Skip to main content

How to watch Nvidia’s Computex 2022 keynote

Nvidia just wrapped its Computex 2022 keynote, which was packed with announcements about the company’s enterprise business. Although the presentation was light on next-gen GPU news, Nvidia still shared some announcements related to liquid-cooled GPUs in the data center and the processors that could power next-gen cloud gaming.

The keynote happened late Monday night in the U.S., lining up with the morning in Taiwan,where Computex takes place. To get you up to speed, we have a replay of the keynote here, as well as all of Nvidia’s announcements.

Recommended Videos

How to watch Nvidia’s Computex 2022 keynote

NVIDIA Keynote at COMPUTEX 2022

Six different Nvidia executives spoke at Nvidia’s keynote, which mostly covered the company’s data center and AI efforts. The presentation aired during late in the U.S., so if you missed the live keynote, you can catch a replay of it above.

Get your weekly teardown of the tech behind PC gaming
Check your inbox!

Nvidia’s keynote came just half a day after AMD kicked off Computex with its executive keynote (where it officially revealed Ryzen 7000 CPUs). Intel skipped this year’s show, where we expected to hear about Arc Alchemist GPUs.

What Nvidia announced at Computex 2022

Nvidia A100 liquid-cooled data center GPU.
Image used with permission by copyright holder

Although we hoped to hear about next-gen RTX 40-series graphics cards, Nvidia focused on its data center GPUs and Grace CPU during its Computex 2022 keynote. It appears we’ll have to wait on next-gen GPU announcements until the fall, when Nvidia has been long-rumored to launch its next-gen GPUs.

We still got some GPU news, though. Nvidia launched liquid-cooled versions of its A100 and H100 data center GPUs in an attempt to reduce energy consumption in mainstream servers. These GPUs will be available as add-in PCIe cards, as well as in Nvidia’s HGX server racks. The A100 is arriving in the second half of the year, while Nvidia expects the H100 in early 2023.

Nvidia says the new GPUs can cut power consumption by 30% in the data center, as well as free up 66% of the space a normal GPU would take up. The cooler uses a direct-to-chip connection, which Nvidia says provides the best cooling potential possible. Liquid cooling is common in supercomputers, but this is the first time we’re seeing it on mainstream servers.

Nvidia also revealed reference designs for its Grace and Grace Hopper processors. These chips are built to handle massive AI workloads, as well as power the next generation of cloud gaming technology. Nvidia announced the Grace Hopper chip, its first-ever dual CPU/GPU chip, at GTC earlier this year.

Nvidia says dozens of models will be available from Asus, Gigabyte, and Supermicro, among others, and they’ll arrive in the first half of 2023. Hopefully by the end of next year, we’ll start to see what these chips can do with platforms like GeForce Now.

Computex wasn’t the most exciting show for PC enthusiasts this year, as Intel skipped it and Nvidia focused mostly on its enterprise business. The show brought news of AMD’s upcoming AM5 platform, though, including the new X670 chipset.

Jacob Roach
Lead Reporter, PC Hardware
Jacob Roach is the lead reporter for PC hardware at Digital Trends. In addition to covering the latest PC components, from…
Nvidia’s new GPUs are already running into problems
Nvidia Blackwell chips.

Nvidia's latest Blackwell GPUs are running into problems in the data center, reports The Information. According to the report, Nvidia's customers are worried about how well the AI accelerators will hold up, as overheating issues have caused delays in server racks being deployed for AI training.

The Blackwell architecture is at the heart of both Nvidia's next-gen AI accelerators and its upcoming RTX 50-series graphics cards. In the data center, the architecture was previously delayed due to "design flaws," pushing the deployment of the B100 and B200 GPUs back. That's despite big orders with AI players like Meta, Microsoft, and Google.

Read more
Nvidia RTX 50-series GPUs: everything we know so far
RTX 4070 seen from the side.

Nvidia already makes some of the best graphics cards, but it's also not resting on its laurels. The RTX 40-series, which has been bolstered by a refresh, is now over two years old, and Nvidia is hard at work on the next generation of GPUs.

The release date of RTX 50-series GPUs is still at least over a month away, but various rumors and leaks give us a better idea of what to expect. Here's everything we know about Nvidia's upcoming generation of graphics cards.
RTX 50-series: pricing and release date

Read more
Nvidia just scaled down DLSS 3, and that’s a good thing
The RTX 4080 Super graphics card sitting on a pink background.

Nvidia's signature tech, DLSS 3, just got yet another update -- and although it's subtle, it actually seems like a good thing for some of the best graphics cards. The latest version, 3.8.10, bundled with the GeForce 566.14 driver, doesn't seem to introduce any major changes, but Nvidia enthusiasts noticed that it's about half the size that it used to be. Where's that difference coming from?

No, Nvidia didn't downgrade DLSS 3 -- at least not in any major way. Although this hasn't been confirmed by Nvidia itself, it appears that the company removed a whole bunch of DLSS presets and replaced them with just two. These presets make it easier for gamers to choose the type of focus they want to apply to each game.

Read more