Conclusive "Proof" that higher resolution audio sounds different

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Rob, I had a look at the spectrum plots in Audacity & don't see any problem - the spectrum plots overlay pretty exactly (by eye) between the different sample rate files (obviously the higher sample rate files plot to higher frequencies). The stopband frequencies seem correct for each sample file. Am I missing something?
 

Robh3606

Well-Known Member
Aug 24, 2010
1,471
461
1,155
Destiny
Am I missing something?

Probably not could certainly be me. I thought that the lowest res file was limited to 16K not 8K. Maybe I misundserstood.

Rob:)
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Probably not could certainly be me. I thought that the lowest res file was limited to 16K not 8K. Maybe I misundserstood.

Rob:)
The numbers in file names are sampling rate, not max frequency. Divide that by two and you get the max frequency.

BTW, when I downloaded them the lowest was 32 Khz sampling or 16 Khz max response. He added the file you tested later.
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Amir,
I'm following the conversation on AVS & see that you seem to be saying that ArnyK messed up the downsampling to 16/44? Is this what you maintain? Have you done this as Tim on AVS has recently reported
"I got an odd result. I made my own downsample to 16/44 with all the options disabled, and then upsampled back to 24/96. I then inverted my downsample and mixed it with the original full band version. I get a perfect null for frequencies below c 20K.

However if I do the same with the supplied 16/44 sample (which I understand was used for the ABX) it doesn't null well at all. Maybe the two files are not perfectly synched?

Apart from the possibility that the two files have different offsets (which would cause a nulling issue) have you reason to believe that the 16/44 is a messed up downsample of 24/96 file?
 

Orb

New Member
Sep 8, 2010
3,010
2
0
Amir,
I think koturban is going on about the following IM test (3rd of the way down) : http://xiph.org/~xiphmont/demo/neil-young.html
TBH IM can be measured with standard-ish/accepted tests so not sure the point of this and surprised it is this most of them are getting hung up on (especially when real world means diverse systems-recordings-environment and purpose is hearing differences that do exist) *shrug*.
I would be more leery about the true native hirez file (higher bit depth and sampling rate) vs how the downsampled file was created myself :)

Cheers
Orb
 

Orb

New Member
Sep 8, 2010
3,010
2
0
Just to add,
I feel if they really are concerned with IM, then they really should also be concerned about the strength/weakness of stop-band rejection of the filter/system as well (in fact real world systems this would be more of a concern IMO in the context of influencing audibibilty) - sticking with headphones for now otherwise this potentially brings in the Gedlee papers relating to speakers.
After all the context is about digital, but this is just my own take on the subject.

Cheers
Orb
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Amir,
I'm following the conversation on AVS & see that you seem to be saying that ArnyK messed up the downsampling to 16/44? Is this what you maintain? Have you done this as Tim on AVS has recently reported

Apart from the possibility that the two files have different offsets (which would cause a nulling issue) have you reason to believe that the 16/44 is a messed up downsample of 24/96 file?

Amir, serious question - any evidence ArnyK's downsampling is messed up & this is why you are able to distinguish his 16/44 from 24/96 files? Have you tired downsampling the original 24/96 file to 16/44 using your favourite SRC & ABXed the resultant files?
 
Last edited:

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Amir, serious question - any evidence ArnyK's downsampling is messed up & this is why you are able to distinguish his 16/44 from 24/96 files? Have you tired downsampling the original 24/96 file to 16/44 using your favourite SRC & ABXed the resultant files?
I have not had any time to do further investigation.

That is almost beside the point though. The first and foremost point I want to make is that there are differences between listeners. Files have been put forward that the experimenter and many others say they can't hear the difference yet I could. And so did other people. Including some non-believers. So we have invalidated this strong claim that "there are no golden ears."

