• 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!

Why aren’t there more university level controlled tests?

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
6,123
Likes
6,202
I would be far-far more useful experimenting with stuff that are not so defined yet,as stereo perception (3-D perception better) ,co-sensory associations,etc.
I have asked for published papers about strictly defined thresholds and all I get back is either the 0.1% THD+N pro standard or an 115db THD+N absolute which is thousands of times lower.
So...
 

DanielT

Major Contributor
Joined
Oct 10, 2020
Messages
4,829
Likes
4,766
Location
Sweden - Слава Україні
I would be far-far more useful experimenting with stuff that are not so defined yet,as stereo perception (3-D perception better) ,co-sensory associations,etc.
What you say is probably close to what actually being researched in the interdisciplinary research field of psychoacoustics.:)

I have asked for published papers about strictly defined thresholds and all I get back is either the 0.1% THD+N pro standard or an 115db THD+N absolute which is thousands of times lower.
So...
Are there any? Isn't that very individual? But okay, it is possible to plot in a diagram the results of the participants, which is done in this test:


Edit:
How to read those results? Is it due to individual differences regarding detecting distortion? Or can the differences be explained by the fact that the test is done with different headphones, speakers? Or a combination of both aspects?
 
Last edited:

DanielT

Major Contributor
Joined
Oct 10, 2020
Messages
4,829
Likes
4,766
Location
Sweden - Слава Україні
It is rather 0.1% for THD and 115dB for N.
Which then has to be tested with speakers that have below 0.1% distortion, which not many have.

Headphones probably cannot be used in such a test because below 0.1% distortion is a criterion. Do such low distortion headphones exist? I'm not a headphone guy so I don't know.
 

MaxwellsEq

Major Contributor
Joined
Aug 18, 2020
Messages
1,750
Likes
2,645
Comparing DACs would have got university research budgets in the 60s and 70, when telephone digitisation was developing. By the 70s, pretty much all the academic models were confirmed in the lab.

Current research involves neural models. Here's a fascinating paper on how the brain works differently when decoding speech vs music.

 

popej

Active Member
Joined
Jul 13, 2023
Messages
281
Likes
185
As an example, how long would it take to do a proper, double-blind, controlled test on a handful of DACs with varying degrees of SNR, THD+N, jitter, etc.?
That implicit suggest, that this kind of consumer products aren't designed on scientific basis. IMO it is not that interested scientists and designers miss knowledge, but consumers.
 

Down South

Active Member
Joined
Jul 20, 2022
Messages
137
Likes
61
Yorkshire Mouth,
the reaction to your initial post was 'as expected'. I take issue with your start point. When the Japanese first presented their direct drive decks to the world at the beginning of the 70s' most of the West (Germany excepted) ruthlessly and in many cases racially reacted to these totally superior turntables. The Japanese had started by asking - what is the function of a t/t - to rotate at precise speeds and to operate silently or as close to. So their teams of engineers stripped away all the preconceived mentality and literally started from ground zero. More than 50 years later present day t/ts have not advanced one iota in engineering terms/precise speed control or silent or near silent operation.

You started with DACS why? surely the place to start is ground zero. What is the most important part of a house - not what kind of windows/decoration/kitchen bla bla bla. The most important part of a house if you want to achieve a long lasting structure is the part you cannot see - the sub-strata the house is to be built on and secondly the foundations. If you don't take account of the sub strata and how you create the foundations then you may well find out as many across Europe have this year that your expensive house has serious structural problems. Houses are built across the world using crap outdated materials and methodology. Raft foundations would have taken care of structural problems because of very dry/ very wet seasons. Temperature control using thermally efficient materials would mean far less heating/cooling would be needed. Using fire proof material would have meant the devastation of property in Australia, the US, Europe couldn't have happened. It's no good building thermally efficient housing, commercial property if you don't deal with 'sick building syndrome' caused by there being no 'air change systems' designed in from the planning stage.

psychoacoustics - what a stupid place to start. Start at the beginning - electronically produced music systems all require electricity to function yes? How often have I heard the futile, negative comments about the electricity supply - so it's not clean, stable, or supplied at the correct voltage. Isn't this where you start. Isn't this the foundation on which to create an optimal system. Ah but it's s not sexy, no but it is essential if you want to optimize your listening experience and for your equipment to operate at it's maximum potential.

Then there are the connectors - RCA/XLR etc. when were these developed - some pre WW11 are they optimum? no, not at all. How does a musical signal travel from one piece of equipment to another. What is the optimum material to use that will interfere the least with this A-B travel. To waste time and money on dealing with these issues so that a bat can optimise it's hearing capabilities is irrational but not at the level that humans can hear. Once you start investigating these issues you realise very quickly the so called scientific gurus/know it alls' pay no attention to these issues what so ever but crap in/crap out applies.

