| Author
|
320 vs. wav
|
distantcontact
Started Topics :
4
Posts :
79
Posted : Mar 30, 2009 05:20
|
Thanks for all the responses everyone!
Quote:
|
On 2009-03-30 03:41, RK9 wrote:
Obviously the signal is different and there will be an auible difference signal. That has no bearing on whether someone can tell the difference between the two in a normal situation.
|
|
+1...what I was really trying to get at was - is there is a noticeable audible difference between a 320 and a wav or a wav and a flac on a large sound system?
Thanks again and keep the responses coming!  |
|
|
dj_Asura
Started Topics :
1
Posts :
20
Posted : Mar 30, 2009 05:57
|
Quote:
|
On my alesis mk2 monitors I can't hear any specific audible difference between the two (I conducted a blind test) although i did feel the wav track was a tad punchier in the bass.
|
|
Exactly... If you go to 192 or lower it sounds like crap, but at 320 you only notice some decrease on low frequencies... |
|
|
|
Kitnam
Mantik
Started Topics :
110
Posts :
1151
Posted : Mar 30, 2009 09:20
|
Quote:
|
On 2009-03-30 03:41, RK9 wrote:
Obviously the signal is different and there will be an auible difference signal. That has no bearing on whether someone can tell the difference between the two in a normal situation.
|
|
it is exactly the only true way that proove the facts (besided subtible non-objective A/B-Testing) based on the mathematics of waveforms, because sound is a waveform, allways.
dj_asura
i call this an urban legend.
just keep it up guys
|
|
|
|
Spindrift
Spindrift
Started Topics :
33
Posts :
1560
Posted : Mar 30, 2009 12:04
|
Quote:
|
On 2009-03-30 09:20, Kitnam wrote:
it is exactly the only true way that proove the facts (besided subtible non-objective A/B-Testing) based on the mathematics of waveforms, because sound is a waveform, allways.
|
|
You obviously have not understood the principles behind perceptual encoding since your test proves nothing about quality of the audio.
The point is not to produce an identical waveform, but one that sounds identical to the original.
And that is not the same thing due to the way our hearing works, so that the signals doesn't phase cancel has no relevance when discussing the perceived quality.
The only way to determine how well an algorithm works is through blind listening test. It is after all the established scientific way to determine perceived audio quality. Calling ABX-tests non-objective is missing the point...audio quality is subjective and cannot be determined mathematically or by comparing waveforms.
Instead of doing a phase cancellation test, stick your wavs in WinABX. Then please post your logs here.
I'm still to see a log of someone consistently telling 320kbps from wav, and even at 192kbps VBR it's pretty rare.
 (``·.¸(``·.¸(``·.¸¸.·`´)¸.·`´)¸.·`´)
« .....www.ResonantEarth.com..... »
(¸.·`´(¸.·`´(¸.·`´``·.¸)``·.¸)``·.¸)
http://www.myspace.com/spindriftsounds
http://www.myspace.com/resonantearth |
|
|
|
Kitnam
Mantik
Started Topics :
110
Posts :
1151
Posted : Mar 30, 2009 12:28
|
ok - seems like i am pretty alone here with my opinion.
when the point is "not to produce an identical waveform, but one that sounds identical to the original", then this means that this can only be done with an identical waveform, in case we speak in a range of 20 to 20000 hz.
2 different waveforms within this range will sound different, this is a not negable mathematical fact. the point is most people on most soundsystems wouldnt get it, and it also depends a little bit on the material.
with the tool of layering inverted phases you have the possibility to extract the difference between 2 waveforms. i was thinking that for some people it would be more interesting to hear this exact definable difference instead of just trying to A/B Test which is like looking for the needle in a dark room.
so i am giving an alternative here, which also shows why it is so hard to get the difference, because this phasing-results are very very quiet. meaning only some overtones are beeing missed, probable most wont be able to tell any difference.
keep it up!
"You obviously have not understood the principles behind perceptual encoding since your test proves nothing about quality of the audio."
what else is giving the quality of audio besides the waveform. because the waveform is the audio is the waveform is the audio.
anyway then show me please. i am always willing to learn.
|
|
|
|
Spindrift
Spindrift
Started Topics :
33
Posts :
1560
Posted : Mar 30, 2009 13:00
|
What happens with perceptual encoding is that it removes parts of the signal we don't hear.
For example it uses masking, which means that a loud part of the signal will make it impossible to hear another softer part of the signal in nearby parts of the spectrum.
So just because audio is within our audible range doesn't have to mean we actually hear it.
Of course the waveform is the audio and hence determine the audio quality, and with a technical analysis obviously mp3 produces an incorrect waveform.
But you cannot determine the perceived quality of audio simply by analysing or comparing waveforms....blind tests is the only way to do that.
(``·.¸(``·.¸(``·.¸¸.·`´)¸.·`´)¸.·`´)
« .....www.ResonantEarth.com..... »
(¸.·`´(¸.·`´(¸.·`´``·.¸)``·.¸)``·.¸)
http://www.myspace.com/spindriftsounds
http://www.myspace.com/resonantearth |
|
|
|
x-rayz
IsraTrance Junior Member
Started Topics :
11
Posts :
576
Posted : Mar 30, 2009 14:11
|
last night I spoked with some guy who recently started producing in FL.. I told him about FLAC format and how he can convert it from wav. But he didnt knew to export from FL into wav, only to mp3.. So I told him to export to wav and guess what was his reaction.. Wow, I can hear the difference!
