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

Lossless vs 128kbps mp3 vs 320kbps mp3 blind test

Discussion in 'Sound Science' started by chewy4, Jan 15, 2013.
First
 
Back
1 2 3 4 5 6 7 8
10 11 12
Next
 
Last
  1. KamijoIsMyHero
    Experience
     
  2. Dark_wizzie
    I've been fooling around with OGG recently. I got to like 75kbps with OGG, and it sounded SHOCKINGLY close to my flac originals. I was blown away, and the size of many of the tracks are 1 mb or under! I didn't know this type of feat would be possible. Definitely something mp3 cannot replicate. Going to that final compression setting for mp3 on Foobar kills the sound quality. All of a sudden now I have more than enough space for all of my tracks in my crappy cell phone. MP3's not really "the best" for audio quality at any bitrate/file size AFAIK, so why test with mp3 besides the fact that it's recognizable?
     
  3. Head Injury
    Precisely this reason. You want to test using a format people are familiar with, otherwise they either won't learn anything from the test or they'll just have another "flaw" to pick at ("Sure I couldn't tell the difference between these files, but the differences between my FLACs and 320 kbps MP3s are like night and day!")
     
    So in most cases this means testing with MP3s (the most widely available and used format, for better or worse) or AAC (iTunes, for worse [​IMG]). Or, for personal tests like yours, whichever format you want to use!
     
    I'm thinking of re-encoding my portable songs to mid-level OGG files to save space, because while it's still plenty transparent to me I don't feel comfortable going below V2 with LAME.
     
  4. CharlesC
    If I said that I can tell the difference between 256k aac and alac what I would mean is that sometimes I can tell the difference. Doesn't a test like this require you to be able to tell the difference consistently to pass? It seems that even if you can only tell the difference 10% of the time then its worth it to rip in a lossless format.
     
  5. CharlesC

    One more thing - anyone with a $2000 mp3 collection is also going to be biased in that they want to believe that their files are just as good as lossless files. Blind testing does nothing to mitigate that bias, so any statistics derived from this will necessarily be pessimistic.
     
  6. skyd171
    That's assuming anyone participating actually has a $2000 mp3, or very expensive, mp3 collection. I think it's safe your concern is mitigated by the fact that most here are audiophiles and want to experience the best sound, and would want to hear a difference. I personally did. But as mentioned I have a $3k+ setup, SE846's with audioquest victoria interconnects,dac, amp, tube amp, etc etc.... I did several listens and actually guessed that the 128kbps was the lossless and lossless was 128. I fancy myself an audiophile and have been a musician since my teen years.
     
  7. CharlesC

    If someone starts off as a teenager spending $200 a year on mp3/aac files then ten years later they have a $2000 collection.  If at that time they become interested in improving their sound quality and invest in better listening equipment, it's not that hard for me to imagine that they have a bias toward hoping their collection is top notch and that they don't have to think about repurchasing music.  This doesn't seem terribly farfetched to me. My point is simply that there is an assumption here that only the lossless lovers have bias but it seems wrong to suppose that the lossy lovers don't have any.
     
    Personally,  I'm very confused by the whole thing.  Mark me down as one who does terribly on AB testing but continues to believe there is a difference. True, my confidence is not all it was before but it has lead me to think that there may be some problems with AB testing - such as: 1) you should only test music where you already think you can hear a difference - what's the point of testing anything else, why should you be expected to hear a difference on a piece of music where you never claimed to hear any difference in the first place?  2) listening to a small sample of music over and over again may well blind you to subtelties - I don't think you can listen to something 6 times in a row and expect to hear anything useful.  It's like food:  the first time you taste something it may be spectacular, but each time you eat the same thing over and over it loses its special quality - and not because the food has changed.  We're talking about a subjective experience of an intangible phenomena. 
     
  8. luberconn
     
    i actually didn't get very many of them right.  i pretty much just mess with 320kbps via google play all access.  good enough for this guy.
     
  9. Dark_wizzie
    I can't successfully abx 320kbps MP3 vs FLAC but I'm still storing my music in FLAC because I can. Life is good.
     
  10. Brooko Contributor
     
    Don't worry - this is very common among the community here.  Despite the actual evidence when we test properly, there are still many people who want to believe there is still an audible difference. I guess some of it is because of the large numbers of people on this site who claim lossless as superior - without actually testing properly.  Crowd/peer think.  The desire to conform.
     
    My advice - if you are more comfortable with lossless - go lossless.  BUT ...... when threads come up comparing the two, or asking if there is a difference, simply refrain from posting. One of the biggest issues remains that people continue to perpetuate a myth (audible difference for proper high bit-rate lossy transcodes) despite evidence to the contrary.  The difference between Head-Fi and sites like HydrogenAudio is that over there  - the expectation that differences are audible is extremely low - and burden on those that claim to hear differences must be proven.  Over here - anecdotal evidence runs rife sadly.
     
    Ultimately though - listen in the format that makes you happy.  Just be careful with any claims to others.
     
  11. Tempo
    The 1st set is the easiest one. Perhaps, it is because the most of instruments used in this song are acoustic.
    The 3rd set is the a little bit harder.But by comparing the very end of this song, You can tell the difference easily wih various perfomance at the high pitch.
    The 2nd setis the most difficult one to me. Maybe it is  because everything is electronics. By comparing 3 files from 0:14 to 0:25, I can only say one of them generate a firmer sound. However, I think I find out the pattern of the answers.[​IMG]
     
    1. b>a>c
    3. c>b>a           
    2. a>c>b
    see! "randomly scrambled"!!
     
  12. itszvz
    Lol, I can't tell a difference between any of them. I tried group 3.
     
  13. shanky
    damn i mixed up group 3's 320 and lossless. got the rest though ! woohoo!
     
  14. bzippy
    just found this older thread. i tried testing with foobar2000 w/ ABX Comparator plugin and i admit i can't hear any difference at all. i didn't even bother completing the tests because i knew i would just be guessing.
     
    so is it my equipment? my ears? is it simply not possible to hear the difference? i do have mild tinnitus so maybe that's the issue? well whatever the deal is, i may have wasted my time and memory re-ripping all my music lossless.
     
    my rig:
    Asus Xonar DX sound card (Cirrus-Logic CS4398 DAC) in HiFi mode (i.e. all processing off) LINE OUT >> FiiO E12 >> Sennheiser HD600
     
    maybe next i'll load them onto my FiiO X3 and see if that changes anything.
     
  15. krismusic Contributor
    I apologise if this is off topic. I am thinking that the fact that I have Tinnitus effectively excludes me from making audiophile judgements. The interaction between equipment and my Tinnitus is likely to be a greater factor in my perception than other subtleties of sound.
     
First
 
Back
1 2 3 4 5 6 7 8
10 11 12
Next
 
Last

Share This Page