[FFmpeg-user] FFPLAY plays some audio files at their native amplitude while others are playing 3dB down in amplitude

Rick Keniuk daddyo323 at gmail.com
Tue Mar 10 03:51:34 EET 2020


You found it Ted...  FFPLAY interprets a mono (single channel WAV) file 
to be played at -3dB from full scale while the other programs set the 
mono file at full scale on both stereo channels. I simply modified my 
Matlab code to generate a stereo file with identical channels (from the 
original single channel data) and that fixed it.  I guess they are 
"power dividing" the single channel signal.  I don't think I agree with 
FFPLAY's interpretation of the amplitude of a single channel WAV file 
but that is what is happening.  Maybe there is a configuration to fix 
this effect.

Thanks for the help,

Rick

On 3/9/2020 4:05 PM, Ted Park wrote:
> Hi,
>
>> I have a sine wave audio test file generated with Matlab in FLAC and also WAV formats which load into several audio applications at full scale amplitude (16-bit sine wave peaks at +32767 and -32767).  The other applications play this file at maximum full scale amplitude yet FFPLAY plays the file at 3dB down from maximum amplitude.  Converting the file with FFMPEG to other formats results in an identical amplitude file which is full scale also. Other programs play this file at maximum amplitude but FFPLAY again is at -3dB from full scale.  Is there a command that I need to issue to FFPLAY to make it play at the native volume?
>>
>> Here is my command:
>>
>> C:\ffmpeg\bin\ffplay -volume 100 Sine_65s_997Hz at FS_44.1k_16b_int16.flac
>>
>> The sine wave FLAC file is 1.7MB and I can provide that if that would be useful.
>>
>> Odd thing here is that I have tried several song files, and even live audio streams in FFPLAY, and they play at the correct full scale amplitude.
> Where are you measuring the level to be 3dB lower than normal? You might have a mono file that is being panned left and right each at half power, which is sqrt(/ about -3dB gain. So to get a 0dB total, you see the 3dB attenuation per channel. Does this sound plausible?
>
> Regards,
> Ted Park
>
> _______________________________________________
> ffmpeg-user mailing list
> ffmpeg-user at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-user
>
> To unsubscribe, visit link above, or email
> ffmpeg-user-request at ffmpeg.org with subject "unsubscribe".


More information about the ffmpeg-user mailing list