How to know if a FLAC is a real FLAC?

Discussion in 'Audio Hardware' started by mickael57280, Aug 7, 2017.

  1. mickael57280

    mickael57280 Member Thread Starter

    Ah thank you so much, I think the best way for checking lots of FLAC is first check with Audio Checker, and if for some files it's a doubt to check with a spectral analyser.

    Sometimes it's hard to know like when you have a solo of conga, no high frequency at all.
  2. stetsonic

    stetsonic Forum Resident

    Yep, but when somebody re-encodes that mp3 into a flac, re-uploads it and somebody else grabs it, turns it again into an mp3 which is later turned into yet another flac and uploaded and then some, at some point the lossy generations start to affect the sound of even the worst field recordings. It's like the cassette days all over again.

    So I kind of understand the point of those "don't convert to mp3" clauses although it's pretty much a futile effort.
    shaboo and Tim Müller like this.
  3. Tim Müller

    Tim Müller Forum Resident

    Yes. :agree:

    With digital, it all could be that easy: Create the master mix for an album in the digital domain, and keep this forever.
    No problems with detoriation of analog tapes, no hassles with bad vinyl masterings or bad vinyl pressings, or dull sounding pre-recorded cassettes.

    Just copy (more: clone) this digital master mix onto the various formats, and always in every territory and on every formats have the same sound quality: The original sound quality of the recording from the day it was mixed and mastered.

    But in reality, for each re-release and re-issue on different format or different territory, someone tinkers with the master mix to "re-master" it. Or, orders a new transfer from worn-out analog "master" tapes. Only, to apply a lot of compression and EQ and other tinkerings...

    Best regards
    stetsonic likes this.
  4. head_unit

    head_unit Forum Resident

    Los Angeles CA USA
    Mmmm. I thought/assumed the posters simply wanted to preserve the highest quality, which in some instances just seems absurd. But I understand what you're saying-though that problem to me is that nobody should ever be encoding MP3 into FLAC, under penalty of listening to high resolution audio ruined by watermarking.
  5. One of the ways that happens is by people burning lossy files to a CD (as CD Audio). Then someone rips it. That's how some of the lossy sourced audience recordings were getting passed around. It makes for a mess.
    head_unit and stetsonic like this.
  6. Claude Benshaul

    Claude Benshaul Forum Resident

    There are instances where the source is so bad that it doesn't really make any difference if the FLAC was originally encoded in MP3. example: The album by Odessa. I first tried it on Tidal, got frustrated when listening to the distortions and tried to find a lossless version. There is one to download in FLAC but I listened to the sample and it sounds just as bad. Maybe it's an up-sampled MP3, maybe it's not, I suspect that it's just a case of a bad recording and mastering.

    Anyway, It got me thinking about HDTracks. I'm pretty confident they don't re-encode inMP3 sources in FLAC but I'm not so sure about other paid download services.
  7. mickael57280

    mickael57280 Member Thread Starter

    Hi I have encountered an another problem, audio checker tell me a file is sure at 100% it's a mpeg, but when I open it in Speak the spectrum go up to 22khz but at around 21khz it's a horizontal green line through all the song.

    What is that?
  8. brimuchmuze

    brimuchmuze Forum Resident

    Likely just some artifact of the equipment in the recording chain, etc.
  9. mickael57280

    mickael57280 Member Thread Starter

    Not necessary a lossy file?
  10. Carl Swanson

    Carl Swanson Resident blabbermouth

    I use AC too, but it doesn't really tell you the original format, it just deals in "probablity". Still handy.
  11. davmar77

    davmar77 I'd rather be drummin'...

    clifton park,ny
    that's why cool edit/adobe audition and audacity are good. you can see the analysis. look at this comparison.

    lossY analysis AUDIO - Bing images
  12. Carl Swanson

    Carl Swanson Resident blabbermouth

    No, you just need to reconcile yourself to the fact that, like every other aspect of life, there are sometimes problems that have no complete, reliable solution.
    gregorya and shaboo like this.
  13. Janet

    Janet Forum Resident

    United states
    Bandcamp does not upsample or alter the audio files that the artist uploads. If there is a 16-44 resolution claim for a song which turns out to be from an mp3 it's the fault of the person who submitted it. Bandcamp allows artists to exert full control over their material at all times, even after it has been published. Some artists publish high resolution FLACS and don't even advertise that aspect. Bandcamp utilizes a unique and sane business model...and I don't work for them...
    Robert C likes this.
  14. I'm not blaming Bandcamp for those lossy source files. I like Bandcamp. It's the artist or label that is uploading the files that is responsible for the quality of the source files.

