is this lossy? help needed

[vold911]

this is from a silver cd bootleg of an audience taped sabbath show

the frequency drop at 20k instead of 16k and the same in the spectral analysis, which would normally tell me right off the bat its mp3 sourced, has me stumped





if its lossy, whats the reasoning behind that conclusion? same if its not
if it was truly lossless shouldnt it in the spectral view go all the way to the top without any alteration , even above 20k?
can a true lossless show a dropoff at all?

and heres a pic of the spectral analysis zoomed in at 2-3 seconds

[tonebloke]

Not eMPty3. Lossless.

The drop off could be some post work done to the source by the bootlegger. May also be because of the settings on the recording machine.

[Audioarchivist]

How old is the show? Sourced from a cassette tape master I'd think, depending on the mics used and the way it was copied off tape to digital, it'll probably not have a full hot and loud audible spectrum of sound up to 20000hz or more. Doesn't mean it's lossy, though. Looks pretty analog to me... Maybe some digital NR was applied over 20000 to minimize the tape noise?

With a decent spectral display, if you zoom in to about a 2 second range, you don't just look for a 16khz cutoff point, but look for a blocky "lego skyline" where the cutoff point is. It'll be pretty drastic. This one looks more like a gradual fade out at the top, and there's some smooth curves (relatively speaking) not squares and a harsh on/off line where frequencies just stop....

[bertrand]

When you dither 48khz or 96khz to 44.1khz without the white noise (the added white noise is an inaudible mastering technique to smooth the dither) it creates that little drop you see there around 21k. May be more or less noticeable depending on the dither algorithms/amount and type of white noise.