I don't think that universities are the place where this research should take place, it's technical colleges that should be used. Yes funding is, like it or not an important point that many have raised and yes it's not in the interests of companies selling $1k + power cables to fund this research. PCBs are these the best way to transport electronic signals, or as they are constructed. Are they used and in their present form because this is 'convenient and cheap'.

The best form of dielectric is air - is this true, then just look at what kinds of dielectric are used - convenience/cost being the only considerations. Maybe there is a technical college somewhere that has a head of dept. with an open mind to set their students/post grad students a project to set about designing connectors that interfere as little as possible with the weak musical information passing through. that using a certain type of fluorocarbon tubing made a difference. That designing in air as part of the design made a difference that can actually be heard. Blind testing yes but a waste of time until these primary questions have been asked and effectively answered. To produce any digital equipment that operates far beyond the hearing capability of bipeds is just plain stupid. How much equipment is made where heat dissipation is an after thought if at all. White papers were written in the 1950s about the effect excessive heat had on longevity and function of components. Some idiot on diyaudio made a statement that mosfets sounded better when operating when hot. He never did an experiment with mosfets operating using a cooling method - so how did he know what he stated was a fact.

Sadly there is so much arrogant 'received wisdom' that with examination turns out to be pure b/s. It really should be a case of accept nothing that cannot be repeatedly proven.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,511
Likes
25,350
Location
Alfred, NY
I have learned a lot from this thread, so I am now redirecting my students to confirming that acids neutralize bases and that all electrons have equal charges.
 

ahofer

Master Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
5,045
Likes
9,152
Location
New York City
Funny you don’t see a lot of experiments validating that gases expand as pressure decreases either.

I mean, 16/44 and MP3 are direct results of the necessary audiological research, done ages ago.
 
D

Deleted member 48726

Guest
Funny you don’t see a lot of experiments validating that gases expand as pressure decreases either.

I mean, 16/44 and MP3 are direct results of the necessary audiological research, done ages ago.
Yeah, it is actually older than I would've guessed.

1707581867635.png
 

ahofer

Master Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
5,045
Likes
9,152
Location
New York City
Yeah, it is actually older than I would've guessed.

View attachment 348718
So they designed a format around known, tested, audiological limits, it should not be a surprise that humans can't tell the difference.

This is like, in my business, when people design trading algorithms around past market behavior and then demonstrate that they back-test well. Of course they do, you are just running the same analysis back-to-front.
 

ahofer

Master Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
5,045
Likes
9,152
Location
New York City
OP
Yorkshire Mouth

Yorkshire Mouth

Major Contributor
Joined
Apr 22, 2020
Messages
1,354
Likes
1,296
Location
God's County - Yorkshire
The point is, after certain things are proven (example, humans can’t hear beyond c.110 dB dynamic range), we can extrapolate from that the idea that humans can’t hear a difference in dynamic range between two DACs with a SNR beyond that.

But science usually repeats that basic concept in particular situations.

And there’s the other side, too. We know humans can’t hear beyond that dynamic range. But the question remains, can humans hear beyond 105 db, 100 dB, etc., in ‘real world’ listening conditions?

Is 110 dB actually necessary? If not, then what is? AFAIK, there isn’t a solid, settled scientific view on this.
 
D

Deleted member 48726

Guest
The point is, after certain things are proven (example, humans can’t hear beyond c.110 dB dynamic range), we can extrapolate from that the idea that humans can’t hear a difference in dynamic range between two DACs with a SNR beyond that.

But science usually repeats that basic concept in particular situations.

And there’s the other side, too. We know humans can’t hear beyond that dynamic range. But the question remains, can humans hear beyond 105 db, 100 dB, etc., in ‘real world’ listening conditions?

Is 110 dB actually necessary? If not, then what is? AFAIK, there isn’t a solid, settled scientific view on this.
That's good questions.

But how to define "real world" listening conditions? -What is the average noise floor?
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,771
Likes
37,635
That's good questions.

But how to define "real world" listening conditions? -What is the average noise floor?
It is unlikely to be below 30 db SPL. What is more important is the spectrum of that noise floor. Rooms with a 40 db noise floor if looked at with an FFT may be within 10 db of the limit of our hearing up in the 3-5 khz range. Remember we can hear into the noise a little bit. Most of the noise is lower frequency with a pinkish profile. So where we can hear most sensitively the needed noise floor is actually pretty low. So asking for at least 100 db range to work with is not unreasonable. Distortion on the other hand can be much higher and we don't notice.
 

ahofer

Master Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
5,045
Likes
9,152
Location
New York City
You want a standard that is flexible for different conditions, including isolating headphones. A “strict” standard, as it were….

 
Top Bottom