Computex 2021: NVIDIA Announcement Recap

Computex 2021: NVIDIA Announcement Recap

June 3, 2021 0 By Lorena Mejia

Officials with NVIDIA couldn’t physically attend the virtual Computex 2021 event, so they “flew in” virtually using the popular GeForce RTX 3080! They touched ground with tons of information in tow, including the transformational power of accelerated computing from gaming to the enterprise data center.

Jeff Fisher, Senior VP of GeForce, started off the keynote with gaming, the most prevalent pastime of the COVID-19 pandemic, whose revenue grew to $180B. “Every person born this year will be a gamer, 140 million of them,” said Fisher. “In fact, Gen-Z prefers video gaming as their favorite entertainment activity over just about anything.”

RTX 30 Series

And while the current RTX 30 series have been extremely popular, NVIDIA announced not one but two more additions. The GeForce RTX 3080 Ti is the new flagship gaming GPU based on the Ampere architecture. With 2nd generation RT Cores and 3rd generation Tensor Cores, Ampere is the greatest generational leap ever which is why the 3080 Ti is being dubbed the best of NVIDIA’s gaming lineup.

The RTX 3080 Ti is 1.5X faster than its predecessor, the 2080 Ti, and can power through the latest games with all the settings cranked up. The GPU comes equipped with 12GB G6X memory and has 34 Shader-TFLOPS, 67 R-TFLOPS, and 273 Tensor-TFLOPS. This GPU is set to be released on June 3. Moving on, NVIDIA announced the GeForce RTX 3070 Ti, which is 1.5X faster than a 2070 Super. This GPU has up to 8GB G6X memory, 22 Shader-TFLOPS, 42 RT-TFLOPS, and 174 Tensor-TFLOPS. It’s set to be released on June 10.

NVIDIA AI

Then, NVIDIA focused on artificial intelligence. “It is time to democratize AI by bringing its transformative power to every company and its customers,” said Manuvir Das, Head of Enterprise Computing. He went on to explain that NVIDIA made NVIDIA-Certified, a program for servers that incorporate GPU acceleration. This will help system manufacturers create AI-optimized designs. Currently, there are dozens of new servers that are certified to run NVIDIA AI Enterprise software. Some of the world’s leading manufacturers with this new technology include Dell, Gigabyte, HPE, Lenovo, and Supermicro.  

NVIDIA BlueField DPU

Finally, Das announced that NVIDIA will expand its NVIDIA-Certified systems with BlueField DPUs (data processing unites). “Going forward, the DPU will be an essential component of every server, in the data center and at the edge,” he added. The DPU has become the third member of the data-centric accelerated computing model, joining GPUs and CPUs. “The CPU is for general-purpose computing, the GPU is for accelerated computing, and the DPU, which moves data around the data center, does data processing,” explained NVIDIA CEO Jensen Huang during an interview in 2020.

To watch the entire NVIDIA keynote, click here.