NVIDIA's highly-anticipated GTC 2024 Conference is just days away, and the firm has given developers a sneak peek of its next-gen Blackwell AI GPUs.
NVIDIA's GTC 2024 Will Set The Tone For The Upcoming AI Markets Will Feature New Hardware & Software Announcements Including Next-Gen Blackwell AI GPUs
The eyes of the industry are on how NVIDIA will move into the future in the era of AI, and the GTC 2024 does note a pivotal moment in the company's journey ahead. More importantly, this year's GTC will also be a physical event, & it is expected to welcome 16,000 members, a massive increase from the last time we saw an in-person GTC.
Reuters reports that NVIDIA has given developers a sneak peek about what they should anticipate, showcasing the upcoming Blackwell AI GPUs and the significant advancements in the firm's CUDA platform.
Starting with Blackwell GPUs, NVIDIA sees it as the next big "AI weapon" of the market, potentially replacing the H100 in popularity.
In terms of the specifications, we aren't sure what it could come with, but seeing the rumors in the past, the Blackwell GPUs could utilize a chiplet (MCM) design or stick with a monolithic design for now and move to a multi-package solution in its revision next year. The accelerator will also feature HBM3e memory, possibly higher memory capacities, upgraded specs, and enhanced features such as 16-Hi stacks though a refreshed model could also adopt the next-gen HBM4 standard. Moreover, Dell's SvP recently pointed out that they are making sure their cooling solutions provide up to 1000W cooling capacity for these chips, which isn't huge considering the computing power you get from these things.
Apart from new advancements in the firm's AI hardware portfolio, NVIDIA is expected to showcase generational upgrades within CUDA's platform, which plays a vital role in the AI performance the accelerators can achieve. While we aren't sure what kind of improvements we can see within the stack, it will help refine the NVIDIA ecosystem, not allowing competitors to fill in the gaps.
NVIDIA's GTC 2024 is expected to be held on 18th March and is set to revolutionize the AI industry in every way possible. It will be interesting to see what sort of approach Team Green takes from hereon, as it would define the firm's market presence and financial performance.
NVIDIA Data Center / AI GPU Roadmap
| GPU Codename | Feynman | Rubin (Ultra) | Rubin | Blackwell (Ultra) | Blackwell | Hopper | Ampere | Volta | Pascal |
|---|---|---|---|---|---|---|---|---|---|
| GPU Family | GF200? | GR300? | GR200? | GB300 | GB200/GB100 | GH200/GH100 | GA100 | GV100 | GP100 |
| GPU SKU | F200? | R300? | R200? | B300 | B100/B200 | H100/H200 | A100 | V100 | P100 |
| Process Tech | TSMC A16? | TSMC N2P? | TSMC N3P? | TSMC 4NP | TSMC 4NP | TSMC 5nm | TSMC 7nm | TSMC 12nm | TSMC 16nm |
| CPU | Rosa | Vera | Vera | Grace | Grace | Grace | N/A | N/A | N/A |
| Memory | HBM4e/HBM5? | HBM4 | HBM4 | HBM3e | HBM3e | HBM2e/HBM3/HBM3e | HBM2e | HBM2 | HBM2 |
| Launch | 2028 | 2027 | 2026 | 2025 | 2024 | 2022-2024 | 2020-2022 | 2018 | 2016 |
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