Also note that I have also passed Scott's test which was meticulously level matched and produced.
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Here are more results posted this morning on AVS:

OK, for this test I used my DIY PC. Audio is coming from of the video card's HDMI output (ATI Radeon R9 200 series). Processing is via Pioneer Elite SC-55 in Pure/Direct mode. I listened through a pair of Behringer B215XL speakers. I made one goof, but ultimately it was just as easy to tell 16/44.1 from 24/96 as it was with the laptop/headphones combo.

foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/21 09:48:16

File A: E:\AVS\Foobar ABX\Jangling Keys\keys jangling band resolution limited 4416 2496.wav
File B: E:\AVS\Foobar ABX\Jangling Keys\keys jangling full band 2496.wav

09:48:16 : Test started.
09:50:38 : 01/01 50.0%
09:50:45 : 02/02 25.0%
09:50:54 : 03/03 12.5%
09:51:02 : 04/04 6.3%
09:51:09 : 05/05 3.1%
09:51:14 : 06/06 1.6%
09:51:22 : 07/07 0.8%
09:51:29 : 07/08 3.5%
09:51:37 : 08/09 2.0%
09:51:43 : 09/10 1.1%
09:51:47 : 10/11 0.6%
09:51:53 : 11/12 0.3%
09:51:56 : Test finished.

----------
Total: 11/12 (0.3%)
 

Orb

New Member
Sep 8, 2010
3,010
2
0
Ok this is bad form of me to say this but..
I really think I need to stop reading the discussion going on regarding this topic on AVSF, serious face and palm moments with much sighing.
I doubt anyone there will accept that you are passing them using valid equipment combined with valid methodology that IMO is the only way to successfully pass subtle ABX with music or complex sounds.
Mind boggling considering how much equipment is tested and measured for IM sigh, and yet do not consider potential stop-band rejection/attenuation implications in general - not to do with you passing the test but of interest in general IMO when it comes to digital systems and the diverse filters-DACs and sampling rates, also includes impulse response ripple/ringing potentials (FR and time domain behaviour).

Keep up the good work.
Cheers
Orb
 
Last edited:

mgagneur

New Member
Jul 23, 2014
1
0
0
Hello Amir,

I have been following the threads here and over at AVS with much interest. At the risk of being attacked I applaud your stance. However you achieved it, you (and others) have taken the files and consistently been able to detect a difference. The differences may be subtle but that was not the point of the exercise. It is most amusing to see the AVS crowd talk themselves into nonsense circles attacking everyone who questions their religion. :)

Well done,

Marty
 

Phelonious Ponk

New Member
Jun 30, 2010
8,677
23
0
Hello Amir,

I have been following the threads here and over at AVS with much interest. At the risk of being attacked I applaud your stance. However you achieved it, you (and others) have taken the files and consistently been able to detect a difference. The differences may be subtle but that was not the point of the exercise. It is most amusing to see the AVS crowd talk themselves into nonsense circles attacking everyone who questions their religion. :)

Well done,

Marty

Yeah. Thank God that never happens here. :)

Tim
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Hello Amir,

I have been following the threads here and over at AVS with much interest. At the risk of being attacked I applaud your stance. However you achieved it, you (and others) have taken the files and consistently been able to detect a difference. The differences may be subtle but that was not the point of the exercise. It is most amusing to see the AVS crowd talk themselves into nonsense circles attacking everyone who questions their religion. :)

Well done,

Marty
Thanks Marty. I must say, in no way did I imagine finding myself in this situation. It has been years since my listening abilities were put to test critically and I just automatically assumed I had lost them. It was just an amazing turn of events to have multiple challenges put forward and managing to meet them. And a couple of others following me just the same, at least partially.

And yes, the tone of interactions on AVS are just out of this world. People keep acting as if nothing has happened. That if they try hard enough, they can just dismiss the results. I never know what obstacle they are going to throw at me next but so far, so good :).
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Yeah. Thank God that never happens here. :)

Tim
It has Tim but that accusation is a one-way one. That is, the subjectivists are accused of never accepting new data and changing their views. And that they are the "anti-science" crew and not them.

