Head-Fi.org › Forums › Equipment Forums › Sound Science › Blind Test - Take Now!
New Posts  All Forums:Forum Nav:

Blind Test - Take Now! - Page 5

post #61 of 88

This will get interesting now that we are at higher bitrates. For some reason these tests always seem to be conducted at 128 kbps or some rate where it has already been demonstrated that some can hear a difference. Much more interesting is testing at the higher rates most of us use in the real world, and where things become increasingly difficult to the point that the samples become essentially transparent even to the experts (unless you cheat by looking at a frequency plot that is ;-)

post #62 of 88

from OP A is the lossless one, the piano has more reverberance..

post #63 of 88

A is lossy actually, really obvious from the lack of high treble

post #64 of 88

I read far enough that the OP had revealed the answer, but the second set has A sounding better in the bells, vocals and drums near the end.

post #65 of 88

When I heard it for the first time, I was convinced that B is lossless. The more I was listening, the harder was to tell which is which. But when I focused at some parts of song, I started to be convinced that B is lossless. My setup was very simple, just soundblaster audigy se soundcard, diy mosfet class a amp with akg k530ltd headphones. No "real hifi".

 

It was very hard, but I think that it was because of chosen song. I've experienced many times artifacts in mp3 files, so I would never ripped my CDs into mp3. No matter how much space I would save.

post #66 of 88

After about 5 seconds of listening i'm already ready to post my conclusions:

1. I feel there is a difference between the 2 files (i always consider the possibility the OP tries to fool everyone by posting identical files)

2. To me B is higher quality. (hear more details right away) 

 

(If i'm right) how did i figure it out so easily and quickly? : 

i'm a gamer - when you play games against other people you quickly learn to prefer headphones over speakers

using headhpones was the trick for me

 

my setup: US $25 Creative audigy 2 card (My first X-fi died on me, 2nd one on the way)  Sennheiser hd280 about US  $99 (closed & w/o "bass boost" - good for gamers) 

Also i enabled highest quality settings for the audigy 2 card 24bit, 96, and trying out foobar2000 audio player - (usually use winamp with 24bit enabled in preferences)

 

Is it a huge difference that any kind of majority would care about - i would have to say no - i used to watch people listening to videos and music on youtube back when all videos on youtube had 64kbps audio that was (to me) tiring to listen to

Is the difference enough for me to care? Yes.  

post #67 of 88

Considering the first post;

 

Took me a while but I got it right. I'm surprised there is that small of an audible difference. I think a huge factor here is the fact that this music is a real live recording. I am used to listening to a lot of synthetic stuff, or extreme walls of sound (ambient, or artists that include it all the time like Devin Townsend). Seems like it is much easier to encode real things as this is how MP3 is designed (all psychoacoustics afaik) and synthetic sounds will be prone to all kinds of artifacts when run through those functions.

 

I could not tell, by listening to one, and then the other, fully, over and over. But I did not try enough. All of the information got through, the notes, the timing, the stage. What let me do it, is ABXing in foobar, and picking tiny .5 to 1 second intervals, and using the keyboard with eyes closed to repeat the sound back and forth between three letters. For example, A, B, then C, and seeing if A was the same as C or B was the same as C, ignoring D. Some passages let me do it 5/5 tries. It was clear, some echos, tones, certain things I can't name or recall were present or not muddy. It was mostly when all of the instruments were sounding at the same time. Others I couldn't tell apart at all.

 

For sure, there is benefit between even 320/V0 and lossless, audible or not. Certain albums will have too much information that you are not supposed to hear. The information makes it into the ear like a digital signal of electrical impulses and may influence the brain directly, bypassing conscious analysis or hearing. It wouldn't raise thoughts by association but by programming. I bet if during a certain emotional state one were to play a unique piece of noise, and replay it another time to the same listeners, at least some may enter that emotional state again. If the noise was encoded however, it wouldn't be the same "digital" data (pattern of zero crossings on the eardrum maybe) and wouldn't do it. As much information as possible always makes the difference even if it isn't conscious, and that's why vinyl is loved for the way it makes you feeeel, not how it sounds. Good vibrations

post #68 of 88

Just took the test 12:30am EST lol ~_~ pretty interesting and thanks for this little experiment.

I used the ABX Foobar2000 for ABX'ing and here are my results:

 

.Capture.JPG

 

It was pretty difficult trying to ABX this track because of the various instruments and fast-paced, dynamic nature of the song IMHO (and I'm a little bit tired lol). I don't recall whether it was A or B that was the lossless (Foobar doesn't tell you I don't believe), but there definitely was a difference, at least to my ears.

 

The difference was in decay of the rapid thump sounds and their slight tonal differences. I'm sure when people isolate a short sample and focus on an instrument they can tell the difference.

For the ABX, I just plugged my SE425s in the front I/O panel of my computer lol. Nothing special at all.

 

Thanks for the little experiment, but most of all, thanks for sharing such an interesting piece of music to me ~ biggrin.gif

post #69 of 88

Hey can you do another one of these please?  gs1000.gif

post #70 of 88

I can't get the download links to work.. any suggestions?

post #71 of 88

I am posting before reading the other post.  I was listening from the digital out on my motherboard to my old Pioneer AV and a DT 990/600.  To me the B recording has just a tad more attack so I'm picking it.  Close call, and I proberly got it wrong. dt880smile.png

post #72 of 88
Answer has likely already been posted but I've not yet looked through the pages of posting yet. At anyrate I find that B is the lossless version and A is the lossy version. Its easy to tell when listening to the reverb and echo of the instruments. Non the less I am impressed at how closely lossy follows behind on this song.

Just saw results on page 3 and I was right haha. This was a lot of fun it would be fun to try again with maybe a higher quality mp3? Around 256? Thanks for doing this it was fun.
Edited by Gatepc - 4/23/11 at 7:50pm
post #73 of 88

the sound stage and instrument positioning are different. in the beginning, when the second guitar starts, in version A, the sound stage is more compact and closed while B, their is greater separation and a larger sound stage. 

post #74 of 88
First, I didn't read any of the comments before making my decision.

In answer to the OP's set, I could easily and obviously tell the difference that B was the lossless. It just sounded more dynamic. And FYI, I was using VLC Media Player on Win 7 -> Total Bithead - > ATH-M50.

In using the same setup for the second set, I found that A sounded better than B, but this set was much harder to tell. Post the answer?
post #75 of 88

Wow, this makes me glad I have all my music in lossless I instantly noticed that B sounded much better, it is hard to explain why, but for one there was a large amount more detail and dynamics in the guitar at the start.

I actually only listened to the first few seconds of each track before I was 100% certain B was lossless.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Blind Test - Take Now!