[mikeaj]

Quote:

Of course you can't tell which is the original if you've never heard the original! Same as if I showed you three versions of a photo, 2 of them post-processed then asked you which was the original. You'd have no idea!

 
I agree with this idea in general, but we're looking at something in specific.  There are a lot of claims about inherent superiority of high-res, audible and irrevocable harm if downsampling and reducing bit depth.  The goal was to determine which is best, which "should" have been apparent, in the way that changing most images to 8-bit color actually should be apparent (and people would recognize the original there).  The reality was that the lowest-fidelity reproduction sounded fine:  the same as the others according to some, not the same according to others—the ones not listening blinded.

 

[autumnholy]

I was rather skeptical on the differences. But then, I tried to ABX between all of them. A&B I couldn't hear any difference. But then, I tried B&C. The first 5 attempts were rather confusing, as I didn't know where to find the difference. Until I tried a smaller sample, around the 4-6sec period, then I scored a straight 8 corrects on ABX between B&C, until i started to get anxious and failed the last 2 attempts. So, my score was 11/15 on B&C. Still didn't wanna give up because I screwed up on the last 2 attempts, I tried a fresh ABX again, and this time, I failed miserably. Don't know if it's fatigue or I lost concentration or I was just imagining all the differences up, I can't consistently ABX both B&C.

 

[analogsurviver]

Quote:

And always changing the subject to avoid facing up to the truth. Not much more to say to him if he's failed the test and he still keeps claiming super human abilities.
I guess you can just add his name as a notch in your belt and wait for the next one to come along, xnor. Good job.

Never did claim superhuman abilities. My ability to hear pure sine wave tones is about what should be for my age. I did correctly identifiy the most mangled sample, the b with original recording replaced with white noise above 21 kHz as the most unnatural and placed it at the third place as the worst. It sounds to me the same/similar as MP3 vs DSD - state again, never heard such a manipulation before. It took about 2-5 seconds to hear there is something wrong with it. Both others are far better, as my original post in this thread clearly stated. 

 

[stv014]

Quote:

I did correctly identifiy the most mangled sample, the b with original recording replaced with white noise above 21 kHz

 
It is not the most mangled sample, A is, which you chose as the "best". A was created from B, and carries all of its degradation, with the additional mangling of quantization to 16 bits.

 

[nick_charles]

My comments in bold
 
Quote:

The quality goes as follows, from best to worst:
 
1. a
2. c
3. b
 
a has the most natural sound, especially decay of percussion at the beggining, with a quite nice sense of depth. 
 
b is shrunk in all dimensions and is unpleasently grainy in the treble, it sounds flat and lifeless after hearing a
 
c is a welcome relief after b, but still not up to a - but is much closer to a than to b.------ No - see XNOR segment below
 
My guesstimate is that a is originally 24 bit at at least 88,2 kHz sampling rate, b some MP3 and c 44,1-48 kHz / 16-24 bit.

 
Quote:

...I did correctly identifiy the most mangled sample, the b with original recording replaced with white noise above 21 kHz as the most unnatural and placed it at the third place (but A is the technically worst sample not B, A adds another level of degradation) as the worst. It sounds to me the same/similar as MP3 vs DSD - state again, never heard such a manipulation before. It took about 2-5 seconds to hear there is something wrong with it. Both others are far better (but A is worse i.e not far better than B as it has the limitations of B and also lowers bit-depth from 24 to 16 bits with crude dither - you placed A in 1st position) , as my original post in this thread clearly stated. 

 
And the facts....
 
 
 

C is the original, unaltered sample.

B is what I call fake 96 kHz. Let me explain.
The file was filtered exactly like resampling to 44.1 kHz would, removing everything above 22.05 kHz.
To make the file indistinguishable from C, I had to add some high-frequency content. I chose completely random (white) noise, shaped to match the waveform.

A is CD quality, the "worst" in the set, because additionally to the processing done to B the bit depth was changed to 16 bits. 1 bit TPDF dither was used, no noise shaping (so no one can complain about moving the noise beyond 22.05 kHz).

--> C > B > A
 
 

 
Your ratings best to worst A-----C------B
Actual quality best to worst C------B------A
 
C is closer to B than A
 
Why do you keep trying to convince yourself that you did not get it utterly wrong that you most likely merely imagined that you could hear differences. The ability to self-deceive is a strong one in humans (I have two degrees in Psychology).
 