    My request of Bandcamp would be to be transparent about what the source resolution is. Especially if it's high-res. Right now you just don't know what you're going to get after you hit the Buy button. I've been surprised a few times and ended up with high-res files when I was expecting it to just be 16/44.1. The only clue was that the download size was really big compared to what a 16/44.1 download would be. But until you buy it, download it and check the bit depth and sampling rate there is no way to know what the file actually is.
  15. gregorya

    gregorya Forum Resident

    Well, that's no way to drive yourself to a nervous breakdown... ;)
  16. anorak2

    anorak2 Forum Resident

    Berlin, Germany
    I understand your question to mean, was the FLAC converted from a CD or other PCM source, or some mediocre format such as MP3?

    There will be differences, but depending on the quality of the MP3 they might be so subtle as not to be audible.

    The MP3 format does not require a frequency cutoff. While many decoders apply one by default, that behaviour can be switched off. It's not inherent to the format. Therefore a spectrum analyzer will not definitively tell if a file was converted from MP3.

    Listen to them. Listen if you can detect MP3 artefacts, such as the typical gargling noise, or short sharp noises such as percussion sounding washed out.

    Other than that you could get hold of the original track e.g. from a CD rip and do a substraction in an audio editor. If they sum to zero, they are identical. If you hear weird artefacts instead, you have some distortion in your FLAC possibly from MP3. Someone did that here:

  17. eeglug

    eeglug Forum Resident

    Chicago, IL, USA
    Here's a case of CDs apparently being sourced from lossy files: the 2013 Edsel reissues of Robert Palmer's catalog. You can read the discussion in this thread starting from Post #13. Post #168 links to an article from comparing spectrograms of Edsel vs original Island cds which show the frequency cutoff for this particular case.
  18. Carl Swanson

    Carl Swanson Resident blabbermouth

    Better take a cab. No collateral damage.
    gregorya likes this.
  19. As fate would have it for participating in this thread. I bought a CDR from a busker a week ago. Just listened to it today. Good music. Well recorded. Professionally recorded in a good studio in Nashville with good supporting musicians. DR numbers for the tracks are 11-13. More compression and EQ on the vocal than I would like, but otherwise very well done. Unfortunately, it's likely a lossy source. I listened to the first track and was impressed with the music and recording, but my audiophile ears perked up and told me this could be lossy. Ran the files through Trader's Little Helper and it says 95% likely MPEG source. Spectrogram shows at 16 kHz haircut. Ears tell me the recording is lacking in breath and breadth (which for me is an indicator of lossy compression). Bummer.

    Still, it's good music and a good recording. Just a bummer that it's only available as a lossy source. The only way to get the music is to buy it from her at a show or while she's busking. Even knowing all this, I'd still buy it again in the same circumstances. It's good singer songwriter style music, and otherwise a good recording.
  20. Grant

    Grant A 60s, 70s & 90s Lovin' Musical Free-Spirit


    You mean lossless ve. not lossless.

    If a file's content, not spikes, goes up to 20,000, it is probably lossless. Even so, such a file should have density within the frequency range.
  21. Grant

    Grant A 60s, 70s & 90s Lovin' Musical Free-Spirit

    Even that can fool you. Sound quality alone is not an indication of the quality of a file.
  22. brimuchmuze

    brimuchmuze Forum Resident

    No, it would not indicate lossy.

    What is the track?
  23. Janet

    Janet Forum Resident

    United states
    I agree that full transparency is what the customer deserves, and the artists who upsample lossy files to FLAC on bandcamp or anywhere else give the entire practice a bad name.
    Cockroach likes this.
  24. junclj

    junclj New Member

    It's hard to determine with a lossy iTunes Plus m4a with tool such as Spek.
    Above is a 264kbps iTunes Plus lossy m4a and below is a FLAC. So how should I know if I convert the m4a to a fake FLAC?


  25. Try using TAU True Audio Checker and Lossless Audio Checker to test the files in addition to visually checking with Spek. Well done AAC encodes from high-res source can fool these tools. Similar to how it is very hard to tell from a spectrogram whether well done AAC looks lossy. The tools are useful, but be prepared for false negatives and false positives.
    JimmyCool likes this.

Share This Page