Intel says its latest Xe2 graphics architecture marks a major step forward in video decoding capabilities, becoming the first to support VVC (Versatile Video Coding) decoding. This innovation propels Intel ahead of competitors like AMD and Nvidia in the video codec space. The graphics are part of the company's new Lunar Lake platform, announced at Computex 2024.
Breaking Down VVC and Its Impact
Versatile Video Coding, also known as H.266, emerged as the successor to HEVC (H.265) in 2020. The codec is designed to offer improved video quality while dramatically reducing file size, achieving up to 50% better compression efficiency than its predecessor. Despite its benefits, VVC adoption has lagged behind the more popular AV1 codec. Intel's endorsement of VVC might shift the industry's focus, potentially increasing its adoption. According to Intel, H.266 can further reduce file size by approximately 10% compared to AV1.
The Xe2 graphics architecture represents a notable enhancement over Intel's previous Xe graphics, integrated into Meteor Lake processors. Intel has reported the new architecture performs 1.5 times faster owing to several architectural improvements. Specifically, in tessellation, Xe2 shows a 20% boost in some areas, with performance gains of up to 12.5 times in certain scenarios.
Competing with AMD's RDNA 3.5
Intel's Xe2 graphics cards face competition from AMD's recent RDNA 3.5 integrated graphics processor, part of the Ryzen AI 300 series. While Xe2 offers 67 TOPS (Tera Operations Per Second) in AI performance, AMD has not yet revealed the AI capabilities of its RDNA 3.5 iGP. The new graphics architecture performance from both companies will be closely observed as these products hit the market.
Intel's inclusion of H.266 (VVC) decoding in Xe2 graphics highlights its commitment to advancing video codec technology, which is particularly vital for efficient video streaming and data compression. The H.266 codec, introduced in 2020, aims to enhance compression efficiency by 50% compared to HEVC/H.265. This increased efficiency translates to reduced storage needs and improved quality in video streaming, supporting high-definition content from 4K to 8K.
Intel Lunar Lake Chips
Lunar Lake made its official debut in April. The company's CEO, Pat Gelsinger, showcased the processor, highlighting its capability to deliver over 100 trillion operations per second (TOPS) in AI workloads. Notably, 45 TOPS of this performance is attributed solely to the Neural Processing Unit (NPU), setting a new benchmark for AI computing in personal computers.
