Log in

View Full Version : Encoding Lossless Audio to E-AC3 5.1


Pages : 1 [2]

j7n
27th May 2026, 19:53
Whether the compression is precalculated to save processing power or done on the fly doesn't matter whe making the decision to apply it. Ffmpeg is a general purpose tool that is often used to transparently convert files to another format without the user knowing it, for example, when loading them into a website. You can only pick compression if you know the playback conditions, perhaps the type of device that will be used.

It's a good thing that this parameter is not rejected for other formats and can exist in conversion preset always.

I heard they have some form of compression in USAC too. But it probably can't be a gain for every block if they want to target low bitrates without bytes to spare.

Balling
2nd June 2026, 09:34
USAC just like TrueHD DRC is not implemented yet.

I mean I personally do not care much but here. https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/22846/files


"Whether the compression is precalculated to save processing power or done on the fly"

It is precalculated when encoding and written into bitstream.
The way Dolby reference encoders work (ffmpeg does not support DRC and is still same quality as Dolby's) is that you can decode DRC compressed stream that is lower volume and takes less CPU processing power as decoding fully the DRC original more lossless sound would take even more computations (and that makes sound really lossy as it no longer the same amplitute and stuff and caused a lot of headaches, which is just advanced loudness normalisation, anyways)