPUs like the B200, rack-scale platforms like the GB200 NVL72 and associated products like networking switches—came in at $39.1 billion for the first quarter, making up nearly 90 percent of its total revenue.
With AMD’s data center revenue coming in at $3.7 billion and Intel making $4.1 billion from its Data Center and AI Group, this meant that Nvidia commanded more than five times greater spending in the data center market than its two most direct rivals combined.
While Nvidia’s data center business grew both sequentially (10 percent) and year over year (73 percent), AMD’s revenue for the segment declined 4.8 percent sequentially but was 57 percent higher than it was a year ago. Intel, on the other hand, saw data center sales decline by 5 percent sequentially but increase by 8 percent year over year.
Data Center Sales Trends For Nvidia, Intel And AMD
Some 87 percent of Nvidia’s data center revenue came from compute products, with Blackwell-based products such as the GB200 NVL72 platform helping drive a 5 percent sequential increase and a 76 percent year-over-year increase for the sub-category, according to commentary submitted by Nvidia CFO Colette Kress this week.
While Nvidia didn’t provide a more specific breakdown of product sales, the company said in its fourth-quarter earnings report that Blackwell-based products generated $11 billion, representing 31 percent of its data center revenue for the period. This made Blackwell its “fastest product ramp” yet, Kress said back in February.
The Blackwell product ramp expanded to “all customer categories” in the first quarter, according to Kress’ latest CFO commentary.
The rest of Nvidia’s first-quarter data center revenue came from networking products, which grew in sales by 64 percent sequentially and 56 percent year over year. The company attributed this sequential increase to the growth of the NVLink compute fabric underpinning its GB200 systems as well as Ethernet-based products for AI solutions.
The largest source of data center revenue for Nvidia was cloud service providers, which represented just under 50 percent of sales for the entire segment.
The company’s data center business did take a substantial hit, however, from the Trump administration’s new restriction on the export of Nvidia’s H20 GPU into China. This caused Nvidia to incur a $4.5 billion charge in the first quarter that was “associated with H20 excess inventory and purchase obligations as the demand for H20 diminished,” the company said in its first-quarter earnings announcement.
AMD’s data center business, on the other hand, was driven by “strong” demand from hyperscale and enterprise customers for its EPYC CPUs as well as “significant” double-digit percentage growth year over year for its Instinct GPU business, AMD CEO Lisa Su said in the company’s earnings call in early May.
On the CPU side, Su said cloud service providers launched more than 30 new EPYC-based instances in the first quarter. At the same time, the number of EPYC-based instances activated by Forbes 2000 customers more than doubled year over year.
When it came to on-premises data centers, Su said EPYC CPU sales “grew by a large double-digit percent year over year for the seventh straight quarter.”
For AMD’s Instinct GPUs, the company saw revenue increase by “significant double-digit percent year over year,” which was driven by shipments of its Instinct MI325X GPUs for new enterprise and cloud deployments, according to Su.
“Several hyperscalers expanded their use of Instinct accelerators to cover an increasing range of generative AI, search, ranking and recommendation use cases,” she said.
“We also added multiple tier-one cloud and enterprise customers in the quarter, including one of the largest frontier model developers that is now using Instinct GPUs to serve a significant portion of their daily inference traffic,” Su added.
In Intel’s earnings call late April, its CFO, David Zinsner, said that while the company’s data center business was down sequentially, it performed “above expectations,” thanks to “hyperscaler demand for host CPUs for AI servers and storage compute.”
On the same call, Intel Products CEO Michelle Johnston Holthaus said that while there has been excitement around its latest Xeon 6 CPUs for data center consolidation and edge AI projects, the company is “seeing strong demand on older-gen parts in data center.”
Holthaus also noted that there is “good competition” in the data center market, adding that Intel has “strength both in hyperscalers and enterprise,” but the company is seeing “market share challenge” in the “rest of the world.”
