Ethan Winer may be on the verge of proving expensive interconnects don't matter.

[RCHeliGuy]

While it is an interesting exercise, incorporating a speaker into any set of measurements is folly for it introduces too many uncontrolled variables which can skew the results and render the test inaccurate and invalid. A much better way to measure, say interconnects, is via signal analysis instrumentation equipment ..... and done up until the time the signal is inputted to the speaker. Analysis of the distortion spectra can then be easily referenced to known hearing thresholds and conclusions drawn thereafter.

I agree.

It sounds like you are describing the box that Ethan is working on.

 

[Pneumonic]

Mark, not sure what Ethan's box comprises but I am talking about doing signal analysis work using scopes and analyzers.

 

[RCHeliGuy]

Mark, not sure what Ethan's box comprises but I am talking about doing signal analysis work using scopes and analyzers.

His null box will measure an analog signal going in and an analog signal going out and then create a signal that is only the difference between the two.

By comparing the delta between electronics or cables, you will be able to measure any effect that they had on the audio signal.

I'm not sure if it can scale to handle the gain difference after the amplifier, but that seems doable. I'll ask him.

Assuming the measurements were precise enough and you recorded both signals, you should be able to generate an amplitude multiplier for the lower signal that was purely a mathematical scaling feature so as not to cloud the results and assuming he is running 64 bit math or better so there is no loss after calculations.

 

[User211]

While it is an interesting exercise, incorporating a speaker into any set of measurements is folly for it introduces too many uncontrolled variables which can skew the results and render the test inaccurate and invalid. A much better way to measure, say interconnects, is via signal analysis instrumentation equipment ..... and done up until the time the signal is inputted to the speaker. Analysis of the distortion spectra can then be easily referenced to known hearing thresholds and conclusions drawn thereafter.

The bolded bit is important.

However, what I really like about the mic test is EVERYTHING is in the chain. I also really like the fact that the in room sound is being measured by the mic too, in other words together with room effects etc. That corresponds much more closely with what the ear will hear, despite the obvious flaws.

Though TBH it is hard to be sure about anything, and who knows how good Ethan's device will actually be. Given that he is biased anyway, anything is possible....

 

[Gordon Gray]

His null box will measure an analog signal going in and an analog signal going out and then create a signal that is only the difference between the two.

Mark,

Interesting do da.

How does this relate to the differences one may hear?

Gordon

 

[Rich]

His null box will measure an analog signal going in and an analog signal going out and then create a signal that is only the difference between the two.

By comparing the delta between electronics or cables, you will be able to measure any effect that they had on the audio signal.

I'm not sure if it can scale to handle the gain difference after the amplifier, but that seems doable. I'll ask him.

Assuming the measurements were precise enough and you recorded both signals, you should be able to generate an amplitude multiplier for the lower signal that was purely a mathematical scaling feature so as not to cloud the results and assuming he is running 64 bit math or better so there is no loss after calculations.

The problem with this route is that it doesn't take into account how the electrical properties of the cable interact with the load presented by the speakers. Woofers, for instance, will present a very different type of load than, say, electrostatic panels. Electrostats present as a more capacitative load. Two different cables could very well send the exact same signal when measured by an oscilloscope, but react differently under load. You are trying to measure the cable independently while ignoring the entire rest of the system it is hooked up to, and then trying to conclude there aren't sufficient differences that should be audible once the cable is connected to that system. Do you see how ridiculous this is? At least Justin is measuring what our ears are actually hearing: the output of the entire system, with a cable change being the only variable.

 

[RCHeliGuy]

Mark,

Interesting do da.

How does this relate to the differences one may hear?

Gordon

I can honestly say that it will never be able to predict how your brain interprets the signals from your ears

But then again that is the whole point. It should be able to measure the signals that eventually hit your ears, but can't predict anything past that point.

