Blue or red pill?

[esldude]

22 hours ago, Audiophile Neuroscience said:

 

Dennis, I don't think it presents a problem.The switching tells (or lack of) and  hearing sound differences can be viewed separately. One is a potential confounder to the outcome and the other is the outcome. There is a logical explanation why the sound differences might be more apparent with one switching *methodology* but the actual switching mechanism itself remained constant and controlled.Sure the sequence of switching changed but a "click" for A and a "clack" for B (or whatever the tell might be) was consistent throughout.

I think that is no more solid than unfounded conjecture with the info available.

[Audiophile Neuroscience]

On 4/27/2018 at 8:59 PM, manisandher said:

 

Back to clutching at straws.

 

The two test protocols were different wrt identifying X correctly:

 

 

But the protocols were identical wrt the influence of any potential 'tells'.

 

There were clearly no tells present during the first protocol... and they didn't magically appear before the second.

 

Mani.

 

On 4/27/2018 at 8:46 PM, Audiophile Neuroscience said:

 

 If there were "tells" Mani would have passed the ABXXXXXXX trials. One thing that links the two test methods as being alike is the potential for click based "tells" ie being the same in both instances. Its like saying maybe room temperature or lighting cued Mani to know the results. However these variables, like the clicks, were held constant between the two tests so in that way the tests were not dissimilar but alike. Something else other than room temperature or lighting or clicks differentiated between the tests.

 

On 4/28/2018 at 5:32 PM, Audiophile Neuroscience said:

 

There is a logical explanation why the sound differences might be more apparent with one switching *methodology* but the actual switching mechanism itself remained constant and controlled.Sure the sequence of switching changed but a "click" for A and a "clack" for B (or whatever the tell might be) was consistent throughout.

 

2 hours ago, esldude said:

I think that is no more solid than unfounded conjecture with the info available.

 

Controlling for variables is standard scientific test procedure. I agree with Mani that clicking or other tells were held constant between the two test methodologies ie consistent=unchanged=a *controlled* variable.

If the same variable was held constant then logically any difference in the results is attributed to something else other than the *unchanged* variable.You and some others seem to imply the tells just magically appeared (as Mani puts it). In my book *that* is conjecture.

 

 

 

[adamdea]

You’re basically going round in a circle of false premises and irrelevant deductions from the same. 

 

The presence of an audible difference caused by the software setting change was also a hypothesised (by you and mani) factor present in ABxxxx and ABx therefore it too “logically”  can’t explain the difference between the ABxxx and the ABx!

 

Except no one is explaining that difference with that (non-)variable. “Logic” doesn’t help with confused premises. 

 

The only thing being explained is the 9/10 in the ABxxx. And there are competing theories as to what might have been detected. No Occam’s razor. 2 different hypotheses. Ontological parsimony not really helpful

[Audiophile Neuroscience]

1 hour ago, adamdea said:

You’re basically going round in a circle of false premises and irrelevant deductions from the same. 

 

The presence of an audible difference caused by the software setting change was also a hypothesised (by you and mani) factor present in ABxxxx and ABx therefore it too “logically”  can’t explain the difference between the ABxxx and the ABx!

 

Except no one is explaining that difference with that (non-)variable. “Logic” doesn’t help with confused premises. 

 

The only thing being explained is the 9/10 in the ABxxx. And there are competing theories as to what might have been detected. No Occam’s razor. 2 different hypotheses. Ontological parsimony not really helpful

 

Adam, I have no idea what you are talking about but I suspect it makes no sense.

Edit: maybe Latin is more your thing?

[manisandher]

1 hour ago, adamdea said:

You’re basically going round in a circle...

 

You can't bring yourself to accept that I heard a difference in the sound of the music between A and B, despite the 9/10 ABX score, and zero evidence to the contrary.

 

Fine. Move on. Nothing for you here.

 

Mani.

