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: MP3 48kHz vs 44.1kHz encoding waveform true peak level differences (Read 400 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

MP3 48kHz vs 44.1kHz encoding waveform true peak level differences

I noticed that if you encode to MP3 at 320kbps CBR using 44.1kHz, the resulting waveform is altered adding level to the peaks causing a higher true peak than the original lossless file, but if you resample the 44.1kHz lossless file to 48kHz using Audacity's resampler before the MP3 conversion, the resulting file has more consistent peaks that do not go above the original peak levels of the lossless file.

Here is an example:

44100.wav peaks at -10dB
after converting it to mp3,
44100.mp3 peaks at -9.075dB

so after encoding at 44.1kHz, there was a +0.925dB true peak boost, almost a whole dB

but

48000.wav peaks at -10dB
and after converting it to mp3
48000.mp3 peaks at -10.12dB

which means that at 48kHz the conversion caused the signal to be attenuated by 0.12dB

I do not know whether the signal at 48kHz is always lower in true peak level than the original signal, the total opposite of 44.1kHz which is if not always, almost every time higher in true peak level than the original signal after the MP3 conversion, but of the little times that I have converted at 48kHz the result has always been a lower true peak level than the original signal.

I do not know how this works, if MP3 is designed to perform "better" at 48kHz based on the consistency of the resulting peaks or if the resampler of Audacity is the cause of this but I find this beneficial as I do want consistent true peak levels below -10dB because that is as high as I want to play the files to prevent distortion and this trick helps achieve it without having to adjust the level of the lossless file which at 44.1kHz I would have to do the conversion many times using different true peak values below -10dB to compensate for the true peak level boost that is always very variable, it can go as little as some fractions of dB to up to 1.5dB above and with this trick I know that it will always stay a few dB below the original 48kHz signal, at least that is what has happened so far.


[Moderator] Please keep music clips under 30 seconds

Re: MP3 48kHz vs 44.1kHz encoding waveform true peak level differences

Reply #1 –
The 48kHz file is 1.3 dB quieter. Seems like your resampling does more than resampling.
 

Re: MP3 48kHz vs 44.1kHz encoding waveform true peak level differences

Reply #2 –
When I resample your -10 dBFS 44100.wav file to 48 kHz I get a result with -8.68 dBFS peak. Looks like you then normalized this back to -10 dBFS, so the content in 48000.wav is 1.3 dB quieter than in 44100.wav.