Listening tests

Status
Not open for further replies.
Jump to Latest
#62
Just saw an article about a week ago, not sure where I saw it, but the basic conclusion was that the mind actually processes information at a much higher speed than was agree upon before. So even our understanding of how we process information is still changing and many surprises are still in store. So those who base things like what level of jitter is audible may be off by a factor of more.
Click to expand...

Our minds are very, very clever at processing the sound coming in, and will "dig out the dirt" if there is some to be found, if we go about it with concentration, in a focused way.

My suspicion is that "good sound" occurs when the quantity of 'valuable' information, sounds directly relevant to the musical event recorded, reaches a balance point with respect to the "muck" - insufficient useful data, and too much irrelevant "noise", the sound is rejected as being bad - there's no pleasure in it ...
 
#64
Waly said:
The 'scientists' would be more than happy to investigate "these sort of areas", provided you can prove they exist.

The other way around, unfortunately, they can't prove a negative.
Click to expand...
Unfortunately, we have a logical impossibility - if I were able to "prove they exist" then I would be one of those investigative scientists ...

Science progresses because people want to investigate empirical data, not because someone has "proven" that the data is real. Medical research takes serious account of people complaining of sore or bad backs, they don't ask them to "prove" their back is a problem ...
 
#65
jcx said:
as far as I can find actual professional publication controlled listening test thresholds for jitter audibility with RedBook CD is 10s-100s of ns
Click to expand...

Your data is old. Which is one of my points -- we keep learning and thresholds keep dropping. 🙂
But, has nothing to do with CFA and why it would be preferred by so many.

Studies by AES - thresholds of audibility -- 120psec p-p jitter for 16 bits and 8 ps for 20 bits. 20psec p-p of data-correlated jitter for certain freqs.

So, ? Which differences in amps (any topology) can be audible? Other than - FResponse and THD and IM and S/N and Zo? Any which a CFA would be inherently better at doing?....

[not same thing as could possibly be made to be similar to CFA, somehow.]

Thx-RNMarsh
 
Last edited:
#68
please provide references for Listening tests? - I've seen the estimates

in fact I've just skimmed the latest article found in a AES e-library search "jitter audibility listening test" - hosted by wolfson: http://www.wolfsonmicro.com/documents/uploads/misc/en/Specifying_Jitter_Performance.pdf

which states up front that it is not about listening tests, but does include a ref list

next hit that includes actual listening result is Benjamin,Gannon1998, summarized by a 3rd party as:
What level of jitter do careful experiments tell us we can actually hear? A careful study found that none of their test subjects could detect random jitter of 250ns or less. That is several thousand times higher than the levels we can hope to approach and that will start to introduce real, measurable, errors with the worst case signal. Other studies have come to similar conclusions.

What sort of jitter do digital sources you can go out and buy have? Well, the validity of a lot of the data I can find trawling the Internet isn't that clear, but it looks like a humble Squeezebox may deliver around 60ps, with "a cheap CD player" coming in at 270ps and "the worst ever measured" at 1100ps. So - the "worst ever measured" is about 1ns ... 250 times lower than seems to be audible.

Wow! It seems very unlikely that you will hear any difference between sources due to their jitter performance.
Click to expand...
 
Last edited:
#72
I talk about a real difference, and then it's discredited as being a drawback..?? How would you measure the expansive nature of the current on demand on the CFA devices, when all tests are more or less static...??

Why is it that altering the mechanics of an amplifier changes performance..??

Recently I build a really good phono-stage, then I boxed it and the magic was gone, why was that..?? what was the ill effect of boxing..?? and how do you measure the effects of that..??

Why is my system performing better late evening and nights..??

Trouble is that we have such a sensible hearing that everything matters and neglecting that just makes us shut our eyes and limit ourself in our engineering.
We may not have all the answers, but if we just approach everything with limited tools and an uncanny lust and desire for "simple" math and theorems that are nothing but a coarse caricature of reality.

Then much chance of progress is simply lost..!!

I think we all can agree that a simulation is nothing but a coarse scenario of a real circuits performance, If math really was "the" only solution then why are simulations not true and accurate..??
 
#74
RNMarsh said:
Studies by AES - thresholds of audibility -- 120psec p-p jitter for 16 bits and 8 ps for 20 bits. 20psec p-p of data-correlated jitter for certain freqs.
Click to expand...
I too would like a link to these studies.

My own small research suggests what jcx summarized for Benjamin & Gannon are the relevant ones.

This is relevant to the thread as it touches on Listening Tests; the one difference that IMHO, hasn't been explored properly between VFAs & CFAs.

IMveryHO, this might be the most important difference of all 🙂
 
#75
I'm not going to add fuel to this.

The thresholds of freq response amplitude variations has dropped over the decades. Distortion thresholds have dropped. Many things have been identified. So, there is something here that has yet been itdentified as the contributor to the many 'patients' description/symptoms. We dont know what it is and have no clue apparently but there are some parameters we can test for thresholds of audibility..... the first one is Slew rate... and NOT based on what the input signal requires but what the listener can hear.... do your DBLT on it, if you like. We could make an amp that only changes SR and not distortion etc. I dont know how best to do that but I am sure one of you do. We only have an assumption about what the amp needs to do in order to accurately follow the input signal. test the assumption via audibility tests.

Or what is the audibility of PSRR? Have we tested for that audibility threshold? Do other topologies affect the threshold of audibility for PSRR? How much PSRR do we need to be inaudible? There are so many others we also dont know and thus fill in the blanks with assumptions. I am sure others can think of some assumptions that have not been tested for audibility thresholds. new info about digital clipping distortion..... what is the audible threshold re digital clipping and how does it vary with interpolation algorithms?

THx-RNMarsh
 
Last edited:
#76
RNMarsh said:
I'm not going to add fuel to this.

The thresholds of freq response amplitude variations has dropped over the decades. Distortion thresholds have dropped. Many things have been identified. So, there is something here that has yet been itdentified as the contributor to the many 'patients' description/symptoms. We dont know what it is and have no clue apparently but there are some parameters we can test for thresholds of audibility..... the first one is Slew rate... and NOT based on what the input signal requires but what the listener can hear.... do your DBLT on it, if you like. We could make an amp that only changes SR and not distortion etc. I dont know how best to do that but I am sure one of you do. We only have an assumption about what the amp needs to do in order to accurately follow the input signal. test the assumption via audibility tests.

Or what is the audibility of PSRR? Have we tested for that? What is the PSRR threshold of audibility? Do other topologies affect the threshold of audibility for PSRR? How much PSRR do we need to be inaudible? There are so many others we also dont know and thus fill in the blanks with assumptions.

THx-RNMarsh
Click to expand...

But many engineers try to make other people follow to their taste. They think they are superior and ignore other people taste. The truth is many engineers don't have any taste, they only have logic brain and failed to understand about taste.
 
#77
A factoid --- the more time you spend on a single subject... the more you learn and know about that one subject.... BUT that also means less time is spent learning about something else. So the smarter one is in one area, the dumber he/she is in other areas.

[there is a message somewhere in this for the engineers/scientists here]

Thx-RNMarsh
 
#79
Look at your ac line noise levels during day and during night. Especially, if you share power with neighbors -- like in an apartment/condo/industrial invironment. I found the noise level drops late at night. get a power line ac filter that works over the widest freq range and has high isolation between outlets.

Thx-RNMarsh
 
Status
Not open for further replies.
Top Bottom