Intel did not provide a sales update for its Gaudi 3 AI accelerator chips. However, the company said late last year that it wouldn’t meet its modest revenue expectation of $500 million for 2024. And while Intel continues to sell its Gaudi 3 chips, Holthaus admitted last December that the Gaudi product line “does not allow me to get to the masses.” Intel is now working on a rack-scale platform based on the next-generation Jaguar Shores chip.
Intel’s PC Lead Diminishes As Nvidia And AMD Grow
Intel commanded more spending in the PC market than AMD and Nvidia combined in the first quarter, even when including Nvidia’s professional visualization revenue.
The semiconductor giant’s Client Computing Group revenue was $7.6 billion. AMD made $2.9 billion for its client and gaming segment. Nvidia generated a record $3.8 billion for its gaming segment and $509 million for the professional visualization segment.
However, Intel’s client revenue declined by 13 percent sequentially and by 8 percent year over year. AMD’s client and gaming revenue was up 2.3 percent sequentially and up 28 percent year over year. Nvidia saw the fastest growth, with gaming sales growing 48 percent sequentially and 42 percent year over year. Its professional visualization business was roughly the same from the previous quarter but up 19 percent year over year.
PC Sales Trends For Intel, AMD And Nvidia
From Intel’s Client Computing Group revenue, the company said $6.5 billion came from laptop and desktop processors, which was nearly $600 million lower than it was a year ago. Other segment revenue was $1.1 billion, down $45 million year over year. While the company didn’t identify the products in this secondary segment, previous statements have identified wireless and connectivity products as part of this sub-category.
In Intel’s latest earnings call, Zinsner said that overall client revenue was “below typical seasonality and in line with expectation, with higher-than-expected volumes offset by product mix and competitive pressure.”
Zinsner said that sales of Intel’s AI PC processors, including its latest “Lunar Lake” CPUs, were lower than expected. However, Holthaus noted that the company is still seeing “very strong commercial demand” for such products.
But Holthaus seemed to indicate that Intel is seeing “much greater demand” for its two previous generations of client CPUs, including “Raptor Lake.”
What Intel didn’t make clear was how much Client Computing Group revenue came from edge processors that were counted as part of the Network and Edge Group prior to the company moving that product category under the former segment in the first quarter.
In AMD’s breakdown of its client and gaming businesses, it showed that the CPU segment generated $2.3 billion in revenue, slightly less than the previous quarter but up 68 percent from the same period last year. In the gaming GPU segment, AMD brought in $647 million, up 15 percent sequentially but down 30 percent year over year.
In AMD’s latest earnings call, Su said desktop channel sell-out “increased by more than 50 percent year-over-year.” Sell-through of notebooks using AMD’s Ryzen processors, on the other hand, was “very strong,” with sales of its latest Ryzen AI 300 processors “increasing by more than 50 percent quarter on quarter,” according to Su.
Su said that “demand for AMD-based commercial PCs was also very strong,” with Ryzen Pro PC sell-through growing more than 30 percent year over year. This was “driven by new end customer wins and an 80 percent increase from 2024 in the number of AMD-powered commercial systems from HP, Lenovo, Dell, and Asus,” according to the CEO.
As for AMD’s gaming business, Su noted that the company’s semi-custom system-on-chip sales for video game consoles declined year over year, but this was more than offset by “strong demand” for the company’s Radeon 9070 graphics cards.
“First week sell-out set a record and was more than 10 times higher than our previous best Radeon launch,” she added.
In Nvidia’s CFO commentary, Collete said that the company’s record gaming revenue was driven by sales of its Blackwell-based GPUs.
“Strong adoption by gamers, creatives and AI enthusiasts have made Blackwell our fastest ramp ever,” she said on Nvidia’s earnings call.
As for the company’s professional segment, the CFO said “broader adoption” of its Ada RTX workstation GPUs led sales to grow from the same period last year.
