Blog

Jan 08, 2024

Apple completes its ARM lineup with 5nm, 134billion transistor ...

As previously suggested, the M2 Ultra uses Apple’s UltraFusion interconnect and

As previously suggested, the M2 Ultra uses Apple's UltraFusion interconnect and silicon interposer to combine two of the M2 Max die with a unified memory architecture. This has 10,000 signals, providing over 2.5TB/s of low-latency interprocessor bandwidth to support up to 192GB of memory capacity, 50 percent more than M1 Ultra, with 800GB/s of memory bandwidth, twice that of M2 Max.

The move to the latest 5nm TSMC process has boosted the CPU clock by 20% over the M1 Ultra with 16 next-generation high-performance cores and eight next-generation high-efficiency cores, alongside a larger GPU with 60 or 76 cores and a 32 core neural engine. The neural engine has a performance of 31.6TOPS, up 40% on the M1 ultra.

The engine has twice the capabilities of M2 Max, further accelerating video processing. It has dedicated, hardware-enabled H.264, HEVC, and ProRes encode and decode, allowing M2 Ultra to play back up to 22 streams of 8K ProRes 422 video. The display engine supports up to six Pro Display XDRs, driving more than 100 million pixels.

"M2 Ultra delivers astonishing performance and capabilities for our pro users’ most demanding workflows, while maintaining Apple silicon's industry-leading power efficiency," said Johny Srouji, Apple's senior vice president of Hardware Technologies. "With huge performance gains in the CPU, GPU, and Neural Engine, combined with massive memory bandwidth in a single SoC, M2 Ultra is the world's most powerful chip ever created for a personal computer."

www.apple.com