• 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!

Post research here that casts doubt on ASR objectivism

ahofer

Major Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
4,947
Likes
8,694
Location
New York City
The prevailing null hypothesis here at ASR can be summarized as follows (I'll edit with any suggestions):
  • There's very little difference between competently designed and adequately powered electronics (amps, DACs, streamers), and nothing audible in electronics that is unmeasurable.
  • Measurement instruments are vastly more sensitive than the human ear, and we have yet to identify strictly audible phenomena that can't be measured.* there is nothing strictly audible that can't be measured.*
  • Even where there are audible differences between electronics, they are hard to hear, and many would have trouble getting more than 70%c orrect in a blind trial a likelihood ratio greater than 8 (of audible differences vs none) in ABX trials won't happen, except:
  • Audible differences between amplifiers can indeed arise from (a lack of) power, differing frequency response (>0.2db), frequency response effects of impedance mismatch/poor design, or high distortion (typically tubes, > 0.5%, but varies by type and in which frequency). Therefore a proper suite of well-known measurements from the amp and companion speakers will describe the entire audible story with amplifiers.
  • Cables make no audible difference are audibly transparent given LCR measurements close to the generic 12G cable here. they are copper, proper gage, stranded, and not incredibly long; However, cables can also introduce potentially audible frequency response deviations when they are longer, higher in inductance or DC resistance, are used with amplifiers with higher output impedance, and/or are supplying loudspeakers with large deviations in their impedance curves. Expensive cables are snake oil. likelihood ratio greater than 8 (of audible differences vs none, given compatible measurements and lengths) in ABX trials won't happen
  • Many other tweeks, including exotic digital cables, disks and stones placed around the room, stickers, phone calls, hockey pucks, green magic markers on CDs, cable risers, etc., are a waste of time and also likely to be snake oil. A likelihood ratio greater than 8 (of audible differences vs none) in ABX trials won't happen.
  • Being more expensive does not necessarily mean higher fidelity
*There are more psychological effects that affect your perception of sound than can be counted, however.

We believe these It appears that these positions have yet to be refuted by research or controlled blind comparisons. In fact, we tend to believe they appear to be *supported* by the majority of audio engineering research, and many of us at ASR simply view them as simple truths at this point.

This is the place you can submit any research you have the might cause us to that may reject these null hypotheses. Many of us would be interested to see what you come up with.

We are only interested in published research and controlled tests with transparent and reproducible methods. Anecdotal sighted listening tests will not convince anyone. Nor will mistreating Shakespeare with vague references to the unsolved mysteries of the universe ("there are more things on heaven and earth...") or quantum mechanics.

[last updated on 1/14/2022 - qualifiers to amplifiers, emphasis]
 
Last edited:

BDWoody

Chief Cat Herder
Moderator
Forum Donor
Joined
Jan 9, 2019
Messages
6,948
Likes
22,625
Location
Mid-Atlantic, USA. (Maryland)
This will be interesting...

One picky point... I'd say between 'competently designed and appropriate to the task.'

Easy to come up with a well made low power amp that isn't suitable to drive Salon's to reference level...
 

BDWoody

Chief Cat Herder
Moderator
Forum Donor
Joined
Jan 9, 2019
Messages
6,948
Likes
22,625
Location
Mid-Atlantic, USA. (Maryland)
Some differences can obviously be heard but then the issue is do we have a measurement that correlate to it.

So far, yes.

I don't believe there has ever been a blind test result that couldn't be explained with existing measurements, has there? I think we all would have noticed that one.
 

dajoe

Member
Joined
Sep 7, 2020
Messages
12
Likes
6
Intersample peaks seem to be an interesting phenomenon. I understand that such peaks are only an issue when during production the loudness is pushed too hard to 0dBfs and if the DAC doesn't allow for enough headroom in case the analog waveform might clip.

My experience is that high hats often do sound unnatural especially in lossy audio but sometimes also in lossless. Maybe that's what I'm hearing.

It seems straightforward that high sample rate audio would be less prone to such an issue.

But overall I would say that it is more a mastering issue than a DAC issue.
 

