• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Could someone help me to think through my ABX result using Bayesian reasoning?

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,223
Likes
17,799
Location
Netherlands
I am sure you‘ve heard of covert information channels. And the Tempest spec…
Sure, if you think that there is a 1 in 100 chance that this is going on we might consider it a serious comment.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
That is not how jitter works. You cannot have this in one channel only. How do you you even know it was jitter?
By the 1/4 sample rate tone used to see jitter. And yes it did work that way in one channel only.

Have I mentioned the DVD player that each time you start a new track has most horrid jitter which gradually subsides and has very little right around the 15 second mark after starting or changing tracks.
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,223
Likes
17,799
Location
Netherlands
By the 1/4 sample rate tone used to see jitter. And yes it did work that way in one channel only.
How would that be possible then? The data of both channels goes into the dac chip basically together (though alternated), and is synced using a single channel clock. Only once both channels are clocked in, the sample is “played”.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
How would that be possible then? The data of both channels goes into the dac chip basically together (though alternated), and is synced using a single channel clock. Only once both channels are clocked in, the sample is “played”.
I wished I knew. I measured a few times and using different ADC's. It acted the same in all cases.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
9/10 can’t simply be ignored, and if confirmed by the test rerun what else explains it?

It was actually 10 in 12 and if Mani tests again and gets 18 out of 20 (witnessed test by someone that knows how to test) and the analog signal is recorded as well then there is no getting around it and further research is warranted and will require Mani's cooperation in that case as he seems able to hear it. There is still doubt because of the small sample size (when not counting any X's after the first X in the first 20 attempts.

looking at 30 attempts it surely looks like (lucky) guessing.

So no conclusions can be drawn other than what's already been mentioned.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I wished I knew. I measured a few times and using different ADC's. It acted the same in all cases.

Chances are some of the digital signals or power supply nasties made it into one channel only due to layout issues.
The layout of a board with digital and analog signals and one power supply situated on a single board, especially with 3 or more ground planes or power supply routing differently over the board can be challenging especially when the grounds are connected in the following amp via RCA shields.
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,223
Likes
17,799
Location
Netherlands
Chances are some of the digital signals or power supply nasties made it into one channel only due to layout issues.
That might be a plausible explanation. But not something that would manifest in a SPDIF or USB connection.

Even if there were different jitter profiles on the SPDIF output due to “loading”, it must have been one hell of a crappy DAC to make this audible.

With his current Okto, we already know the jitter performance: it’s nowhere near audibility.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
But not something that would manifest in a SPDIF or USB connection.

I would agree, Blumlein mentioned he had this with an old M-Audio card. That makes me think it was a PC card.
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,223
Likes
17,799
Location
Netherlands
I would agree, Blumlein mentioned he had this with an old M-Audio card. That makes me think it was a PC card.
Yes he did. Point is, we can’t just translate this to @manisandher’s case.
 

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
Did I miss it or was the harware setup ever disclosed?
- machine type
- OS
- Player (and settings) -- how was bit-identity verified?
- Connection to DAC
- DAC (and USB type)
- Amp
- Speaker/Headphones
- Wiring and Grounding Scheme details
- ...
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
What are you talking about?

It would be really easy to measure the analog output in both 'conditions' using a steady state signal. I would assume, given the equipment Mani owns, he doesn't use an old PC card as a DAC but mentioned this:
DAC was connected to PC via BNC spdif. So that means external DAC with its own power supply and chances are, given BNC SPDIF that the source signal was galvanically separated (the more expensive DACs usually had a data transformer in its path.

It is a common belief that USB means poor sound quality and coax SPDIF 'sounds better' in audiophile circles.
There is nothing in Amirs tests showing this is the case in reality though. Aside from a few DACs with poor USB inputs (mostly older DACs) the USB performance is clearly 'better'.
 
Last edited:

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,223
Likes
17,799
Location
Netherlands
What are you talking about?
We’re have a sound card with analog audio on the one side, and an spdif + DAC on the other. Those setups are vastly different when trying to identify sources of audible differences.
 

charleski

Major Contributor
Joined
Dec 15, 2019
Messages
1,098
Likes
2,239
Location
Manchester UK
Interesting hypothesis as we have all had experiences as described. While it might be tempting to ascribe these perceptual changes to mood, I don't believe that accounts for it entirely. There are of course two muscles that influence the transfer function of the ossicular chain, the tensor tympani and the stapedius, both of which serve to decouple the tympanic membrane from the inner ear (according to some sources, only the stapedius functions as such in humans) and prevent damage to the hair cells (which of course is why alcohol inebriation (relaxes the stapedius) and high volumes are a bad mix. I've seen papers that measure the transfer function in freshly dead humans and living cats, but unaware of any that look at that as a function of muscular tone. The other muscle that may impact transmission is the tensor veli palatini muscle that helps to open the eustachian tube ("popping the eardrums" to equalize pressure between middle ear and ambient pressure). In any event there is a lot going on well before we get to "mood," much of which we are unaware of.

