Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: HQ AC3 decoding with FFMPEG (or not ?) (Read 1756 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

HQ AC3 decoding with FFMPEG (or not ?)

I need to squeeze out the maximum possible audio quality out of cameras' (256Kb/48kHz) AC3 audio recordings.

Here's the commandline I wanna use:
Code: [Select]
ffmpeg -drc_scale 0 -i <filename.ac3> -c:a pcm_f32le <filename.wav>

...any suggestion (for example about dithering) ?
...do you think is better to choose foobar instead of ffmpeg ?

Thanks in advance !
Hybrid Multimedia Production Suite will be a platform-indipendent open source suite for advanced audio/video contents production.
Official git: https://forart.it/HyMPS/

Re: HQ AC3 decoding with FFMPEG (or not ?)

Reply #1
for long-term archival, better to keep the original.
who knows, what kind of technology gets invented in the future.
plus it's way more compact that way.

if you have to re-encode (e.g. nontrivial editing other than a coarse trimming), then yes, this makes sense.
any subsequent steps would not be specific to the original format, at that point it's just (floating point) audio. all the general best practices apply.
a fan of AutoEq + Meier Crossfeed

 

Re: HQ AC3 decoding with FFMPEG (or not ?)

Reply #2
Quote
...any suggestion (for example about dithering) ?
Don't dither floating point.    (AC3, like all lossy compression, doesn't store individual samples and there is no actual bit depth.)

You're supposed to dither when you reduce the bit depth

Dither is added noise that's supposed to sound better than quantization noise.    You shouldn't add noise unnecessarily, and you shouldn't dither more than once.      

Under normal conditions you can't hear dither or quantization noise at 16-bits or better, so it's the last thing I worry about.   But it's "good practice" when you downsample.   It's probably a good idea at 8-bits where you CAN hear quantization noise.