[manisandher]

On 4/26/2018 at 10:02 AM, PeterSt said:

But only if you have some time for it.

 

Here are some captures hot off the press.

 

- digital and analogue captures taken simultaneously, both at 24/176.4

- Tascam clock (digital captures) slaved to MOTU clock (analogue captures), which I'm hoping will aid a little in alignment between digital and analogue captures, should that be something that you'd like to explore

- ID-tags edited to keep track of setups

 

https://drive.google.com/open?id=1oIon6NSPETqtmbXG23Z9hIMzeuKlAIGk

 

Mani.

[PeterSt]

3 hours ago, manisandher said:

- Tascam clock (digital captures) slaved to MOTU clock (analogue captures),

 

Hey Mani, that is nice ! I don't think that will make it worse.

Meanwhile at this moment I am not up to it. Tax form day. 

[manisandher]

3 hours ago, PeterSt said:

Tax form day. 

 

Ooh, lucky you.

 

Mani.

[manisandher]

6 hours ago, manisandher said:

Here are some captures hot off the press.

 

I took a listen to the analogue captures I linked to earlier today and they sounded worse than the previous ones from the MOTU ADC. They also seemed to have a slight channel imbalance. I then realised that the MOTU's input was set to 'variable' instead of 'fixed', meaning that the signal from the DAC was passing through additional pots and circuitry in the MOTU.

 

I've re-done the captures, just as earlier, but with the ADC set to its 'fixed' input. I think these captures should be used for any listening/analysis rather than the earlier ones:

 

https://drive.google.com/open?id=1lu6CdvCAlDqGcxV6sQwcaTEFfbhOlKQz

 

Mani.

[PeterSt]

4 minutes ago, manisandher said:

Ooh, lucky you.

 

If course I avoid tax paying.  But it requires forms to fill out. hehe

Regards, Peter

 

[PeterSt]

1 minute ago, manisandher said:

I've re-done the captures, just as earlier, but with the ADC set to its 'fixed' input.

 

Clocks still connected ?

I ask because only theoretically this is a good idea. I mean, via WordClock connections ... I know, a different subject, but please take into account that such a connection can be a culprit for best sound (high jitter).

[manisandher]

10 minutes ago, PeterSt said:

Clocks still connected ?

 

Yes, exactly as before. (But to be clear, the DAC clock isn't involved - there's no wordclock I/O on the DAC unfortunately.)

 

I have no idea what happens when a clock drifts during a digital capture, i.e. no conversion taking place. I suspect nothing much, provided the drift is within a specified range. But I felt that slaving the digital capturing device's clock to that of the analogue capturing device couldn't do any harm, and might perhaps prove useful for alignment.

 

Mani.

[fas42]

Sorted ... should have read the next posts, first, .

[fas42]

4 hours ago, manisandher said:

 

I took a listen to the analogue captures I linked to earlier today and they sounded worse than the previous ones from the MOTU ADC. They also seemed to have a slight channel imbalance. I then realised that the MOTU's input was set to 'variable' instead of 'fixed', meaning that the signal from the DAC was passing through additional pots and circuitry in the MOTU.

 

Which is how it works. Every unnecessary part in the circuit path, unless it is 100% pristine, is liable to degrade the sound - this relates to every area of a playback rig; optimising is merely locating the worst offenders, and removing them, replacing them with greatly improved quality units, or redesigning how it works.

[fas42]

Well, the drift has been solved, but the two versions aren't aligned to the beat of the clock. Just looking at one spot, comparing 23 and 24, the samples are out by about 2u secs - resampling will be necessary, to get decent nulling.

[pkane2001]

23 hours ago, fas42 said:

I'll leave it to Paul to get a "proper" mechanism to fire up,

 

Still working on it. Seems the FFT library I picked isn’t the fastest, but it does the job. May need to switch to something faster before too long (used FFTW in the past, and that’s blazingly fast).  

