Head-Fi.org › Forums › Equipment Forums › Sound Science › Double blind test 128Kbps vs lossless? I'll be amazed if you can tell much difference
New Posts  All Forums:Forum Nav:

Double blind test 128Kbps vs lossless? I'll be amazed if you can tell much difference - Page 8

post #106 of 257

Using my im616 and straight from my 2006 MBP headphone jack, I can tell 10 out of 10 times which is the FLAC.

 

METHODOLOGY-I used vlc, then clicked randomly on the skip track button with my eyes closed. After AB-ing several times, I can tell which one is cleaner sounding. I only listen to the first 10 seconds or so, and often just the first chord. The first chord has the tendency to sound jagged on the 1st track. After I've made my decision, I open my eyes and then check if I'm right. With the Etys, I chose the same track as the 'cleaner' track every time.

 

I do NOT have golden ears at all. In fact, when I tried with my fairly newly acquired RE0, I couldn't hear a difference between the two. Not a single difference, and that's considerin I already knew what t look for based on my experience with the im616.

 

I'm chalking this down to amazing isolation with the ety's and the fact that there is a noisy airconditioner near my laptop.

 

Though I've recently been enjoying the RE0 more than my im616's, perhaps the fact that I've lived with the im616 daily and for so long means I 'hear' them better. 

 

Maybe something about BAs vs. Dynamics and the way they pick up detail has something to do with it?

 

Wow. I love this hobby! I learn something new everyday. I didn't think I'd be able to tell the difference with ANY headphone. I have a lot of my music ripped from years ago on a lower bitrate, Am glad I started ripping to higher bitrates when hard drive space got cheaper. :)

post #107 of 257

Of course, I ASSUME I'm choosing the better track. Otherwise, my ears are actually preferring the lower bitrates to the FLAC! That would be a little embarrassing, but also pretty liberating.

 

How do I find out the right answer?

post #108 of 257
Quote:
Originally Posted by Pio2001 View Post




The ripper, no, exept if you include the encoder in it. Then yes, there are big quality differences between the mp3 encoders.

 

The track also is very important. The same kind of encoding can lead to transparent results of a given track, and unlistenable results on another.

 

Depeche Mode - Something to do is a track that I found very easy to ABX.

 

That's what I was wondering. I used WMP to rip the track I ABXed, and like I said I got 9/13. I ripped one as a 128 kbps MP3 and one as a WMA Lossless, converted them both to wav and then ABXed them. 
 

post #109 of 257
Thread Starter 
Quote:
Originally Posted by AtomikPi View Post




I think better yet would be posting a v0 or v2 since there's really no need to compress all the way down to v5. With those settings, I doubt anyone will pass the DBT except in the case of a killer sample. Those also naturally have higher low pass settings, around 19.5 khz for v0, so that would be corrected. And as mentioned above, I would just stick to the V setting and not try to micromanage LAME.


I chose V5, or 128Kbps, because this test was meant to challenge people's perception of 128Kbps are crap. Using V2 or V0 would become VERY difficult to tell, as shown as people are struggling to tell a V5 and lossless. The people who do tell the difference are mostly using pretty expensive gear (notably Currawong >2K setup)


Edited by chengbin - 7/11/10 at 7:57pm
post #110 of 257


 

Quote:
Originally Posted by chengbin View Post




I chose V5, or 128Kbps, because this test was meant to challenge people's perception of 128Kbps are crap. Using V2 or V0 would become VERY difficult to tell, as shown as people are struggling to tell a V5 and lossless. The people who do tell the difference are mostly using pretty expensive gear (notably Currawong >2K setup)


Or a $40 setup. *cough*

 

Though I'm sure a more expensive setup would reveal the differences a bit better. And on the lower price range, isolation is really key. I could tell the difference with my FiiO E5 > Head Direct RE2 but probably wouldn't be able to on my Alessandro MS-1i just because of the influence of background noise.


Edited by Spelaeus - 7/11/10 at 8:11pm
post #111 of 257

I wonder if FR makes a big difference, what range are most of these artifacts in anyways? I'm thinking a treble emphasis would make them really 'glaring' ... Off course, I might be off the ball since encoding is not my area. 