My answer to this is that subjectivists don't make claims, not always anyway :), that their belief in audio is based on DBTs, engineering, etc. The extreme objectivists do. So they need to set the example of learning from new data. Your reaction Tim on this forum is what I expected them to have. But that is not remotely the case there. Just yesterday Arny announced that my "homework is not finished" and if I don't run his latest "IM Distortion Test" that my results are "bogus."
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
Ok this is bad form of me to say this but..
I really think I need to stop reading the discussion going on regarding this topic on AVSF, serious face and palm moments with much sighing.
I doubt anyone there will accept that you are passing them using valid equipment combined with valid methodology that IMO is the only way to successfully pass subtle ABX with music or complex sounds.
Mind boggling considering how much equipment is tested and measured for IM sigh, and yet do not consider potential stop-band rejection/attenuation implications in general - not to do with you passing the test but of interest in general IMO when it comes to digital systems and the diverse filters-DACs and sampling rates, also includes impulse response ripple/ringing potentials (FR and time domain behaviour).

Keep up the good work.
Cheers
Orb

Thanks Orb. Like you, every time I go there and read a post, my jaw drops. There is just no common sense in those discussions.
 

esldude

New Member
I notice in the files for Arny's keys, there is a slight shift in time between the tracks downsampled vs the full hirez version. Do you think that with you listening for very short periods of time this is perhaps what is letting you hear a difference? The 44.1 downsample has less lag than those of lower sample rate filtering.

The downsampled files at AVS also have a time lag for downsampled files. The AVS files also don't null out spectacularly well even when lined up. Arny's files null out better though not perfectly. Yes, the lags are small fractions of a second. Not saying you aren't hearing a difference just pondering some other reasons why it might be you are getting good blind results.
 

Phelonious Ponk

New Member
Jun 30, 2010
8,677
23
0
It has Tim but that accusation is a one-way one. That is, the subjectivists are accused of never accepting new data and changing their views. And that they are the "anti-science" crew and not them.

My answer to this is that subjectivists don't make claims, not always anyway :), that their belief in audio is based on DBTs, engineering, etc. The extreme objectivists do. So they need to set the example of learning from new data. Your reaction Tim on this forum is what I expected them to have. But that is not remotely the case there. Just yesterday Arny announced that my "homework is not finished" and if I don't run his latest "IM Distortion Test" that my results are "bogus."

My reaction is nothing new. I've believed you were hearing things, real things, that I don't hear for a long time, Amir. But I also believe that your training and listening abilities are very rare, and that it's not surprising that most of the folks over on AVS don't hear a difference. And are these small difference you've had to train yourself to listen for in isolated, ideal passages somehow coming through in the casual listening of the hundreds of audiophiles who have long insisted on the clear, analog-like superiority of hi res over Redbook digititis? Of that, I remain very skeptical.

Army's IMD test? Is he trying to show that the difference you hear is IMD in the audible range created by your system's reproduction of supersonic content? Pretty curious about that one myself, but it would do nothing to render your results "bogus." The fact that he swore there was no audible difference, and you demonstrated that you could differentiate them audibly, would remain.

Arny is wrong.

Have any of our most vocal subjectivists taken the test yet?

Tim
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Thanks Marty. I must say, in no way did I imagine finding myself in this situation. It has been years since my listening abilities were put to test critically and I just automatically assumed I had lost them. It was just an amazing turn of events to have multiple challenges put forward and managing to meet them. And a couple of others following me just the same, at least partially.

And yes, the tone of interactions on AVS are just out of this world. People keep acting as if nothing has happened. That if they try hard enough, they can just dismiss the results. I never know what obstacle they are going to throw at me next but so far, so good :).

Amir, I started a thread on another forum (quiet soon after this one started) where the Obj Vs Subj row has always raged which recently had discussed using Evidence-based Vs Faith-based to define the two camps - I called my thread "Sorting out evidence-based from faith-based" & posted your results. The furore was very like the AVS crowd - denial, poo-pooing the results, questioning the integrity of the solo testers & here's the real clincher, denying that one test has any significance. For people that had demanded blind test proof of anybody that reported a subjective impression, when faced with the truth, they just couldn't handle it (as Jack Nicholson once said). I actually don't believe any of them had encountered Foobar ABX before or knew of it's existence or the notion of statistical significance :) I had always called their "blind tests" half-arsed, pseudo-science & this proved it - no consideration for control of any biases other than the sighted bias.
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing