Exactly so, yes. The canonical "wife in the kitchen heard the difference" is a brilliant example of Clever Hans in action.
-
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!
Ethan Winer Builds a Wire Null Tester
- Thread starter amirm
- Start date
- Joined
- May 25, 2019
- Messages
- 158
- Likes
- 218
FYI, if it's not "even coin toss level" that means that you can identify cables. Coin toss level implies the least amount of differentiation between cables
solderdude
Grand Contributor
Last edited:
nice article
Thanks guys.
Tks
Major Contributor
- Joined
- Apr 1, 2019
- Messages
- 3,221
- Likes
- 5,496
Just wanted to say I was a massive fan of your Youtube videos on audio science. Helped me understand so many things so clearly when getting into audio.
Ethan's videos on audio science are great, but the one I still like best is the cello rondo... 
Completely lost the context for this in the overall thread, but my thought was the "wife heard" comment was code for: "would you PLEASE stop your infernal mucking around already...!"
As for the Wire Null Tester, I'm not sure I'm in the market for one, would be interested to know if it will ever be released as a kit or assembled product, @Ethan Winer ?
PS, love your contribution to the science of music enjoyment!
Hypnotoad
Active Member
It's like the guys dog that played chess with him, people said it was amazing how smart he was, the guy said he's not that smart I beat him every game.
No, no, no! "Wife heard the difference from another room" is the replacement method for double-blind testing practiced by many audiophiles. It is much more sensitive than DBTs and ABX testing, apparently.
Thanks very much guys.
Gary, I considered selling this but it would require a lot of effort to correct a few minor problems, make new boards, etc. If it could test more than wires it would be a viable product. I'm glad to share the complete schematic with you (or anyone else here) if you email me from my home page:
http://ethanwiner.com/
Gary, I considered selling this but it would require a lot of effort to correct a few minor problems, make new boards, etc. If it could test more than wires it would be a viable product. I'm glad to share the complete schematic with you (or anyone else here) if you email me from my home page:
http://ethanwiner.com/
solderdude
Grand Contributor
One of the issues could also be adjusting the trimmers which would need either a specific high frequency squarewave and monitoring circuit or a good scope and signal generator to get that right.
I realize that would mean using input attenuators when nulling power amps (I have that in my nulling device)
EDIT: nevermind the circuit shown here is not the complete schematic as the meter, headphone out and power supply also isn't shown and it is also mentioned in the article.
Last edited:
solderdude
Grand Contributor
The setup in this case is super simple.
You can do it the complicated way (and band limited way) and use Paul's excellent nulling software.
Easy to compare 2 cables of equal length this way.
You can even use audio software when you record 2 channels on the same ADC at the same moment (no time drift)
You can do it the complicated way (and band limited way) and use Paul's excellent nulling software.
Easy to compare 2 cables of equal length this way.
You can even use audio software when you record 2 channels on the same ADC at the same moment (no time drift)
solderdude
Grand Contributor
I know... that's why I shared the link because it is already done.
What kind of complexity do you think is involved when using a DAC (as the source) and an ADC ?
There is at least an equal amount of components in the path in case of an ADC just tucked in a different box.
Ethan's device (so is mine) is also a single box and a LOT less complex than an ADC-DAC loop.
Well... with the null tested you, of coarse, also use a DAC as a signal source in most cases.
What kind of complexity do you think is involved when using a DAC (as the source) and an ADC ?
There is at least an equal amount of components in the path in case of an ADC just tucked in a different box.
Ethan's device (so is mine) is also a single box and a LOT less complex than an ADC-DAC loop.
Well... with the null tested you, of coarse, also use a DAC as a signal source in most cases.
solderdude
Grand Contributor
that means the null is not perfect. could be because of component tolerances between 2 channels.
In the null circuit this can be optimized.
In the end the null circuit has less components and does not rely on 2 potentially differing ADC channels and electronics.
You can see in Ethans video you can null cables so only noise remains.
My experience is the same. For nulling longer cables you need a delay line (cable of the same length) though.
When audio remains audible the null is not perfect.
I had to use a coarse and fine pot to get perfect nulls.
Without a shadow of a doubt the device Ethan has can obtain better nulls when one has a real-time device to test. DAC's are not real-time devices so that should be done with something like deltawave.
solderdude
Grand Contributor
I only use a null tester like the one used by Ethan W.
Not really interested in complex devices and manipulation.
I just want to know what a cable does not how capable AD-DA conversion is.
Though it is interesting to see how well the method performs.
Not really interested in complex devices and manipulation.
I just want to know what a cable does not how capable AD-DA conversion is.
Though it is interesting to see how well the method performs.
solderdude
Grand Contributor
A cable does not behave like a a weak EQ or filter thing. At least not within and well outside the audible range. People often think this because they have convinced themselves they hear a tonal change (talking about interlinks not speaker cables).
It does add minute (inaudible) noise of course.
There is nothing in nulls I made that indicates it acts as a filter.
Only when you have very high source impedances the cable properties can create high frequency roll-off and phase differences.
With speaker cables things differ, in that case the resistance plays a role when we talk about 'weak EQ' effects (due to voltage division).
All changes you see in the lows are most certainly coming from the used electronics, and those differences near nyquist are due to digital filter properties.
This does not happen with nulling. Only the delay due to cable length needs to be taken into account.
Much easier, less components and no need to 'process' things.
It does add minute (inaudible) noise of course.
There is nothing in nulls I made that indicates it acts as a filter.
Only when you have very high source impedances the cable properties can create high frequency roll-off and phase differences.
With speaker cables things differ, in that case the resistance plays a role when we talk about 'weak EQ' effects (due to voltage division).
All changes you see in the lows are most certainly coming from the used electronics, and those differences near nyquist are due to digital filter properties.
This does not happen with nulling. Only the delay due to cable length needs to be taken into account.
Much easier, less components and no need to 'process' things.
Last edited:
solderdude
Grand Contributor
why would one worry about a null as deep as -100dB ?
This means there is no worthwhile change.
yes, when the differences are small you have to be carefull about noise.
The ADC adds noise and the nulling circuit adds noise.
Any significant noise added will be by the electronics, not the cable its contribution is way smaller.
The 8 bit resolution scope has little to do with it as the difference signal is amplified before it goes to the scope and even in the 10mV setting the scope still has 8 bit resolution.
What's not to be trusted about the nulling device and what cannot be trusted from a bandwidth limited system with a lower and higher (extremely sharp filtered) system. Even a small ripple at the higher frequencies can cause differences.
This means there is no worthwhile change.
yes, when the differences are small you have to be carefull about noise.
The ADC adds noise and the nulling circuit adds noise.
Any significant noise added will be by the electronics, not the cable its contribution is way smaller.
The 8 bit resolution scope has little to do with it as the difference signal is amplified before it goes to the scope and even in the 10mV setting the scope still has 8 bit resolution.
What's not to be trusted about the nulling device and what cannot be trusted from a bandwidth limited system with a lower and higher (extremely sharp filtered) system. Even a small ripple at the higher frequencies can cause differences.
Last edited:
solderdude
Grand Contributor
I have no idea what point you are trying to make.
With a 2% error your null will never be better than -34dB
When you want to achieve -100dB the error must be smaller than 0.001%
With a 2% error your null will never be better than -34dB
When you want to achieve -100dB the error must be smaller than 0.001%
Similar threads
- Locked