Gregss

Active Member
Joined
Jan 26, 2021
Messages
160
Likes
208
Agree with the fact that there can be differences, but when it is just an opinion without any type of decent testing:

Hitchen’s Razor: That which can be claimed without evidence may also be dismissed without evidence.
 

Kvalsvoll

Addicted to Fun and Learning
Audio Company
Joined
Apr 25, 2019
Messages
878
Likes
1,643
Location
Norway
The prevailing null hypothesis here at ASR can be summarized as follows (I'll edit with any suggestions):
  • There's very little difference between competently designed and adequately powered electronics (amps, DACs, streamers), and nothing audible in electronics that is unmeasurable.
  • Measurement instruments are more sensitive than the human ear, and there is nothing strictly audible that can't be measured.*
  • Even where there are audible differences between electronics, they are hard to hear, and many would have trouble getting a blind comparison more than 70% correct
  • Cables make no audible difference given they are copper, proper gage, stranded, and not incredibly long; expensive cables are snake oil
  • Tweeks, generally speaking, are a waste of time and also likely to be snake oil, as are special digital cables
  • Being more expensive does not necessarily mean higher fidelity
*There are more psychological effects that affect your perception of sound than can be counted, however.

We believe these positions have yet to be refuted by research or controlled blind comparisons. In fact, we tend to believe they are *supported* by the majority of audio engineering research.

This is the place you can submit any research you have the might cause us to reject these null hypotheses. Many of us would be interested to see what you come up with.

We are only interested in published research and controlled tests with transparent and reproducible methods. Anecdotal sighted listening tests will not convince anyone. Nor will mistreating Shakespeare with vague references to the unsolved mysteries of the universe ("there are more things on heaven and earth...") or quantum mechanics.

Those statements do not belong to this forum or any other brand or entity, they are general statements describing basic principles that are all known and proven.

"It is known."
 

escksu

Addicted to Fun and Learning
Joined
Jul 16, 2020
Messages
965
Likes
397
  • Cables make no audible difference given they are copper, proper gage, stranded, and not incredibly long; expensive cables are snake oil.
I would say this statement needs to be better defined since we are talking about science and measurements here.

1. You mention they are copper. How about plated copper or copper alloys? Since 100% pure copper does not exist, at what purity do you consider it copper?

2. Proper gauge. What is proper gauge? Could you also show me a test where cable out this "proper gauge" produces an audible difference?

3. I assume stranded means anything that is not solid core. Any measurements to show that there is a difference between solid core and stranded? Do number of strands make any difference?
 

Phorize

Major Contributor
Forum Donor
Joined
Apr 26, 2019
Messages
1,533
Likes
2,060
Location
U.K
The prevailing null hypothesis here at ASR can be summarized as follows (I'll edit with any suggestions):
  • There's very little difference between competently designed and adequately powered electronics (amps, DACs, streamers), and nothing audible in electronics that is unmeasurable.
  • Measurement instruments are more sensitive than the human ear, and there is nothing strictly audible that can't be measured.*
  • Even where there are audible differences between electronics, they are hard to hear, and many would have trouble getting a blind comparison more than 70% correct
  • Cables make no audible difference given they are copper, proper gage, stranded, and not incredibly long; expensive cables are snake oil
  • Tweeks, generally speaking, are a waste of time and also likely to be snake oil, as are special digital cables
  • Being more expensive does not necessarily mean higher fidelity
*There are more psychological effects that affect your perception of sound than can be counted, however.

We believe these positions have yet to be refuted by research or controlled blind comparisons. In fact, we tend to believe they are *supported* by the majority of audio engineering research.

This is the place you can submit any research you have the might cause us to reject these null hypotheses. Many of us would be interested to see what you come up with.

We are only interested in published research and controlled tests with transparent and reproducible methods. Anecdotal sighted listening tests will not convince anyone. Nor will mistreating Shakespeare with vague references to the unsolved mysteries of the universe ("there are more things on heaven and earth...") or quantum mechanics.

Do spoof papers lampooning the insistence of randomised controlled trials in health academia count?

https://www.bmj.com/content/327/7429/1459
Full copy:

https://journals.sagepub.com/doi/pdf/10.1177/154510970400300401
 

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,700
Location
Hampshire
I would say this statement needs to be better defined since we are talking about science and measurements here.

