1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

128k MP3's versus WAV/Lossless

Discussion in 'Sound Science' started by anaxilus, Jun 12, 2010.
1 2
4 5
  1. snoopy007
    128Kbit mp3 from ten years ago do sound a lot worse than 128kbit encoded (latest lame) today. ( yes I still have some files from over ten years ago)
    actually even quite a lot of 256/320kbit cbr mp3 from ten years ago doesn't sound as good as the latest 192 vbr v2 lame encoded files now. Most of the improvement is probably in the fine tuning of the pyscho acoustic model used + a better use of "joint stereo" technique to improve overall quality of the mp3 files.
    Bottomline, I think we should be thankful for lame becoming the predominant encoder + the use of standardised templates for the improvement in mp3. 10 years ago mp3 were still being created using myrid of different mp3 encoders and our own estimate of what is the best settings to use for each bitrate.
    Personally for me, lame vbr v0 or v2 is more than adequate for portable use. As for home use, I will stick to flac files since hard drives are so cheap now.
    Mp3 has improved since the "early" days when having large collections WAV or lossless compressions was just fond dreams but lossy is lossy and has no place in a quality playback system...IMO of course.
    If everything starts with your media then why mess with it at all?
  3. Anaxilus


    I agree wholeheartedly.  Whether I can hear the difference or not, you don't want to compromise your source.  That's like building a house on quicksand.  If I can't tell the difference, is it my ears?  Is it the phones?  Is it the player or amp?  Does it even matter??  What does matter is that I have all the data from the original available.  That means lossless or WAV period.  Takes up too much room you say?  That's what sd cards are for.  Sorry Apple go @#$% yourself.  I go to CES every year and noway am I gonna audition new audio gear w/ 128k mp3s.  I should go to the McIntosh rep and ask if I could though just to see his face, would be classic. 
  4. j57park
    I think also type of songs you are listening to makes a difference.
    For example, classical music between 320kbps and FLAC is completely noticeable I find
  5. Anaxilus


    Yes slower music that hasn't been compressed in the mastering process is easier to differentiate.
  6. Shike
    MP3 is beneficial in encoding time, doesn't require being converted to work with the majority of players, and also offers space savings.
    Even though space is cheap not everyone wants to deal with the first two . . . not to mention I don't think managing two of the same collection would be highly efficient.
    Actually, I'd go with say a 192CBR file and don't tell them or let them see it's lossy.  If they start talking to you about the sound quality of the system tell them it is lossy.  Watch reaction for lols.
    If they notice I'd be stunned - assuming the music isn't a "killer track" as mentioned over at hydrogenaudio (in other words, easy to DBT).
  7. Publius
    When I participated in the last HA 128kbps test, I was able to reliably distinguish all of the encodes, sometimes finding several different artifacts in the same file. But they all tended to be *quite* subtle. I listened at a fairly elevated volume and it was pretty likely I would not have caught the differences at normal/quietish volumes.
    It says a lot about that MP3 test that the Helix MP3 encoder received a score that was statistically indistinguishable from LAME, even though Helix (at least at the time) had many more known quality regressions than LAME did. What that means is that, while artifacts do exist and will occur, they are infrequent enough to not be of a significant concern for "most" music. That's a pretty surprising conclusion to make.
    Encoding bugs can still cut 128kbps encodes down to size - QuickTime AAC had one or two really fierce bugs for a while IIRC - but when the encoder is well tuned and the bugs are ironed out, and most importantly when VBR is used (and used well), the results are very, very satisfactory, even for MP3. 
  8. fark
    Since I´m new at this audio stuff even tho I´m 34 years of age, can someone please tell me what I´m doing wrong at this site:
    http://www.freemosquitoringtones.org/           ???
    Today was the first time I ever saw a site like this one so I just had to try it out, cause I´m curious of course!  [​IMG]
    I don´t understand what should happen, well of course I understand I´m suppose to hear a signal and that I do, but it also says I´m not supposed to hear many of them since I´m 34, right?    So why can I hear them all?  Please, someone tell me what I´m doing wrong here?   The last one that says 22khz I´m not actually sure I´m hearing, I honestly think I do some of the times but it is so hard that I not sure if I´m lying to my self?  The 21khz one I do hear so pherhaps I´m not testing the right way here, am I supposed to have some fixed volyme of some sort?  Maybe I´m not supposed to use headphones or something like that but only speakers? 
    Regarding mp3 vs lossless I dont know to be honest?   I really must try that as well cause I´m so curious be nature, haha!
  9. Pianist
    IMO There's no way anyone can tell the difference between 320 kbps and lossless. If you think you can, you are just fooling yourself.
  10. krmathis Contributor

    Ok, then your opinion differ from quite a lot of us. But that is fine!
  11. firsthand
    The quality of mp3 is the same as the past.
    The lower the quality will reduce the high-frequency response of your song.
    When you want to compare the quality, try listening to high-frequency instrument like cymbals.
  12. haloxt
    I like how everyone is just contradicting each other, reminds me of a dinner party when I visited China, with my half deaf grandfather talking bad about democratic governments, and half deaf grand uncle talking bad about communist governments, both at the same table.
  13. Spelaeus


    I'd agree that a 320kb/s mp3 does sound pretty decent. And I'd certainly find it pretty damn difficult if not impossible to tell the difference on most recordings just based on sound quality. But lossy is still lossy, and you do still end up with some digital artifacts.
    Mind you, I do find 320kb/s acceptable for listening on my portable device. And given the tremendous size difference between 320kb/s mp3 and lossless wav, I'd probably use 320kb/s exclusively if it weren't for the peace of mind offered by having files which are exact duplicates of the original and will not degrade over multiple generations of rips. But it's still certainly possible to tell the difference if one takes the time to listen for digital artifacts.
  14. Shike

    Except the majority of the time when it isn't.  Very few tracks are able able to be blind tested successfully as LAME has improved that much as an encoder.
  15. Ypoknons Contributor
1 2
4 5

Share This Page