But to close this for yourself why not just take the samples and do a set of DBTs in FooBar, I suggest a minimum of 20 trials per set as we have seen people guess very well and the more trials you do the less likely you are to just guess correctly. Then you can see for yourself. As you thought you heard a massive difference between B and A you should at least be able to replicate your findings with that pair trivially.

 

[bigshot]

The fact that everyone picked a different ranking for the samples shows that they are audibly identical. If you really think you were able to detect something, take the next step and fly blind. But it seems that you really don't want the real answer.

 

[analogsurviver]

Quote:

The fact that everyone picked a different ranking for the samples shows that they are audibly identical. If you really think you were able to detect something, take the next step and fly blind. But it seems that you really don't want the real answer.

Sorry - I DO want the right answer. Xnor provided it and although I do not agree with such manipulations that can never occur in real life or any known unmanipulated recording process, I aceept it as possible means of proving that response above 20 kHz does not matter. I just do not find it necesarry to n-times listen to b, which did not sound OK to me on first and any subsequent listens - and always will, as long as you use equipment with enough high frequency extension.
 
A friend who sells audio also listened to these samples and also chose the same ranking of a>c>b - with the same reservations regarding c as myself.
Most demoing of equipment is done today with CD, less with SACD and for obvious reasons even less with analog - so he has that "CD in ears" for sure as well. The fact that I object to b so much is simple - it just does not sound right. As it is a fake 96/24 with real recording above 21 kHz removed and replaced with white noise and A was apearently made from B by simply making 44,1/16 CD out of it, makes A more or less the same as if it was made from real 96/24 or C, as for CD anything above 21 k or so gets filtered, no matter if original recording or artificially added white noise. A did sound  
correct and normal to me, B - NEVER. The fact that I object b so much must be somehow connected with "hearing/sensing/whatever" that artificially added noise above 21 k or so - and I repeat, I can not possibly hear pure 21 kHz sine tone.
 
All three samples sounded decidedly PCM. Linn Records had for this Xmas for each day one free download - ranging in resolution from 44,1kHz/24 bit all the way up to 192/24 for some more serious stuff. It is only 192/24 that does approach real thing - for the "lighter"music genres, only one jazz piece received 192/24, most pop/rock/indie stuff not exceeding 48 k/24 - Linn did not offer higher resolution for these genres for a reason. All recordings, no matter the resolution, were of at least demo quality.  And I will purchase the whole albums of those recordings that are available as SACDs, that is in other words DSD at 2.8 MHz - which was most probably original master format, although I suspect that real master was even better DSD at 5.6 MHz due to the awesome sound quality. Wish Linn offered DSD downloads besides present PCM downloads of various resolution/price as well.
 
I did listen to some more "CD sound" on vinyl today : http://www.discogs.com/Joe-Williams-Every-Night-Recorded-Live-On-Vine-St/release/3561459
because I do prefer good music performed really well as in this live ( regrettably digital ) recording over some less inspired performance(s) on direct to disc pure analog recordings - adding further to "CD in my ear". My only regret is that very seldom do superb musicianship and recording coincide on the same "whatever" you can buy and listen to at home. Perhaps the most succesful to date is http://www.discogs.com/Harry-Belafonte-Belafonte-At-Carnegie-Hall-The-Complete-Concert/release/1177631 - if you can not hear clear difference between the original LP issue (or many re-releases ) and CD, be it regular or ((gold) limited ) edition(s), then you are in the wrong hobby or profession. Here the first hit I got for "Harry Belafonte Live at Carnegie Hall SACD" search: http://www.thefreelibrary.com/The+Living+Stereo+SACDs+Revisited+and+the+Belafonte-Carnegie+Shootout.-a0131903296 Please note that the SACD mentioned was mastered from CD version/master - NOT directly from analog, which should have resulted in much better sonics. No idea if SACD mastered directly from analog master is available by now - will stick to the original analog LP anyway.
 
But, be it as it may with SQ - enjoy the recording for the music, as it probably still is the best captured live concert in history. I hope you will not disagree with this last statement too.
 
This is the last post of mine regarding abc test for which I thank  the author again.

 

[bigshot]

Before you took this test, you swore up and down that you could easily tell redbook from high bitrate. You said that redbook was decidedly inferior. Clearly, that isn't the case.

The fact that you expressed preferences doesn't prove that you actually were able to hear differences. The only way to do that is to correctly identify which is which in a totally blind comparison.

