Head-Fi.org › Forums › Equipment Forums › Sound Science › 128k MP3's versus WAV/Lossless
New Posts  All Forums:Forum Nav:

128k MP3's versus WAV/Lossless - Page 2

post #16 of 73

It really depends. The most extreme example for me is a certain track encoded in 128 kb/s versus 320 kb/s is a night and day difference even on iBuds.

post #17 of 73


 

Quote:
Originally Posted by eucariote View Post

Higher bitrates primarily allow the encoding of waveforms at higher frequencies - and age-related hearing loss usually affects the hearing of higher auditory frequencies, so your hearing may be the critical factor.  Can you hear frequencies over 17kHz?

 

Quote:
Originally Posted by El_Doug View Post

DAMMIT MAN!  You shoulda posted a warning about that - i had my amp volume at 1 o'clock and everything!!! 

 

crap that hurt
 


 

Look on the bright side. You probably can't hear frequencies over 17kHz any more...
 

post #18 of 73


. that was funny

Quote:
Originally Posted by Spelaeus View Post


 

Look on the bright side. You probably can't hear frequencies over 17kHz any more...
 

post #19 of 73

Quote:

Originally Posted by El_Doug View Post

DAMMIT MAN!  You shoulda posted a warning about that - i had my amp volume at 1 o'clock and everything!!! 

 

crap that hurt
 


 

 

Oops, warning posted..    
 

post #20 of 73

LOL! Hope your ears are ok. I can easily hear that. Then again I am only 16  I just tried this: http://www.noiseaddicts.com/2009/03/can-you-hear-this-hearing-test/ and heard up to 21khz. I got nothing at 22khz lol. 

post #21 of 73

Lossy codecs have been improving.  128 kbps from 10 years ago is not the same quality as 128 kbps now.  How audible those changes are to you?  I don't know.

 

One test would be to get ahold of an old version of an mp3 codec and do a listening test comparing it to a modern mp3 codec.  Maybe find an old version of xing and compare it to the current version of LAME.

 

Unfortunately I don't know where to send you to get an old version of a codec.  Some of the sites for things like that can be pretty shady and I wouldn't trust them.  There used to be a site called "really rare wares" that had old versions of codecs.  But the site is now gone.

post #22 of 73
Quote:
Originally Posted by stang View Post

LOL! Hope your ears are ok. I can easily hear that. Then again I am only 16  I just tried this: http://www.noiseaddicts.com/2009/03/can-you-hear-this-hearing-test/ and heard up to 21khz. I got nothing at 22khz lol. 


 

Those tests are MP3 and have a sampling rate of 44.1 kHz.  A 22 kHz tone at a sampling rate of 44.1 kHz is at the theoretical maximum for what what frequency can be recorded and played back.  And that's assuming perfect digital filters on encoding, decoding, and on your DAC.  Perfect digital filters don't exist.  So at 22 kHz or even 21 kHz you're hearing artifacts rather than the actual test tone.

 

If you want to try listening to a 22 kHz sound you should find test tones that are recorded and played back at a higher sampling rate, say 96 kHz. And the test files should be lossless like FLAC rather than anything lossy.

post #23 of 73

I have ringing in my ears 24/7, so hearing 22khz would prove to be extremely difficult anyway, as I would be hearing the ringing, not the actual 22khz tone. I am not sure what caused this ringing of the ears. Whether it is this dental appliance which moves my jaw back (thus affecting my ears/hearing) or because I went to my first concert last December and it was damn loud and my ears rang loudly for several days after. 

post #24 of 73

The folks who have mentioned the nature of the source material are, IMO, getting at the point.  I have done a lot of casual A/Bing of different codecs, and the differences, where they were evident, were clear on tracks that had a lot of dynamic range, and featured acoustic instruments and the human voice.  As to the OPs point - that compression codecs are improving - I have seen or heard this a few times in the last year or so, but have not seen or heard any corresponding stories announcing the development of same.

post #25 of 73

Let's speak science here, since we're in the science forum. 

 

I am the member that challenged the OP to the test. I converted him from a "128Kbps MP3 is crap" believer, and I will try to convince you with this blind test. 

 

Below is the file of 2 wav files. One is lossless, one is 128Kbps mp3, converted to wav. Can you tell which one is which just by listening? (In my other thread someone used multiple analyzers). 

 

http://www.megaupload.com/?d=EUG6Y4RH

 

In your reply, please state what equipment you used, and what difference you hear. 

 

@OP, I'm amazed you can hear a difference by just using Logitech speakers. I have a Logitech Z-5500, and they're not very high quality speakers, certainly a big step below my RE0.


Edited by chengbin - 6/13/10 at 8:17am
post #26 of 73

 

Quote:
 The folks who have mentioned the nature of the source material are, IMO, getting at the point

 

Exactly. When I had done a comparison for myself, the quality of the recording had a lot to do with how much of a difference I was able to discern. The sweet spot for me (i.e. the lowest bit rate I was willing to accept without compromising too much of a loss in quality) was 192kbps. This is with ogg vorbis though which I only use on my DAP while commuting. Otherwise it's always lossless for me. 

post #27 of 73

Yeah, mp3 has gotten a lot more efficient and better. They didn't just randomly stop developing it, or else it would have been overtaken as the standard vs other competitive codecs.

I read about a year ago that trained listeners can hear the difference between mp3 128 and lossless about 50% of the time. Not a difference throughout the whole song, mind you, just one little difference that occurs on a second or so of a track that they managed to track down and test repeatedly with an abx enabled program. This kind of listening is not fun, and very rigorous compared to what you're used to.

There's no doubt that 192 kb/s is completely transparent, and if you've got a need for space, you're not missing much on 128 kb/s if it was encoded in the the last year. There's no doubt certain rare wavs won't transcode to mp3 properly and will be really obvious, but that's just it... it's very rare and you will know when you hear it.

edit: I've gotta say, I've never been more disappointed in head-fi before. They've actually made my avatar an advertisement for some crappy pair of IEMS. Sure, I could change it, but that is really low. Shows how much this site is just about shopping and myths.


Edited by Chef - 6/13/10 at 2:28pm
post #28 of 73
Quote:
Originally Posted by chengbin View Post


Below is the file of 2 wav files. One is lossless, one is 128Kbps mp3, converted to wav. Can you tell which one is which just by listening? (In my other thread someone used multiple analyzers). 

 


That's a test comparing lossless to modern lossy.  The OP is wondering about ~10 year old MP3 encoding vs. modern MP3 encoding.

 

There is no doubt that MP3 encoding has been advancing over the years.  Just check the change log for LAME and you can see they've been regularly updating it for sound quality improvements along with performance improvements.  The question is whether those sound quality improvements have added up to anything significant compared to what was common for mp3 10 years ago.

post #29 of 73

To the OP, there has been significant advances in MP3 encoding. I remember reading a post on hydrogenaudio comparing lossy formats to lossless under ABX test, where the listener rate the songs from 1-5. MP3 10 years ago received like mid 1.x scores. Fast forward to last year, people rate it at around 4.3ish. I forgot what bitrate it was, but it wasn't high. I can't seem to find the page anymore.

 

Here is a page you might find interesting. 

 

http://wiki.hydrogenaudio.org/index.php?title=Listening_Tests


Edited by chengbin - 6/13/10 at 6:48pm
post #30 of 73

ygpm about comparison
 

Quote:
Originally Posted by chengbin View Post

http://www.megaupload.com/?d=EUG6Y4RH

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › 128k MP3's versus WAV/Lossless