Could someone tell me or lead me to a tutorial on the best way to determine if a show someone has burned for me onto CDR is in fact lossless, or has an mp3 generation somewhere in its lineage.

I use True Audio's Tau Analyzer, but I've read that it's not very accurate for live recordings. I can also examine the WAV file using the Wav editor that came with EAC and look at the spectral view, but I'm just not sure what I'm looking for.

I'd really appreciate some help. I've gotten a few shows that are supossedly lossess, but just don't sound very good quality to me. Sound a little "thin" not as full as most I have. If this stuff is from mp3 I don't wanna circulate it further!!

Scott

Are you talking about bootlegs? You may know that not all recordings have to be good, since tapers may use different equipments in different places. So it may be just the source you got.

"Lossless" means a file its identical to it's source. If a show was poor recorded, it will make a poor lossless file. I do have EXCELLENT shows, but some are really just for collectors.

You probably want Tau Analyzer which uses the AuCDtect algorithms.

This software hasn't been updated in awhile, and in my testing, it often produces false negatives, but seems OK when it comes to false positives.

As was said previously, the sound quality of a bootleg or live recording can vary widely. Something like this is only to determine is transcoding (for your own use, for example for portable listening) or sharing would be appropriate.

PS - I found that if you want to run on Windows 2000, you need to install GDI+.

Try these two links for starters:

http://www.audiohub.org/get/fa/fa-eac.htm

http://www.audiohub.org/get/fa/sa.htm

If you're still unsure of your source I'd recommend signing up at thetradersden.org and checking out this forum.

I would worry about spectral analysis alone to identify encoded content.

If no low-pass filtering was used during encoding, looking at the peaks won't help you, but that encoded file is likely to have more artifacts than one that does have low-pass applied.

It still is a valid indication, just not foolproof.

It is, so far as I'm aware, the best solution though.

On the contrary, looking at a zoomed-in spectral analysis (1 or 2 seconds) is the best method of detecting lossy compression and certainly for attempting to differentiate between an MD recorded show and something with a lossless->MPEG stage.

Thanks alot everyone.

I followed the links that ffooky posted. VERY informative. The forum on Trader's Den was especially helpful. Everything you could want to know about analyzing wavs and interpreting the Spectal and Frequency Analyses. I did quite a bit of experimentation last night with many different tracks - original wavs, low quality mp3, high quality mp3, cbr, vbr - and based on those trials it does indeed seem that the best way of determining if there's a lossless compression in the lineage is to examine the SA and FA of a 2-sec segment of the track.

I also found the following to be true (again, in MY trials):

Tau Analyzer (auCDtect) identified the tracks correctly every time - even though I've read lots of posts on how unreliable it is with live shows.

EAC's built-in wav analyzer performed SA and FA very well - even though I've read lots of posts on how it's far inferior to other programs like AnalFreq, Audition/Cool Edit Pro, and Audacity.