1. You mention they are copper. How about plated copper or copper alloys? Since 100% pure copper does not exist, at what purity do you consider it copper?

2. Proper gauge. What is proper gauge? Could you also show me a test where cable out this "proper gauge" produces an audible difference?
These would be better combined to simply say cables with resistance, capacitance, and inductance suitable for the application. Material is irrelevant. Copper is often used in cables since its electrical (low resistivity) and mechanical (malleable, reasonably strong) properties are both beneficial. When higher tensile strength is required, copper-clad steel is a good option. Power lines are typically aluminium since it is cheap and lightweight. Proper termination is tricky, however, making it a poor choice for things like audio interconnects.

3. I assume stranded means anything that is not solid core. Any measurements to show that there is a difference between solid core and stranded? Do number of strands make any difference?
Stranded wire is more flexible than solid. That's all there is to it.
 

Plcamp

Addicted to Fun and Learning
Joined
Jul 6, 2020
Messages
860
Likes
1,315
Location
Ottawa
I was going to entirely crush your postulates on every point…

…but then you went and ruled out quantum mechanics.
 

Plcamp

Addicted to Fun and Learning
Joined
Jul 6, 2020
Messages
860
Likes
1,315
Location
Ottawa
I would say this statement needs to be better defined since we are talking about science and measurements here.

1. You mention they are copper. How about plated copper or copper alloys? Since 100% pure copper does not exist, at what purity do you consider it copper?

2. Proper gauge. What is proper gauge? Could you also show me a test where cable out this "proper gauge" produces an audible difference?

3. I assume stranded means anything that is not solid core. Any measurements to show that there is a difference between solid core and stranded? Do number of strands make any difference?

I suggest this article scientifically answers questions about what actually matters about speaker cables.

https://benchmarkmedia.com/blogs/ap...unYd-kvCSe7egA-BHTVbqkiwN2ns7txBoCGBoQAvD_BwE
 

Ron Texas

Master Contributor
Joined
Jun 10, 2018
Messages
6,074
Likes
8,906
I can hear all kindds of things. Bill Gates has planted subliminal messages in sorts of music. LOL.
 

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,063
Likes
14,694
Where does the 70% come from? I find the 3rd bullet more open than others to being shot down.
 

Frgirard

Major Contributor
Joined
Apr 2, 2021
Messages
1,737
Likes
1,040
Intersample peaks seem to be an interesting phenomenon. I understand that such peaks are only an issue when during production the loudness is pushed too hard to 0dBfs and if the DAC doesn't allow for enough headroom in case the analog waveform might clip.

My experience is that high hats often do sound unnatural especially in lossy audio but sometimes also in lossless. Maybe that's what I'm hearing.

It seems straightforward that high sample rate audio would be less prone to such an issue.

But overall I would say that it is more a mastering issue than a DAC issue.
The isp is not an issue in mastering.
Isp limiter exists.
A flexibility of 1 or 2 dB avoid the Isp.
Isp is the guarantee the client will be happy.
The debate is: the distortion is audible or not. The loudnesswar compression generate distortion by changing the tone so in reality isp is non audible.

The decrease of the output level of the player fixed the issue.
 

raistlin65

Major Contributor
Forum Donor
Joined
Nov 13, 2019
Messages
2,279
Likes
3,421
Location
Grand Rapids, MI
Where does the 70% come from? I find the 3rd bullet more open than others to being shot down.

And not just the 70%. Making the sweeping generalization that audible differences in electronics are always difficult to hear is a problematic statement. Better to say, "many are hard to hear."
 

Beershaun

Major Contributor
Forum Donor
Joined
Oct 3, 2019
Messages
1,864
Likes
1,910
Why is the op intentionally Vague about the measurement levels in their statements when Amir is not about his audible threshold target measurements for all his tests? A null hypothesis in statistics is very carefully and specifically worded as to be testable and the OPs post is not. Or maybe it is specifically worded as to not be testable so there can be yet another thread on the internet that anonymous people can argue with each other about. I want my 5 minutes I spent reading this thread back.
 
Top Bottom