I think that concludes this series of tests about the quality..
http://www.facebook.com/xrayzproductions
http://www.myspace.com/xrayzproductions |
|
|
|
Spindrift
Spindrift
Started Topics :
33
Posts :
1560
Posted : Mar 30, 2009 15:40
|
Quote:
|
On 2009-03-30 14:11, x-rayz wrote:
last night I spoked with some guy who recently started producing in FL.. I told him about FLAC format and how he can convert it from wav. But he didnt knew to export from FL into wav, only to mp3.. So I told him to export to wav and guess what was his reaction.. Wow, I can hear the difference!
I think that concludes this series of tests about the quality..
|
|
That's a very conclusive test 
Would maybe have some relevance if it is was conducted blind and you included details of bitrate and encoding algorithm used...no-one said a 24kbps is transparent.
(``·.¸(``·.¸(``·.¸¸.·`´)¸.·`´)¸.·`´)
« .....www.ResonantEarth.com..... »
(¸.·`´(¸.·`´(¸.·`´``·.¸)``·.¸)``·.¸)
http://www.myspace.com/spindriftsounds
http://www.myspace.com/resonantearth |
|
|
|
x-rayz
IsraTrance Junior Member
Started Topics :
11
Posts :
576
Posted : Mar 30, 2009 16:59
|
|
RK9
IsraTrance Junior Member
Started Topics :
21
Posts :
210
Posted : Mar 30, 2009 17:44
|
Quote:
|
On 2009-03-30 12:28, Kitnam wrote:
when the point is "not to produce an identical waveform, but one that sounds identical to the original", then this means that this can only be done with an identical waveform, in case we speak in a range of 20 to 20000 hz.
2 different waveforms within this range will sound different, this is a not negable mathematical fact.
|
|
I am a UC Berkeley signal processing student and you are wrong. There are plenty of sets of two waveforms that sound the same.
For example, take a signal that's a 330 Hz sine wave mixed with a 440 Hz sine wav. You can change the phase of one of the two sines by any amount in any direction and it will sound exactly the same. You will end up with a different waveform for sure but the human ear cannot tell the difference. |
|
|
|
Spindrift
Spindrift
Started Topics :
33
Posts :
1560
Posted : Mar 30, 2009 18:38
|
Quote:
|
On 2009-03-30 16:59, x-rayz wrote:
hehe it was a blind test alright, I havent seen nor heard what the guy was doing, I heard just awesome! after it was done
|
|
Exactly how did you perform the comparison?
Did you use ABX software, and in that case you have logs?
Do you know what encoder FL was using and with what settings?
Did you only test once, or did you perform several comparisons to discount chance?
It really doesn't take many minutes to setup and perform a proper ABX test with WinABX, but through the countless discussions on the subject, the only ABX logs and documented tests I have come across indicates that high bitrate mp3's are transparent.
The only tests challenging that conclusion are of the variety "I played an mp3 and a wav for a friend once and he heard the difference", i.e. neither the method used nor the variables involved documented at all.
(``·.¸(``·.¸(``·.¸¸.·`´)¸.·`´)¸.·`´)
« .....www.ResonantEarth.com..... »
(¸.·`´(¸.·`´(¸.·`´``·.¸)``·.¸)``·.¸)
http://www.myspace.com/spindriftsounds
http://www.myspace.com/resonantearth |
|
|
|
Kitnam
Mantik
Started Topics :
110
Posts :
1151
Posted : Mar 30, 2009 23:36
|
@spindrift & rk9
thx for the knowlegde sharing guys.
so far it seems like the human has some problems identifying all signals in the spectrum depending on other signals within the spectrum which has a kind of psychoaccustic reason. did i got that correctly.
this make the nearly inaudible differences from my "technical" phasing-tests even more uneffective for the recieved audio as they allready has been from my point of view. so a 320 mp3 will everybode make happy?
and i realy dont trust A/B testing, the differences people say they can listen seem sometimes esoteric to me.
|
|
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Mar 31, 2009 01:41
|
|
~d2~
Inactive User
Started Topics :
7
Posts :
751
Posted : Mar 31, 2009 01:56
|
The null test or phase reversal test is to check to see if 2 files are identical when people claim they are not.
One flaw with mp3 encoding is that it is designed on a model of the human hearing system which is based on averages. Therefore it is possible for some to hear differences when others can not. This I have found is often overlooked. |
|
|
|
Upavas
Upavas
Started Topics :
150
Posts :
3315
Posted : Mar 31, 2009 07:59
|
Quote:
|
The only tests challenging that conclusion are of the variety "I played an mp3 and a wav for a friend once and he heard the difference", i.e. neither the method used nor the variables involved documented at all.
|
|
That is unfair Spindrift! I had a room full of audio teachers (mixlab) who could all hear the difference, as could I. And I did tell you about it...
Upavas - Here And Now (Sangoma Rec.) new EP out Oct.29th, get it here:
http://timecode.bandcamp.com
http://upavas.com
http://soundcloud.com/upavas-1/ |
|
|
|