If the muscles are still in good shape, it would seem that electrical stimulation and contraction might afford a good look at the phenomenon. There may be such studies but available only in abstract. And that is a subject I could rant about for hours.
There are a small number of people who are able to contract their tensor tympani voluntarily. Audiometric studies show this has dramatic effects on the transfer function for both air and bone conduction, see Wickens et al. 2017 for instance. But most studies of this system are concerned with tensor tympani myoclonus, which may be a cause of some forms of tinnitus.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
It would be really easy to measure the analog output in both 'conditions' using a steady state signal. I would assume, given the equipment Mani owns, he doesn't use an old PC card as a DAC but mentioned this:
DAC was connected to PC via BNC spdif. So that means external DAC with its own power supply and chances are, given BNC SPDIF that the source signal was galvanically separated (the more expensive DACs usually had a data transformer in its path.

It is a common belief that USB means poor sound quality and coax SPDIF 'sounds better' in audiophile circles.
There is nothing in Amirs tests showing this is the case in reality though. Aside from a few DACs with poor USB inputs (mostly older DACs) the USB performance is clearly 'better'.
Wasn't there an Altmann Attraction used for part of this? I know Mani has(or had) one.
 

Raindog123

Major Contributor
Joined
Oct 23, 2020
Messages
1,599
Likes
3,554
Location
Melbourne, FL, USA
Did I miss it or was the harware setup ever disclosed?

Not really. I think the initial excuse for that was that Mani wanted to frame it all as a “statistical exercise of a Bayesian flavor”. Meaning, his approach was: “Forget about the hardware, it is irrelevant to my question. My question is, if I have a very strong believe in something, anything (his ‘high prior‘ of ’0.9’) backed up by some experimental result, rather weak but favorable (his 9/10, or 10/12 however you look at it)… so, would combining these two be a ’convincing scientific proof’ of this something?”

Whereas, the answers in the thread were to two separate, very distinct questions. The first answer - to Mani’s above question - pretty much is: “No. Dragging Bayesian analysis into such weak (small number of samples) experiment is wrong, and as a result you/we cannot conclude with any reasonable degree of confidence that the event is true”...

Meanwhile the answer to the other question — of whether Mani can or cannot hear difference between two particular settings of a particular SW music player, on a particular PC, connected to a particular DAC over a copper S/PDIF, with the digital signal forked, recorded and verified in such and such particular way, and the DAC output amplified and played through a particular analog chain — the answer to this question is: “We do not know. We need more data — a better experiment definition and description, better preparation, better control, more raw data points, and better analysis of those... After we have all this, only then we can with certainty conclude whether Mani does hear something, and if yes look for the underlying physical effect — S/PDIF clock somehow being affected by PC‘s load, or some signal interference/pickup, or mind connection.”
 
Last edited:
OP
manisandher

manisandher

Addicted to Fun and Learning
Joined
Nov 6, 2016
Messages
652
Likes
607
Location
Royal Leamington Spa, UK
Not really. I think the initial excuse for that was that Mani wanted to frame it all as a “statistical exercise of a Bayesian flavor”. Meaning, his approach was: “Forget about the hardware, it is irrelevant to my question. My question is, if I Ihave a very strong believe in something, anything (his ‘high prior‘ of ’0.9’) backed up by some experimental result, rather weak but favorable (his 9/10, or 10/12 however you look at it)… so, would combining these two be a ’convincing scientific proof’ of this something?”

Whereas, the answers in the thread were to two separate, very distinct questions. The first answer - to Mani’s above question - pretty much is: “No. Dragging Bayesian analysis into such weak (small number of samples) experiment is wrong, and as a result you/we cannot conclude with any reasonable degree of confidence that the event is true”...

Meanwhile the answer to the other question — of whether Mani can or cannot hear difference between two particular settings of a particular SW music player, on a particular PC, connected to a particular DAC over a copper S/PDIF, with the digital signal forked, recorded and verified in such and such particular way, and the DAC output amplified and played through a particular analog chain — the answer to this question is: “We do not know. We need more data — a better experiment definition and description, better preparation, better control, more raw data points, and better analysis of those... After we have all this, only then we can with certainty conclude whether Mani does hear something, and if yes look for the underlying physical effect — S/PDIF clock somehow being affected by PC‘s load, or some signal interference/pickup, or mind connection.”

Wow! Pretty much bang on!
 
OP
manisandher

manisandher

Addicted to Fun and Learning
Joined
Nov 6, 2016
Messages
652
Likes
607
Location
Royal Leamington Spa, UK
Did I miss it or was the harware setup ever disclosed?
- machine type
- OS
- Player (and settings) -- how was bit-identity verified?
- Connection to DAC
- DAC (and USB type)
- Amp
- Speaker/Headphones
- Wiring and Grounding Scheme details
- ...

FWIW:

ABX setup.jpg

All equipment fed off single dedicated AC mains circuit.
 
OP
manisandher

manisandher

Addicted to Fun and Learning
Joined
Nov 6, 2016
Messages
652
Likes
607
Location
Royal Leamington Spa, UK
FWIW 2

We took analogue captures of the DAC outputting a 10kHz sine. Here is Mans's analysis:

10k FFT 16m.png

A is SFS=0.1
B is SFS=200
 
Top Bottom