AMD is preparing Instinct MI400 for its future Roadmaps

More From Author

See more articles

Myntra Upcoming Sales 2025: Your Fashion Calendar for Maximum...

Myntra Upcoming Sales 2025 In the ever-evolving world of fashion e-commerce, Myntra continues to be India's go-to destination...

Dimensity 6020 vs Snapdragon 695: Mid-Range Chipset Battle

Dimensity 6020 vs Snapdragon 695: Qualcomm Snapdragon 695 5G (SD695) is a fast mid-range ARM-based SoC found...

My Jio Recharge Plans as of January 4,...

My Jio Recharge Plans: Since its establishment in 2016, Reliance Jio has made a remarkable impact on...

AMD reiterated the specifications of its Instinct MI300 APU, the world’s first data centre chip with an integrated x86 CPU and GPU architecture, at CES 2023. With 146 billion transistors, the chip is truly at the cutting edge of what modern chipmakers can achieve. It also makes use of the advantageous 3D Chiplet packaging technology, combining TSMC’s 5nm and 6nm IPs in a single unit.

AMD
credit: wccftech

AMD’s upcoming Instinct MI300 APU is a true engineering marvel, combining x86 CPU and GPU cores for the server segment, but the company has already confirmed the next-generation MI400 on its roadmap, as confirmed by Lenovo’s Vice President. It’s no surprise that AMD will continue to use this APU approach for the Instinct family down the road, as Lenovo’s Vice President, Scott Tease, has stated.

The VP confirmed that AMD has Instinct MI400 and other products on the way, and they like what they’ve seen so far.

Scott claims that future AMD Instinct GPUs, such as the MI300 and MI400, will consume more than a Kilowatt of power, a 2x increase over current-gen CPU and GPU designs for servers. This massive increase in power would necessitate the use of more expensive liquid cooling solutions to keep these monster chips cool.

AMD
credit: wccftech

It’s one of the issues raised by the Lenovo VP because, while the new architecture and software supporting it can run 70% of the applications available, the remaining 30% will force users to return to traditional x86 designs. The same is true for NVIDIA’s Hopper and Grace implementations, particularly the Grace CPU Superchip, which employs the Arm architecture.

Also Read:

Source

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.

━ Related News

Featured

━ Latest News

Featured