SDRPlusPlus
SDRPlusPlus copied to clipboard
Negative frequencies from .WAV-file
I used the file_source.dll to play an old .wav-file recording (32-bit, from 2009). I was surprised that it showed the spectrum mirrored:

Comparing with SDR-Sharp v1772, I see no negative frequencies:

Is this working as designed?
the "negative" frequencies is because it can't find the center frequency in the filename so it just default to zero.
For the mirroring however I'm not sure what's going on without looking a the wav file, could you upload it somewhere?
Here: http://www.watt-32.net/dx/PR_090226-0558_01200kHz_000_055900_060030.wav (720 MByte!!)
Edit: Hang on. The upload failed. It's ready now.
This file looks nothing like the first screenshot on my install
I'll check the wav file itself, it's behaving a bit strangely
Ok found, the issue, looks like old SDR# versions added a bunch of junk to the beginning of the file, removing it lets it play correctly

This file looks nothing like the first screenshot on my install
Try zooming a bit. The file I uploaded is the same .wav-file in this PR.
looks like old SDR# versions added a bunch of junk to the beginning of the file
You mean that this PR_090226-0558_01200kHz_000_055900_060030.wav from 2009 was recorded with SDR-Sharp?
Hard to believe since the 1st SDR-Sharp was released around 2013.
Oh alright I thought it was SDR# Well whatever recorded it added some metadata in front of the audio data which causes it to read the sames wrong
Well whatever recorded it added some metadata in front of the audio data which causes it to read the sames wrong
Okay. But like I wrote, SDR# v 1.0.0.1630 has no problems with it. I just did this:
- loaded the above: http://www.watt-32.net/dx/PR_090226-0558_01200kHz_000_055900_060030.wav 700 MByte file into SDR#.
- tuned to 1060 kHz.
- noted the voice says: ... Classic Country AM 1060....
No doubt it's the correct frequency and that SDR# handles this metadata (whatever that is) correct. Can you explain what this meta-data is and how SDR++ (mis)handles it?
Non standard RIFF sections. I'm not planning on supporting those. Has to be plain wav at input.
Met the same thing, and then reached here. Just for the information, one way to remove these metadata is via:
ffmpeg -i input.wav -fflags +bitexact -flags:v +bitexact -flags:a +bitexact output.wav
Then SDR++ will handle it correctly.