RAV1E: The "Fastest & Safest" AV1 Encoder

[chithanh]

https://www.servethehome.com/cavium-...rver-option/5/

I you look at the results when a GCC -Ofast compiled version of the SPECrate benchmark is used, the ARM based system beats both the Xeon Gold and the EPYC system but when that same benchmark is compiled with Intel's ICC compiler for the Xeon and AMD's AOCC compiler for the EPYC, the AMD system wins handily.

As AOCC is based on LLVM/Clang, I wonder why they didn't test that. It would have gotten a good baseline on Epyc vs. ThunderX2, plus giving some insight on AMD AOCC optimizations for SPECrate2017.
Also STH claim that the gcc results "are likely more representative of many workloads that use gcc as their compiler", but they use -Ofast??
-Ofast will enable standards violating compiler optimizations (such as algebraic optimizations) and is surely not representative of typical workloads.

[fuzz]

This hand-optimized assembly discussion reminded me about this talk: CppCon 2016: Jason Turner “Rich Code for Tiny Computers: A Simple Commodore 64 Game in C++17”

[mulenmar]

Nice thing about VP9 when doing TV-series is that you can encode several episodes in parallel and thus speed things up cosiderably. I did the first season of Buffy (the 1080p HD remake) yesterday, which is 13 episoded @ about 45 minutes each.

Using the following settings:
ffmpeg -c:v libvpx-vp9 -speed 2 -crf 27 -b:v 0 -tile-columns 4 -frame-parallel 1 -threads 8 -c:a libopus -ac 2 -b:a 128k
And running two episodes in parallel @ about 8.5fps per episode, I managed to finish in less than 12 hours on my Ryzen 1700X. CPU load was about 50% so I could have probably run four episodes in parallel if I wanted to speed things up further.

This took it down from ~45GiB to ~10GiB. Quality is still alright although I did notice a slight decline.

It can be painfully slow when encoding feature lenght movies though.

Thanks, that's quite helpful. 12 hours is still a pain, but at least I can set it to run while I'm at work. 480p DVD video probably won't take so long, either.