Using Single-Pass Look-Ahead in Modern Codecs for Optimized Transcoding Deployment

Vibhoothi Vibhoothi, Julien Zouein, François Pitié, Anil Kokaram

Modern video encoders have evolved into sophisticated pieces of software in which various coding tools interact. In the past, single-pass encoding was not considered for video-on-demand (VOD) use cases. In this work, we evaluate production-ready encoders for H.264 (x264), HEVC (x265), AV1(SVT-AV1) along with direct comparisons to the latest AV1 encoder inside NVIDIA GPUs (40 series), and AWS Mediaconvert's AV1 implementation. Our experimental results demonstrate that single-pass encoding inside modern encoder implementations can give us excellent quality at a reasonable compute cost. The results are presented as three scenarios targeting high, medium, and low complexity accounting quality-bitrate-compute load. Finally, a set of recommendations is presented for end-users to help decide which encoder/preset combination might be more suited to their use case.

Print ISSN
Electronic ISSN
2160-2492
Published
2024-10
Content type
Original Research
Keywords
linear streaming, http adaptive streaming (has), dynamic adaptive streaming over http (dash), av1
DOI
10.5594/JMI.2024/WSSG3140