white iPod earphones THD

Discussion in 'Pro Audio' started by Kevin T, Aug 11, 2006.

  1. Kevin T

    Kevin T Guest

    The vast majority of music is currently being listend to is MP3 over
    iPods with standard earbuds. At some ref level/Freq ( IEC 60268-7 )
    etc. What is the THD of this defacto standard at the listeners ear ?

    Thanks
    Kevin T
     
    Kevin T, Aug 11, 2006
    #1
    1. Advertisements

  2. Kevin T

    GregS Guest

    Ipod check.
    I don't have one.

    greg
     
    GregS, Aug 11, 2006
    #2
    1. Advertisements

  3. what the ****
    yo man this is not cheap hardware inc.
    google it or ask apple or
    try another more appropriate usernet group
     
    audioaesthetic, Aug 11, 2006
    #3
  4. Kevin T

    Scott Dorsey Guest

    The bad news is that it's still sadly relevant to us because too much of
    our material is being distributed this way.

    The worse news is that THD is a useless measure for describing a lossy
    compression system. We don't really _have_ any good metrics that correlate
    with audibility that work with lossy compression systems, which is where
    the problem begins.
    --scott
     
    Scott Dorsey, Aug 12, 2006
    #4
  5. I have a personal metric that I stole from nancy raygun

    Just Say No.... (to using crap storage/playback for music)

    didn't even use a cassette deck for personal pleasure, let alone mp3 or
    apple acc.

    but then I am in the minority

    watching the lemmings run to the cliff!!!
     
    audioaesthetic, Aug 12, 2006
    #5
  6. Just Say No... (to using crap storage/playback for music)

    When the Windows Media Player was amenable to actually working, I sometimes
    listened to KUSC with 96kbps MP3 encoding. I was flabbergasted at just how
    good it sounded.

    But, yes, the stuff on my jukebox is uncompressed WAV files. When you've got
    enough storage space for 30+ CDs, why compress?
     
    William Sommerwerck, Aug 12, 2006
    #6
  7. Kevin T

    John Noll Guest


    The newer 30 and 60 Gig iPods have enough disk space that you can upload
    44.1 WAVs of AIFFs. I've got my iPod loaded with only 44.1 files and
    listen with nice Sennheisers. The stock earbuds have never been removed
    from the package.

    A pretty cool setup that sounds really nice.

    I refuse to listen to MP3s.
     
    John Noll, Aug 12, 2006
    #7
  8. Kevin T

    Arny Krueger Guest

    Measured THD at the listener's ear is probably pretty low. IME IEMs tend to
    be fairly low in distortion at the modest levels that most modern consumer
    gear drives them. IPODs themselves measure pretty clean.

    http://www.stereophile.com/mediaservers/934/index5.html

    So they can be pretty good if you use them to play losslessly compressed
    media.
    Excellent and highly germaine point.
    AFAIK, true to this day. There seem to be some pretenders, though. The good
    news is that lot of portable music players will play losslessly compressed
    music, and some will even record it on the fly.

    Also, if you use the right procedures and keep the bitrates high enough, it
    can be hard to hear the difference with many recordings.
     
    Arny Krueger, Aug 12, 2006
    #8
  9. I'll disagree greatly with the testing methodology on that, if you look at
    not just the pictures, but the comments on how to get the graphs you'll see
    things like "Fig.1 Apple iPod, frequency response at -12dBFS into 100k ohms
    (right channel dashed, 0.5dB/vertical div.)." Since variations from usage
    scenario will change things, often dramatically, it be vastly more
    appropriate to actually connect them to the 16 Ohms that is normal for
    consumer earphones, or some other nominal load point that it will actually
    see. Preferably it should be mapped against a wide range of resistances.

    I personally found the iPod DAC heavily flawed when listening to it, but
    that was a couple of years ago, it is entirely possible they have changed it
    since then. The amp did not seem to be driven in a linear fashion distorting
    the sound enormously. This listening was done on a good pair of Sennheisers
    so, while they certainly added their own flaws, the flaws experienced were
    not the result of listening on cheap 'phones.
    Joe
     
    Joseph Ashwood, Aug 13, 2006
    #9
  10. Kevin T

    Arny Krueger Guest

    No way, Jose'.

