Scaling Your Encoding Backend Using Eve
summary
Eve (Two Orioles) VP9 and AV1 encoder claims actual results: at slow presets it delivers about 10–30% BD‑rate gains versus libvpx and matches x265 quality, while its faster presets approach x264=preset=medium / x265=preset=superfast in complexity giving better bitrate/quality tradeoffs. AV1 results show ~20–30% compression gains but at 10x higher encode cost today, with work to reach x264/x265 fast‑preset speeds; memory use stays in the same ballpark and all speeds are reported as single‑thread measurements.
AI-written summary · about
chapters (6)
- 0:00Introduction to video encoding
- 2:15Scalable low complexity encoding
- 3:40VP9 performance in Eve
- 6:34AV1 quality and speed trade-offs
- 9:04Long-term encoding strategies
- 10:29Q&A
related talks
24mAV1 at Netflix
20mBoosting Video Codec With RISC-V Vector Extension
22mAI Enhanced GPU Video Coding: Achieving Joint High Compression Efficiency and Throughput
28mAn overview of AOM: The Alliance for Open Media
12mMeasuring live video quality with minimal complexity, now available for everyone!
21mWhat to do after per-title encoding
16mSwift: Adaptive Video Streaming with Layered Neural Codecs
20mStreaming at 250 kbps: raise your expectations