 

Meanwhile, I have drift correction mostly functional as of last night. It will need a lot more testing and tuning, of course, as well as careful measurements to ensure I’m not distorting anything while resampling.

[STC]

6 hours ago, manisandher said:

 

I took a listen to the analogue captures I linked to earlier today and they sounded worse than the previous ones from the MOTU ADC. They also seemed to have a slight channel imbalance. I then realised that the MOTU's input was set to 'variable' instead of 'fixed', meaning that the signal from the DAC was passing through additional pots and circuitry in the MOTU.

 

I've re-done the captures, just as earlier, but with the ADC set to its 'fixed' input. I think these captures should be used for any listening/analysis rather than the earlier ones:

 

https://drive.google.com/open?id=1lu6CdvCAlDqGcxV6sQwcaTEFfbhOlKQz

 

Mani.

 

 

Tried this and this is so much easier. No fancy system. Just a $20 Sennheiser earphones and PC. But you can only get this right after being familiar to the particular test. Throw me a new track and I am bound to fail. 

 

foo_abx 2.0.4 report
foobar2000 v1.3.17
2018-04-30 09:11:57

File A: 23. analogue capture _ A.wav
SHA1: a48b6be748d9af8e431795a3d6792754ebbe2ae4
File B: 24. analogue capture _ B.wav
SHA1: 99fca571ad6742e44376e4d8f6f38c2164595624

Output:
DS : Primary Sound Driver
Crossfading: NO

09:11:57 : Test started.
09:12:10 : 01/01
09:12:16 : 02/02
09:12:21 : 03/03
09:12:26 : 04/04
09:12:32 : 05/05
09:12:38 : 06/06
09:12:47 : 07/07
09:12:55 : 08/08
09:13:00 : 09/09
09:13:09 : 10/10
09:13:15 : 11/11
09:13:20 : 12/12
09:13:25 : 13/13
09:14:15 : 14/14
09:14:23 : 15/15
09:14:28 : 16/16
09:14:50 : 17/17
09:14:57 : 18/18
09:15:02 : 19/19
09:15:07 : 20/20
09:15:07 : Test finished.

 ---------- 
Total: 20/20
Probability that you were guessing: 0.0%

 -- signature -- 
d53f34ba1b65e93c2976eed569f43e7bac52e42e

[fas42]

32 minutes ago, STC said:

 

 

Tried this and this is so much easier. No fancy system. Just a $20 Sennheiser earphones and PC. But you can only get this right after being familiar to the particular test. Throw me a new track and I am bound to fail. 

 

 

Interesting. For whatever reason, on a cursory listen I found these latest two analogue captures harder to distinguish from each other, compared to the previous lot - I'm pleased that someone can pick up a clearcut variation, makes investigating them more worthwhile.

[STC]

17 minutes ago, fas42 said:

 

Interesting. For whatever reason, on a cursory listen I found these latest two analogue captures harder to distinguish from each other, compared to the previous lot - I'm pleased that someone can pick up a clearcut variation, makes investigating them more worthwhile.

 

I am so confident that I can do it again so that there will be no whatsoever dispute. However, this is not a real blind test since I am operating the ABX of Foobar. Will give another try with 30 trials. 

[STC]

1 hour ago, STC said:

 

 

Tried this and this is so much easier. No fancy system. Just a $20 Sennheiser earphones and PC. But you can only get this right after being familiar to the particular test. Throw me a new track and I am bound to fail. 