Edited by Ypoknons - 7/11/10 at 9:30pm
post #112 of 257
Quote:
Originally Posted by Spelaeus View Post


Or a $40 setup. *cough*

 


And young ears.  My ears are over 40 and don't hear the high frequencies (like 16K) so well any more.  I'm not even sure how I'm hearing a difference between the files if the main difference is a cutoff at 16K.

 

I assume you've got younger ears than me.  The difference you hear in the files could be more pronounced for you than the very subtle difference that I'm hearing.

post #113 of 257


 

Quote:
Originally Posted by Ham Sandwich View Post




And young ears.  My ears are over 40 and don't hear the high frequencies (like 16K) so well any more.  I'm not even sure how I'm hearing a difference between the files if the main difference is a cutoff at 16K.

 

I assume you've got younger ears than me.  The difference you hear in the files could be more pronounced for you than the very subtle difference that I'm hearing.

Fair enough. I'm 19 and capable of hearing up to at least 18k-19k. After that, I'm not sure if I'm unable to hear it or my relatively cheap equipment is unable to reproduce it, since I don't think I have any equipment with a frequency response higher than 20k. Not that I'd actually need it for anything other than testing my hearing, so...

post #114 of 257

My hearing stops around 15 kHz, and I can hear the differences.

 

Treble emphasis might indeed turn the differences more audible. the mp3 psychoacoustic model is based on flat frequency response. If the frequency response is not flat, artifacts can become audible. Thus, bad gear might reveal differences hidden by good gear.

 

post #115 of 257

I normally can't tell the difference but still rip to flac for archival purposes anyway. That way I can encode to lossy as needed for different uses without fear of extra loss via 2nd generation lossy encode :)

post #116 of 257
Quote:
Originally Posted by Nebby View Post

I normally can't tell the difference but still rip to flac for archival purposes anyway. That way I can encode to lossy as needed for different uses without fear of extra loss via 2nd generation lossy encode :)


That's always a great idea.  That's why I never pay for lossy music.  Maybe I can't hear the differences now, but what about after a future upgrade?  Always plan for the future.

post #117 of 257
Quote:
Originally Posted by maverickronin View Post


That's always a great idea.  That's why I never pay for lossy music.  Maybe I can't hear the differences now, but what about after a future upgrade?  Always plan for the future.


Most definitely. And even if lossy codecs do continue to improve as they have, lossless will always be lossless.

post #118 of 257

 

Quote:
Originally Posted by Pio2001 View Post

My hearing stops around 15 kHz, and I can hear the differences.

 

Treble emphasis might indeed turn the differences more audible. the mp3 psychoacoustic model is based on flat frequency response. If the frequency response is not flat, artifacts can become audible. Thus, bad gear might reveal differences hidden by good gear.

 


That's interesting about the psychoacoustic models being based on flat frequency response.  I hadn't though of or considered that as a factor.  No headphone is flat, and some much less flat than others.  It would be funny if a cheap bad headphone is better able to hear artifacts and differences because the cheap bad headphone does more damage to the psychoacoustic model used for the compression.  I had always assumed that good headphones with more clarity (less color, better transients, clean layers of sound, etc) would be better all around for listening tests.

 

I wonder if careful tweaking with a parametric EQ could also bring out lossy compression artifacts due to making the frequency response non-flat and breaking the psychoacoustic model assumptions?

post #119 of 257
Quote:
Originally Posted by Ham Sandwich View Post

 


It would be funny if a cheap bad headphone is better able to hear artifacts and differences because the cheap bad headphone does more damage to the psychoacoustic model used for the compression.  I had always assumed that good headphones with more clarity (less color, better transients, clean layers of sound, etc) would be better all around for listening tests.

 

I wonder if careful tweaking with a parametric EQ could also bring out lossy compression artifacts due to making the frequency response non-flat and breaking the psychoacoustic model assumptions?



In one of the early large scale mp3 vs CD blind tests the far and away best listener in terms of discrimination had a hearing defect that (he reported) allowed him to hear mp3 artifacts more easily

post #120 of 257

they sound the same to me

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Double blind test 128Kbps vs lossless? I'll be amazed if you can tell much difference