TECHNICAL
GPU alone (GPU), and for a fully loaded system with simultaneous encoding on the CPU and GPU. Optimizer’s efficient implementation running on Intel hardware reduces bit rates, allowing jobs to scale easily without impacting the number of streams that can be processed.
Intel Data Center GPU Flex 140 discrete accelerator. 12 The input test suite is the same as used in the Netflix Technolgy Blog,5 using an average of results across all content types for the top-rung results. Data was collected for software (CPU) and Quick Sync Video hardware (GPU) versions of the AVC, HEVC and AV1 encoders. The results (Table 1) show the bitrate and VMAF score for encoding using FFmpeg only and FFmpeg with Optimizer. The results reveal that bitrate is reduced by at least 15 percent and up to 40 percent, while VMAF scores generally improve or decrease by negligible amounts. 12 Table 2 shows the results for the number of maximum parallel encoding sessions that can run in real-time on the “Tears of Steel” 1080p video clip on the Intel- based server without and with VisualOn Optimizer. Data is shown for the encoder running on the CPU alone (CPU), on the
Comparative Results: Other CAE Solutions
Compared to results from other CAE per- title encoding solutions from the Netflix Technology Blog,5 VisualOn Optimizer delivers much lower average bitrate (Figure 4) and higher average VMAF (Figure 5). 11 Comparative Results: 4th Gen Intel Xeon Scalable Processors and Intel Data Center GPU Flex Series with Intel Quick Sync Video The following benchmarks show comparative results of CPU- and GPU- based encoding running without and with VisualOn Optimizer averaged across different content categories over all Adaptive Bit Rate (ABR) ladder rungs, and for the top rung of an ABR ladder. The benchmark was completed on a dual-socket server running 4th Gen Intel Xeon Scalable processors with the
Summary
Video encoding technologies and solutions have evolved rapidly over the last few years with CAE paving the way for more efficient content delivery. CAE uses AI to analyze content on a frame- by-frame and scene-by-scene basis and adapt encoder parameters, resulting in dramatically reduced bitrates while enhancing visual quality. Intel is working with the ecosystem that supports media delivery companies to address their challenges, using advanced Intel technologies, processors, and GPUs.
Table 1: Encoding results without and with VisualOn Optimizer on 4th Gen Intel Xeon Scalable processors (CPU) and on the Intel Data Center GPU Flex 140 with Quick Sync Video (GPU). 12
Table 2:. Maximum number of real-time parallel encoding sessions on 1080p content without and with VisualOn Optimizer. 12
94
Volume 46 No.4 DECEMBER 2024
Made with FlippingBook - Online magazine maker