- Joined
- Dec 5, 2019
- Messages
- 3,843
- Likes
- 3,790
or we could just measure oxytocin levels
-
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!
"Things that cannot be measured"
- Thread starter ShiZo
- Start date
or we could just measure oxytocin levels
audiophile
Active Member
- Joined
- Oct 7, 2019
- Messages
- 177
- Likes
- 141
- Joined
- Oct 25, 2019
- Messages
- 11,158
- Likes
- 14,849
403 access denied!
EDIT- though I assume this was it https://darko.audio/2021/04/podcast-jonathan-novick-audio-precision-aes-cta-on-measuring-audio-gear/
EDIT EDIT- your link works now. We broke Darko
Raindog123
Major Contributor
With “Verboten!” 
Both links are broken. FYI403 access denied!
EDIT- though I assume this was it https://darko.audio/2021/04/podcast-jonathan-novick-audio-precision-aes-cta-on-measuring-audio-gear/
EDIT EDIT- your link works now. We broke Darko
- Joined
- Oct 25, 2019
- Messages
- 11,158
- Likes
- 14,849
- Joined
- Feb 23, 2016
- Messages
- 20,936
- Likes
- 38,034
Maybe we need a audio-gull rating for audiophile gullibility. Give zero to 10 gulls or should we have an infinite rating?
Cantor proved that not all infinities are equal. If we do chose an infinite rating, let's pick the more infinite of infinities, to make sure we don't run out of ratings too quickly.
audiofooled
Addicted to Fun and Learning
- Joined
- Apr 1, 2021
- Messages
- 537
- Likes
- 600
Sometimes CTRL+C/CTRL+V does wonders.
BluesDaddy
Senior Member
- Joined
- Mar 21, 2019
- Messages
- 342
- Likes
- 497
There's an old joke regarding the acronyms for college degrees: Everyone knows what "BS" stands for, well "MS" is just "more of the same", and PhD is simply "Piled Higher and Deeper". This is not meant to disparage anyone with degrees (I have a couple myself) but to poke fun at the inclination of academicians to confuse pompous verbiage with erudition.
kristiansen
Member
Lots of silence after the podcast and the video with Jonathan Novick which only tells us what we should all know.
That of course we can far from measure everything we hear, we can sometimes be lucky and measure the cause of what we hear, but that is only a fraction of everything that goes on, But we can measure "a lot of inconsequential things" far more accurately and with much lower levels than human can hear. Fine.
The company he mentions is not Danish, but Swedish
Audiograph.se
That of course we can far from measure everything we hear, we can sometimes be lucky and measure the cause of what we hear, but that is only a fraction of everything that goes on, But we can measure "a lot of inconsequential things" far more accurately and with much lower levels than human can hear. Fine.
The company he mentions is not Danish, but Swedish
Audiograph.se
audiofooled
Addicted to Fun and Learning
- Joined
- Apr 1, 2021
- Messages
- 537
- Likes
- 600
Watched it - agree with you - he already posted a number of times - lost interest for further discussion - prefer silence.
solderdude
Grand Contributor
I'll bite..
Everything he says is 100% correct. I think every engineer will agree with his complaint.
You realize what the story is about. It was NOT that everything cannot be measured (in the electrical domain).
It was a plea for manufacturers to show standard measurements and explain what and how it was measured.
What this was about was distortion measurements of power amps and what ends up in the manual in terms of specs.
The specs that manufacturers give you.
I hope you realize that the measurements Amir does often are done to verify specs and sometimes go beyond.
Yes, Amir certainly does not measure everything that should/has to be measured and reports things he finds worth reporting.
That does not mean there are things that cannot be measured. They can. The issue is how to interpret ALL of those measurements and for this you need knowledge of audibility levels and someone that can communicate that to 'the general non technical' public.
Nowhere in his talk did he say anything about your claims (resistors, capacitors or even silver vs copper) so this talk basically is an ode to independent measurements and checking of specs.
Amir shows several aspects that do not show up in specs sheets from manufacturers so add info for those who understand that added info.
It is clear that SINAD is a lousy 'indicator' for SQ as it measures just 1 aspect at one frequency. But this is not the only thing measured.
Amir is often asked or questioned by some regulars (think PMA, RestorerJohn and quite a few others as well) asking for more measurements (in the sense of different types of measurements/test circumstances, but doing all tests all the time means a LOT of reporting also of things that are irrelevant so not feasible if you want/need to churn out 1 or 2 evaluations per day because the audience wants that.
Moral of the story is: You can measure everything there is in the electrical domain. Not everything IS measured or shown. There are no mysteries in audio, only incomplete measurements and conclusions drawn (the latter for many reasons).
- end of rant -
solderdude
Grand Contributor
First one has has to determine they sound different in a scientific way. Then, when this has been proven one needs to measure the amp on a lot of different aspects and then, using knowledge of audibility thresholds one can show why certain (power) amps perform better than others under certain loads and conditions.
Not doing this and or not understanding all needed measurements as a whole is the root of all misunderstandings.
never rely on manufacturers specs only, aside some good ones with lots of plots, and don't think the ears are excellent analyzers either. They aren't.
Not doing this and or not understanding all needed measurements as a whole is the root of all misunderstandings.
never rely on manufacturers specs only, aside some good ones with lots of plots, and don't think the ears are excellent analyzers either. They aren't.
While watching, I saw this (at 38:67):
It's the first time I see the explanation for this weird non-convex behavior w.r.t to the thd+n versus power.
That was the kind of non convex behavior I saw on this amplifier for instance (Hypex NC502MP).
That really puzzled me, because the guy clearly shows that the crossover distortion, the one that happens mostly around the zero amplitude of the waveform is the most detrimental with music signals. He shows the probability density function w.r.t to voltage for a piece of music, and it's a clearly a peaked distribution around zero.
P.S: this could be totally obvious for a lot of people here, but to me it wasn't.
Last edited:
kristiansen
Member
Yes it is, nothing new under the sun
At least it seems that it is doable to match an audible characteristic to a measurable one, which is encouraging !
Comparing an amplifier for which the non-convex part of the distortion is at the same level of another's one but in the clipping part, will tell you that with music the worst offender is the one with the non-convex behavior (thd+n being equal at a given power output).
An audibility threshold may also be needed for this kind of crossover distortion.
Oh and you are saying that is nothing new, but the graph is coming from a class-d amplifier, from a quick google search, crossover distortion is mostly related to class B and/or AB amplifier. So is it something normal to see this in class-d amplifier ?
Last edited:
Similar threads
