Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitchesNvidia shares gained 24.7% during the week covered by a May 31, 2023 report, which linked the rise in part to CEO Jensen Huang’s forecast that data centers would shift from general-purpose computing to accelerated systems for generative AI. His $1 trillion figure described the installed infrastructure he said could transition—not a completed replacement program or a guarantee that every data center would be rebuilt.
What Huang said about data-center infrastructure
In NVIDIA’s May 24, 2023 second-quarter earnings statement, founder and CEO Jensen Huang described two simultaneous shifts: accelerated computing and generative AI. He said that “a trillion dollars of installed global data center infrastructure will transition from general-purpose to accelerated computing as companies race to apply generative AI into every product, service, and business process.” Data Center Knowledge reported the statement and its market context.
The forecast offered investors a large potential demand story: if organizations adopted generative AI across products and operations, some might need more specialized computing hardware. The report connected that prospect partly to the stock’s weekly gain. It did not establish that Huang’s remarks alone caused the move, nor did his statement document a completed replacement cycle. The scale and cost of any upgrade would depend on individual enterprise needs.
Why accelerated computing could matter—and where the case is narrower
Accelerated systems use specialized processors to handle suitable computing tasks more efficiently than general-purpose systems. The investment argument in 2023 was that demanding AI workloads, particularly model training, could increase demand for such hardware. But AI workloads are not uniform, and the most compute-intensive approach is not automatically the best fit for every organization.
#1 Best Overall
Demanding training workloads
Bradley Shimmin, a data and AI industry analyst at Omdia, acknowledged that some demanding model-training work could motivate investment in newer acceleration hardware. He said companies facing those requirements might invest to reduce costs and speed time to market. That is a workload-specific case for accelerators, not an argument that all computing should move to them. The original report includes Shimmin’s comments.
Smaller models and efficient fine-tuning
Shimmin also pointed to a countervailing direction: smaller models, curated datasets, and efficient fine-tuning. If a use case can be served by a less demanding model or a more efficient development process, its compute needs may be lower than a strategy built around training the largest models. The practical choice is therefore between architectures and approaches suited to particular workloads—not a universal switch from general-purpose computing to accelerators.
Rank #2
- Part number 900-53651-2500-000 and model: P3651
- This is the 2 slot version for when there is no empty slots between 2 slot cards. If you have one or more empty slots between the cards or the cards are 3 slot this NVLink will not work. See the attached images showing the card layout.
- NVLink 3.0 for any brand of RTX Ampere model graphics cards: 3090, A30, A40, A100 / H100 (Requires three NVLinks), A800, A4500, A5000, A5500, A6000
- This is the same as PNY part number: NVLAMP-2SLOT-BSP and RTXA6000NVLINK-KIT
- This is the same as Dell part number: 0RWJ7Y
What the infrastructure thesis leaves out
Computing hardware is only one part of building data-center capacity. In its FY2027 Q2 Form 10-Q, NVIDIA identified land, power, facility shells, and capital as important dependencies for customer buildouts. The filing describes expansion as a complex, multi-year undertaking with regulatory, technical, and construction challenges. It also says customers could delay deployments because infrastructure is unavailable, financing is constrained, or technology adoption proceeds more slowly than expected. These are risks disclosed by NVIDIA, not findings about a particular customer project. See NVIDIA’s SEC filings.
Those constraints matter to the forecast: demand for AI computing does not automatically translate into installed capacity on a predictable timetable. Companies must be able to fund and build the facilities, secure power, and decide that expected workload benefits justify the investment.
Recommended Free Tools
Rank #3
- Video/Sound Cards
- Passive Cooling
How later results relate to the 2023 story
NVIDIA reported $89.0 billion in Data Center revenue for the quarter ended July 26, 2026, up 117% year over year, as part of total revenue of $96.2 billion. In its August 26, 2026 release, Huang said, “AI has reached its inflection point. It’s doing useful work. Its tokens are productive and profitable. Now, compute is revenue.” These are later company-reported results and commentary; they provide current context for the scale of NVIDIA’s business, but they do not independently establish what caused a stock move in 2023. NVIDIA’s investor-relations releases.
Earlier, on May 20, 2026, NVIDIA reported $75.2 billion in Data Center revenue for FY2027 Q1, up 92% year over year. Huang called AI-factory construction “the largest infrastructure expansion in human history”; that is the company’s characterization, not an independently established ranking. The results illustrate that NVIDIA continued to report substantial Data Center revenue growth, while the original 2023 share gain remains a dated market event rather than a measure of current performance. The company’s releases cover both periods.
Quick Recap
Rank #4
- CUDA Cores: 4608 / NVIDIA Tensor Cores: 576 / NVIDIA RT Cores: 72
- GPU Memory: 24 GB GDDR6 with ECC / Bandwidth: 624 GB/Sec
- System Interface: PCI Express 3.0 x16
- Four DisplayPort 1.4 Connectors
- 3D Stereo Support with Stereo Connector
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
