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

Poll: Audible Difference between FLAC and 320kbps MP3?

Discussion in 'Sound Science' started by el_doug, Aug 10, 2009.
First
 
Back
1 2 3 4 5 6 7 8 9
11 12 13 14 15 16 17
Next
 
Last
  1. K_19
    I do feel that it matters, but mostly only with more revealing rigs and equipments... it wasn't as noticeable until I got my current Pico -> WA6 ->K701/HD650/RS1 rig, where the separation between the two rates has become very evident IMO. Before with my old Hotaudio DAC -> EF1 I did not find that there were too many differences at all.

    Also, I'm sure this was mentioned already in this thread (I didn't read through it all), but the quality of original recording plays a big factor. For example, I've had instances where I've had a decent LAME 256kb mp3 beat out my Oasis (what's the story monring glory) FLACs in perceived quality and overall enjoyment... as many of you might now that particular recording, as well as most of Oasis albums, were mastered very poorly. So there are many other factors to consider in the overall equation.
     
  2. gbacic
    It also matters how they are incoded, you can have terrible encoded 320 and really good 192 or 256 (such as itunes AAC) and the lower bitrate will win.
    It's just safer to go with FLAC 'cause you know it will be CD quality and you won't have to worry about it being a crappy rip.
     
  3. tarkovsky7
    I don't even have great audio equipment, just shure 420 buds and AKG 240 headphones, and I can tell the difference between 320 ACC and Apple Lossless every single time on either set of headphones. On the 320 ACC, speaking in non-technical term, all the ambience seems to have been removed: the echo-gone, the careful layering of instruments-gone, the bass losses its punch.
    Interestingly enough, I find when i listen to 320 AAC I have to turn the volume up to a high level to get any type of visceral reaction from the music; I can listen to lossless at a low volume and get a visceral reaction from the music.
    I find it suprising that on a hi-fi audio forum that most people can't tell a difference with equipment that is superior to mine. I think the differences in our perception must have to do with how we listen to music and experience it; the people who can't tell a difference I suspect are just listening to all the instruments, and sure they sound clear and seperated but don't notice the "studio production qualities" of the music.
     
  4. PiSkyHiFi
    Quote:

    Originally Posted by tarkovsky7 /img/forum/go_quote.gif
    I don't even have great audio equipment, just shure 420 buds and AKG 240 headphones, and I can tell the difference between 320 ACC and Apple Lossless every single time on either set of headphones. On the 320 ACC, speaking in non-technical term, all the ambience seems to have been removed: the echo-gone, the careful layering of instruments-gone, the bass losses its punch.
    Interestingly enough, I find when i listen to 320 AAC I have to turn the volume up to a high level to get any type of visceral reaction from the music; I can listen to lossless at a low volume and get a visceral reaction from the music.
    I find it suprising that on a hi-fi audio forum that most people can't tell a difference with equipment that is superior to mine. I think the differences in our perception must have to do with how we listen to music and experience it; the people who can't tell a difference I suspect are just listening to all the instruments, and sure they sound clear and seperated but don't notice the "studio production qualities" of the music.




    After my efforts here to clarify what is mathematically and physically feasible, I am surprised by this comment. Quite simply, I don't believe you.

    Its possible you have unique ears that don't fit the psychoacoustic models used in audio compression - hard to tell what that really means for your hearing though.

    To say things like all the ambiance has been removed suggests one just has to listen for missing ambiance - I have and found it present in both MP3 and lossless.

    The bass loses its punch ? I can show that sometimes the bass will gain punch as it has variance either way.

    I hear floors in "Studio Production Quality" of music most of the time - its a fact of life when listening to modern recordings that they are crafted and all reveal their makings in high rate MP3, AAC and lossless.

    I believe its possible to actually quantify this stuff properly using math, I have good felling for it, maybe its about time I sat down and did the correct comparison without regard to unique ears and see if I can state my case with objectivity. I hope I can find the time for this.

    Enjoy the music,

    Corley.

    P.S. I am keen to hear your source for this comparison, if there's a way you can upload a 10-second byte in both AAC and lossless to somewhere - hey I'll even make a space for you - just let me know, I would appreciate hearing the source that reveals.
     
  5. tarkovsky7
    In a side by side A/B experiment, I can tell the difference between 320 ACC and Apple Lossless after 20 seconds well over 90 percent of the time. My previous claim that all ambience is lost in 320 ACC is obvious hyberbole; but, a lot of ambience is lost-enough for me to notice almost every single time. Not only is ambience lost, but the lossless also sounds much more "stable"; maybe, there is a slight variances in volume of certain frequencies or position changing in certain frequences in a 320 ACC, or simply artifacts that i hear. This is especially true of the low and high frequencies.
    I don't claim to have "golden ears," in fact, I'd be shocked if i dont have some hearing loss; i'm almost positive it's the different ways we listen to music.
     
  6. PiSkyHiFi
    Quote:

    Originally Posted by tarkovsky7 /img/forum/go_quote.gif
    In a side by side A/B experiment, I can tell the difference between 320 ACC and Apple Lossless after 20 seconds well over 90 percent of the time. My previous claim that all ambience is lost in 320 ACC is obvious hyberbole; but, a lot of ambience is lost-enough for me to notice almost every single time. Not only is ambience lost, but the lossless also sounds much more "stable"; maybe, there is a slight variances in volume of certain frequencies or position changing in certain frequences in a 320 ACC, or simply artifacts that i hear. This is especially true of the low and high frequencies.
    I don't claim to have "golden ears," in fact, I'd be shocked if i dont have some hearing loss; i'm almost positive it's the different ways we listen to music.