    A good perceptual coder removes information any which way it can get away
    with, which includes significant spectral and amplitude changes, including
    allowing addition of what amounts to being noise.
    More to the point - listening tests involving trained listeners.
    Indeed. Speech is usually the easier source to code with reasonable
    transparency.
     
    Arny Krueger, Aug 13, 2006
    #10
  11. Kevin T

    Arny Krueger Guest

    How does one audition just the DAC in one of these devices?
    Depending which Sennheisers you mean, they have relatively poor voltage
    sensitivity compared to the devices usually used with iPods.
     
    Arny Krueger, Aug 13, 2006
    #11
  12. Kevin T

    Scott Dorsey Guest

    Sure, but it's dependant on the source material. And it doesn't tell you
    anything about audibility.
    My personal feeling is the voice quality measures aren't much good either...
    --scott
     
    Scott Dorsey, Aug 13, 2006
    #12
  13. I did it by making use of known inputs and known behavior, so I used
    uncompressed files ripped from some very familiar jazz CDs (mostly because
    that's what I wanted to listen to), and headphones that I have a great deal
    of experience with it. This placed the varibles within the device itself.
    There were of course other variables involved, the amp, the power supply,
    etc, but these are effectively inseperable from the DAC for this purpose.
    I mostly chose that pair because I have long term experience with them
    plugged into various devices both good and bad, and they have become my
    basic reference headphones. I use them because I am familiar with their
    flaws, not because they are unflawed.
    Joe.
     
    Joseph Ashwood, Aug 13, 2006
    #13
  14. Kevin T

    Daniel Fuchs Guest

    The word is usenet.
     
    Daniel Fuchs, Aug 13, 2006
    #14

  15. tanxs 3% harmonic distorted
     
    audioaesthetic, Aug 13, 2006
    #15
  16. Kevin T

    Scott Dorsey Guest

    There are a bunch of different issues. Is intelligibility all that matters?
    Do you need to be able to identify the caller by their voice? Is tonality
    important (it is if you're speaking in Cantonese)? A system that gives good
    intelligibility in one language may not be appropriate for people speaking
    a different language. This then leads to compatibility issues.... and then
    we get to the re-encoding artifact issues...
    --scott
     
    Scott Dorsey, Aug 13, 2006
    #16
  17. Kevin T

    Kevin T Guest

    Thanks Scott & Arny for keeping this at the level I intended. Unlike
    some I know the fact I dont like MP3 lossy coders is somewhat
    irrelevant in this new world of P2P & PAPs. I work in audio for a major
    cellphone manufaturer that finds itself in the music player biz and the
    benchmark weather we like it or not is iPods. If you had to answer the
    questions from marketing "Are we doing as well as the competion? "
    "Why does customer "X" say our sound is poor"? How would you do it?

    Kevin T
     
    Kevin T, Aug 14, 2006
    #17
  18. Kevin T

    Scott Dorsey Guest

    You can do a couple of things. First, you can have listening panels.
    It can actually be pretty inexpensive to do if you have a nearby college
    and can shanghai some students. And you can probably get a paper out of
    it, which from your company's view is free advertising too.

    Secondly, you can look for research done with existing listening panels.
    If you're using a known codec, you can make the assumption that most of
    the sonic coloration is the result of the codec and ignore the rest,
    thereby applying any other listening tests done with that codec to your
    device. The manufacturer of your codec will give you more propaganda than
    you can shake a stick at too.

    Thirdly, you can ignore the problem. You can say, "Customer X says our
    sound is poor, because it's poor." In the case of an application where
    sound quality isn't really paramount, that's possible. You can also say,
    "our bit rate into the codec is lower than the bit rate into the codec
    of the iPod, and therefore the best we can do will be sonically inferior
    to the iPod." Of course, if your bit rate is higher and the results are
    still sonically inferior, then you have a problem.

    But good sonics don't sell products anyway, and your primary audience isn't
    people who are terribly concerned about sound quality. This is very
    depressing from my perspective, but it's not anything new.
    --scott
     
    Scott Dorsey, Aug 14, 2006
    #18
  19. Kevin T

    Kevin T Guest

     
    Kevin T, Aug 14, 2006
    #19
  20. Kevin T

    Mark Guest

    Mark, Aug 14, 2006
    #20
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.