The problem is, the results that really count here are perfectly clear. You can't tell the difference between redbook and high bitrate. The ringer tossed in to prevent folks from file peeping doesn't matter. It's just a red herring. The point is, you were unable to spot the unaltered high bitrate file in the bunch. In fact, you chose the one with *none* of the attributes of high bitrate sound as the best. That proves two things... Extended frequency response from high sampling rates were inaudible to you, and the lower noise floor of high bitrate sound was inaudible too.

If you were mistaken about something *that* fundimental, perhaps the folks in Sound Science who you were so sure were dogmatic about charts and diagrams might know something about listening that you don't know too.

 

[nick_charles]

Quote:

Before you took this test, you swore up and down that you could easily tell redbook from high bitrate. You said that redbook was decidedly inferior. Clearly, that isn't the case.
The fact that you expressed preferences doesn't prove that you actually were able to hear differences. The only way to do that is to correctly identify which is which in a totally blind comparison.
The problem is, the results that really count here are perfectly clear. You can't tell the difference between redbook and high bitrate. The ringer tossed in to prevent folks from file peeping doesn't matter. It's just a red herring. The point is, you were unable to spot the unaltered high bitrate file in the bunch. In fact, you chose the one with *none* of the attributes of high bitrate sound as the best. That proves two things... Extended frequency response from high sampling rates were inaudible to you, and the lower noise floor of high bitrate sound was inaudible too.
If you were mistaken about something *that* fundimental, perhaps the folks in Sound Science who you were so sure were dogmatic about charts and diagrams might know something about listening that you don't know too.

 
 
It is no longer a matter of audio here but solely one of Psychology which is interesting in and of itself but getting off topic rather, I think this thread has run its course - many thanks to XNOR !

 

[Lenni]

Quote:

 
 
It is no longer a matter of audio here but solely one of Psychology which is interesting in and of itself but getting off topic rather, I think this thread has run its course - many thanks to XNOR !

 
before you start with the psychology theories, if I'm not mistaken the majority of users who participated in this test couldn't differentiate any file (including myself). doesn't that also mean something? it seems you rather (forcefully) focus on the 1% who heard a difference and ignore the rest of the result. just making an observation.

 

[nick_charles]

Quote:

 
before you start with the psychology theories, if I'm not mistaken the majority of users who participated in this test couldn't differentiate any file (including myself). doesn't that also mean something? it seems you rather (forcefully) focus on the 1% who heard a difference and ignore the rest of the result. just making an observation.

I think there is a problem of communication here. As you say the vast majority admitted an inability to differentiate between the files. For these no further explanation is necessary beyond that the differences were inaudible to them. My last remark was solely directed at analogsurvivor who also seems incapable of detecting a difference but also incapable of admitting the fact, that is where the reference to psychology comes in...

 

[Currawong]

Quote:

The fact that everyone picked a different ranking for the samples shows that they are audibly identical. If you really think you were able to detect something, take the next step and fly blind. But it seems that you really don't want the real answer.

 
No it doesn't and, regardless, not everyone "picked a different ranking".
 
Quote:

Before you took this test, you swore up and down that you could easily tell redbook from high bitrate. You said that redbook was decidedly inferior. Clearly, that isn't the case.

The fact that you expressed preferences doesn't prove that you actually were able to hear differences. The only way to do that is to correctly identify which is which in a totally blind comparison.

The problem is, the results that really count here are perfectly clear. You can't tell the difference between redbook and high bitrate. The ringer tossed in to prevent folks from file peeping doesn't matter. It's just a red herring. The point is, you were unable to spot the unaltered high bitrate file in the bunch. In fact, you chose the one with *none* of the attributes of high bitrate sound as the best. That proves two things... Extended frequency response from high sampling rates were inaudible to you, and the lower noise floor of high bitrate sound was inaudible too.

If you were mistaken about something *that* fundimental, perhaps the folks in Sound Science who you were so sure were dogmatic about charts and diagrams might know something about listening that you don't know too.

 
 12 "you"'s. Please don't ruin an interesting experiment by attacking people.

 

[bigshot]

No attack intended. I'm just commenting on someone's analysis of a listening test.

 

[wtaylorbasil]

I have come late into this discussion. Are the three audio files available for me to downlosd?
I would be most pleased if someone can make it available to me.
 
regards
William

 

[xnor]

I will send them to you.