    I propose a test, would you be willing to take part ?

    1. The test will be 2 samples of the same section of music, 1 will be straight FLAC encoded PCM, the other will be converted to 320 Kbps AAC and then back to FLAC. Both samples will be recreated using a filter that will use double precision floating point to change the base sample rate and its amplitude to a pre-determined different sample rate and volume somewhere within 2% of the original encoded sample - the same rate and volume change will be applied to both samples - this is just make it difficult to compare the data to an already known lossless copy of the audio.

    2. I will provide a download location for these 2 samples A and B.

    3. Subject to do ABX testing using the 2 samples, or if they are dishonest, mathematical analysis.

    4. Let me know which one you think is the lossless.

    5. Maybe I'll make a few samples of different music styles this way, then once the test is done, I can open source the code that made the samples for others to scrutinize and we can see how accurate our ABX testers do in a more objective test.

    What do you think ? I'll even take requests.

    I think you may make a good test subject for this.

    Looks like the sample length will be around 20 seconds and there will be an approximately 1 second fade in/fade out.

    Corley.
     
  7. tarkovsky7
    Sure, I'd be willing to particpate in this test; just let me know when and were you post the samples. If i can, i will be willing to give a detailed and specific list of difference between the two samples.
     
  8. PiSkyHiFi
    Quote:

    Originally Posted by tarkovsky7 /img/forum/go_quote.gif
    Sure, I'd be willing to participate in this test; just let me know when and were you post the samples. If i can, i will be willing to give a detailed and specific list of difference between the two samples.



    I will produce some samples that cannot be determined as either lossy or lossless using FFT software so that the test can only be verified by the code that created the samples.

    You will have to be a little clearer on music choices - I am looking for up to 10 samples, if you don't specify, I will choose my own and try to be diverse.

    If you pass the test, your detailed and specific list will carry much meaning.

    I'll need some time to prepare the work - thanks for being a test subject.

    Corley.
     
  9. the search never ends
    I also had to go with (after much time)..........many modern recordings forced mr to
    choose that answer. However well masterd stuff is not hard usually.
     
  10. tarkovsky7
    Maybe it would be best to limit your samples to rock and hip-hop since thats pretty much all i listen too.

    Thanks,

    Andrew
     
  11. tarkovsky7
    On second thought, maybe it would be best to make the samples as diverse as possible.
     
  12. anetode
    Quote:

    Originally Posted by tarkovsky7 /img/forum/go_quote.gif
    I don't even have great audio equipment, just shure 420 buds and AKG 240 headphones, and I can tell the difference between 320 ACC and Apple Lossless every single time on either set of headphones. On the 320 ACC, speaking in non-technical term, all the ambience seems to have been removed: the echo-gone, the careful layering of instruments-gone, the bass losses its punch.
    Interestingly enough, I find when i listen to 320 AAC I have to turn the volume up to a high level to get any type of visceral reaction from the music; I can listen to lossless at a low volume and get a visceral reaction from the music.
    I find it suprising that on a hi-fi audio forum that most people can't tell a difference with equipment that is superior to mine. I think the differences in our perception must have to do with how we listen to music and experience it; the people who can't tell a difference I suspect are just listening to all the instruments, and sure they sound clear and seperated but don't notice the "studio production qualities" of the music.




    Focusing in on individual instruments, the overall soundstage and things like that requires both perceptual and cognitive function. After years of critical listening your brain has developed unique ways of interacting with your sensory nerves and it processes auditory information through different "circuits" depending on your listening preferences at the time. It would indeed be surprising for you to find out that you may be hearing the same waveform in different ways on different occasions [​IMG]
     
  13. MD1032 Contributor
    I would be highly interested to see the results of that test, but it absolutely must use more than one file, otherwise you could simply "guess" correctly. I would suggest using more like 20 files of the same piece of music and that the subject does not have on hand. That's like the minimum...if it were me, it would be done in person because you could always measure the differences.
     
  14. googleborg
    I tested a few tracks using foobar abx plugin, can tell the difference only after abxing them, and only on high frequency things like hihats and cymbal crashes, the rest sounded identical to my hears though my hifi. I still rip to lossless though [​IMG]cd: :p
     
  15. Skylab Contributor
    I actually don't care whether I can hear a difference between lossless and lossy - actually ripping CD's in with a lossy codec is a bad idea, IMO, for the following reasons:

    1. The fact is that lossy coding DOES remove musical information - whether audible or not
    2. Disc space is now DIRT cheap, so there is no reason to use lossy coding when ripping CD's - if you rip using lossless, you can easily transcode to lossy later if you need to for portable use
    3. Archiving in lossless means that if a new, super-great lossy codec comes along later (which it will), you don't need to re-rip to take advantage of that - you just transcode from the lossless files you have.

    So bottom line - I find the debate interesting, but only in the context of what kind of files to put on your portable. For ripping, using anything other than lossless is silly.
     
    moj0 likes this.
First
 
Back
1 2 3 4 5 6 7 8 9
11 12 13 14 15 16 17
Next
 
Last

Share This Page