 

foo_abx 2.0.4 report
foobar2000 v1.3.17
2018-04-30 09:11:57

File A: 23. analogue capture _ A.wav
SHA1: a48b6be748d9af8e431795a3d6792754ebbe2ae4
File B: 24. analogue capture _ B.wav
SHA1: 99fca571ad6742e44376e4d8f6f38c2164595624

Output:
DS : Primary Sound Driver
Crossfading: NO

09:11:57 : Test started.
09:12:10 : 01/01
09:12:16 : 02/02
09:12:21 : 03/03
09:12:26 : 04/04
09:12:32 : 05/05
09:12:38 : 06/06
09:12:47 : 07/07
09:12:55 : 08/08
09:13:00 : 09/09
09:13:09 : 10/10
09:13:15 : 11/11
09:13:20 : 12/12
09:13:25 : 13/13
09:14:15 : 14/14
09:14:23 : 15/15
09:14:28 : 16/16
09:14:50 : 17/17
09:14:57 : 18/18
09:15:02 : 19/19
09:15:07 : 20/20
09:15:07 : Test finished.

 ---------- 
Total: 20/20
Probability that you were guessing: 0.0%

 -- signature -- 
d53f34ba1b65e93c2976eed569f43e7bac52e42e

 

20 minutes ago, STC said:

 

I am so confident that I can do it again so that there will be no whatsoever dispute. However, this is not a real blind test since I am operating the ABX of Foobar. Will give another try with 30 trials. 

 

 

I decided to go for 50 trials.

 

foo_abx 2.0.4 report
foobar2000 v1.3.17
2018-04-30 10:19:22

File A: 23. analogue capture _ A.wav
SHA1: a48b6be748d9af8e431795a3d6792754ebbe2ae4
File B: 24. analogue capture _ B.wav
SHA1: 99fca571ad6742e44376e4d8f6f38c2164595624

Output:
DS : Primary Sound Driver
Crossfading: NO

10:19:22 : Test started.
10:19:41 : 01/01
10:20:23 : 02/02
10:20:38 : 03/03
10:21:24 : 04/04
10:21:44 : 05/05
10:21:58 : 06/06
10:22:12 : 07/07
10:22:32 : 08/08
10:22:46 : 09/09
10:22:59 : 10/10
10:23:12 : 11/11
10:23:26 : 12/12
10:23:54 : 13/13
10:24:11 : 14/14
10:24:29 : 15/15
10:24:42 : 16/16
10:25:01 : 17/17
10:25:15 : 18/18
10:25:40 : 19/19
10:25:59 : 20/20
10:26:13 : 21/21
10:26:28 : 22/22
10:26:39 : 23/23
10:26:48 : 24/24
10:27:04 : 25/25
10:27:18 : 26/26
10:27:44 : 27/27
10:27:56 : 28/28
10:28:09 : 29/29
10:28:17 : 30/30
10:28:25 : 31/31
10:28:35 : 32/32
10:28:42 : 33/33
10:28:53 : 34/34
10:29:00 : 35/35
10:29:09 : 36/36
10:29:16 : 37/37
10:29:26 : 38/38
10:29:35 : 39/39
10:29:43 : 40/40
10:29:50 : 41/41
10:30:08 : 42/42
10:30:19 : 43/43
10:30:25 : 44/44
10:30:33 : 45/45
10:30:44 : 46/46
10:30:53 : 47/47
10:31:06 : 48/48
10:31:14 : 49/49
10:31:23 : 50/50
10:31:23 : Test finished.

 ---------- 
Total: 50/50
Probability that you were guessing: 0.0%

 -- signature -- 
1bbd1bf565b3d157c728aac58e9ae4cee1d96688

[Audiophile Neuroscience]

4 minutes ago, STC said:

 

 

 

I decided to go for 50 trials.

 

foo_abx 2.0.4 report
foobar2000 v1.3.17
2018-04-30 10:19:22

File A: 23. analogue capture _ A.wav
SHA1: a48b6be748d9af8e431795a3d6792754ebbe2ae4
File B: 24. analogue capture _ B.wav
SHA1: 99fca571ad6742e44376e4d8f6f38c2164595624

Output:
DS : Primary Sound Driver
Crossfading: NO

10:19:22 : Test started.
10:19:41 : 01/01
10:20:23 : 02/02
10:20:38 : 03/03
10:21:24 : 04/04
10:21:44 : 05/05
10:21:58 : 06/06
10:22:12 : 07/07
10:22:32 : 08/08
10:22:46 : 09/09
10:22:59 : 10/10
10:23:12 : 11/11
10:23:26 : 12/12
10:23:54 : 13/13
10:24:11 : 14/14
10:24:29 : 15/15
10:24:42 : 16/16
10:25:01 : 17/17
10:25:15 : 18/18
10:25:40 : 19/19
10:25:59 : 20/20
10:26:13 : 21/21
10:26:28 : 22/22
10:26:39 : 23/23
10:26:48 : 24/24
10:27:04 : 25/25
10:27:18 : 26/26
10:27:44 : 27/27
10:27:56 : 28/28
10:28:09 : 29/29
10:28:17 : 30/30
10:28:25 : 31/31
10:28:35 : 32/32
10:28:42 : 33/33
10:28:53 : 34/34
10:29:00 : 35/35
10:29:09 : 36/36
10:29:16 : 37/37
10:29:26 : 38/38
10:29:35 : 39/39
10:29:43 : 40/40
10:29:50 : 41/41
10:30:08 : 42/42
10:30:19 : 43/43
10:30:25 : 44/44
10:30:33 : 45/45
10:30:44 : 46/46
10:30:53 : 47/47
10:31:06 : 48/48
10:31:14 : 49/49
10:31:23 : 50/50
10:31:23 : Test finished.

 ---------- 
Total: 50/50
Probability that you were guessing: 0.0%

 -- signature -- 
1bbd1bf565b3d157c728aac58e9ae4cee1d96688

 

Hi ST

That is impressive. I take your point about not being truly blinded but that could be arranged.Perhaps a silly question but care to comment on the significance/interpretation of this result?

[fas42]

Okay, before anyone gets too excited - there's a tell at 0.9 sec in, which lasts for 0.025 secs. This could be a glitch in the recording process, for some reason - and is probably enough to 'learn' to recognise. It appears to be the only one, without looking more thoroughly at the difference file.

[STC]

3 minutes ago, Audiophile Neuroscience said:

 

Hi ST

That is impressive. I take your point about not being truly blinded but that could be arranged.Perhaps a silly question but care to comment on the significance/interpretation of this result?

 

Hi AN ( I guess you want to remain anonymous  ),

 

I do blind test often but I hate doing ABX to determine small difference. It serves no purpose if we need time, practice and can only prove proof exists with familiar system and tracks. When Mani proved (despite Mansr and Esldude scepticism) that he could hear the difference I decided to pay more attention. I perceived difference but there were too minor to be consistent and yet this intrigued me. It took me one week to understand the difference.

 

I am not sure if I want to reveal it so soon because I think Mansr is close to getting the answer. I am sure you too can hear the difference. Mani already given the hint in page 50s something. I think this pill is going to be bitter as it got nothing to do with equipment. It is just your hearing sense and more concentration than needed to do brain surgery. And this definitely got nothing to do with bit-identical files.  Worth it? 

[Audiophile Neuroscience]

34 minutes ago, STC said:

 

I am not sure if I want to reveal it so soon because I think Mansr is close to getting the answer. I am sure you too can hear the difference. Mani already given the hint in page 50s something. I think this pill is going to be bitter as it got nothing to do with equipment. It is just your hearing sense and more concentration than needed to do brain surgery. And this definitely got nothing to do with bit-identical files.  Worth it? 

 

Okay, color me intrigued.

[acg]

39 minutes ago, fas42 said:

Okay, before anyone gets too excited - there's a tell at 0.9 sec in, which lasts for 0.025 secs. This could be a glitch in the recording process, for some reason - and is probably enough to 'learn' to recognise. It appears to be the only one, without looking more thoroughly at the difference file.

 

STC did 50 ABX comparisons in 12 minutes...that's 14 seconds per ABX...it appears as though he could hear that "tell"