 

[RCHeliGuy]

The problem with this route is that it doesn't take into account how the electrical properties of the cable interact with the load presented by the speakers. Woofers, for instance, will present a very different type of load than, say, electrostatic panels. Electrostats present as a more capacitative load. Two different cables could very well send the exact same signal when measured by an oscilloscope, but react differently under load. You are trying to measure the cable independently while ignoring the entire rest of the system it is hooked up to, and then trying to conclude there aren't sufficient differences that should be audible once the cable is connected to that system. Do you see how ridiculous this is? At least Justin is measuring what our ears are actually hearing: the output of the entire system, with a cable change being the only variable.

You assume that the sensors replace parts of a system and are not just probes in the 100 Mega Ohm range having no impact on the signal where ever the probes are placed.

Also I don't know if this box is designed to handle the output of an amp. It may only be something to measure between components before the amp.

 

[RCHeliGuy]

Looks like you can measure both sides of an amp!

Me: Will your null box be able to compare a signal with a gain difference like before and after a pre-amp or on the other side of an amplifier?

Ethan Winer: Oh yes, of course! Each input has a 70 dB gain range from -50 to +20. The signal generator also has a balance pot with a 20 dB range. At max attenuation the inputs can handle a power amp up to about 10 KW.

Me: Wow! Very cool!

Ethan Winer: It's getting close, I'm testing the first section this weekend, then I'll continue stuffing the PC board. BTW, I've been enjoying your woodworking posts.​

 

[RCHeliGuy]

The bottom line is that this device "should" be a much more fool proof way to measure what if any difference a piece of equipment or cable makes.

If it does work properly we should be able to prove empirically what difference anything really makes to a system.

It should be able to measure distortion, phase, other non-linearities.

 

[amey01]

Mark,

Interesting do da.

How does this relate to the differences one may hear?

Gordon

You could ask the same question to Justin (but you seemed to respect his measurements when they supported your view that there IS a difference).

We all have components in our systems which measure differently. And we all accept that there are tangible differences to the sound we can achieve when we measure a difference. So why then, do we not accept that when there is not a measurable difference, there can still be tangible impacts to the sound?

Honestly though - I understand your question Gordon - and I do wonder this myself.

However, I always come back to the fact that: this audio signal is the only thing which the components are sending to your ears. That is - the ONLY thing we are putting into the sound system is the audio signal. Therefore, it is reasonable to measure that same signal as it comes out.

[I say this for argument's sake - not in an absolute sense] If that audio signal hasn't changed, then there can't be any difference to what you are hearing (quite simply, because nothing else was added to begin with).

It's like looking at a plain piece of white paper. If you see a tree on it, then it is your imagination - it is by definition not there. If you see a car on it, then it is purely your imagination. It is the same with audio - if you hear a difference when there is by definition not a difference present, then I'm afriad you are imagining that difference.

I'm sure even you understand that imagination is quite possible and plausible. That's why a discussion like this is worthwhile and valuable.

 

[Gordon Gray]

Adam,

Thanks for the input. Frankly, I don't look at graphs with all kinds of wiggly lines. Means very little to me but I am sure they are relevant / important to others.

The O v S discussion can have value but it has occurred so many times on so many websites and the end result rarely achieves any meaningful consensus that both camps can agree on.

I am still not sure what this new EW thing does to resolve the O v S disparity.

Best,

Gordon

 

[RCHeliGuy]

What Ethan's box should do is let us test something like an amplifier to see what characteristics we like.

For example Let's say someone wanted to compare amplifiers.

If you found a certain characteristic that you liked, you might be able to find that there are a family of amps that cluster around that sound characteristic.

I am talking about coloration, not perfection, but something that you tend to prefer.

 

[User211]

EWs box will do f'all, just like my squiggly lines.

I sometimes wonder why we bother discussing it TBH. Perhaps it's just because we're all perplexed by it.

It takes a certain mindset to even be interested in it.

 

[amey01]

rarely achieves any meaningful consensus that both camps can agree on.

Does it matter Gordon? Does there have to be consensus?

These discussions have helped me form my opinion; just as they no doubt have helped you form yours. Who said we have to agree?

 

[Gordon Gray]

Adam,

Would be nice but likely impossible.

Don't think anyone in either camp has had their opinion changed.

Thanks Mark for the clarification.

Gordon

 

[Bernard]

EWs box will do f'all, just like my squiggly lines.

I sometimes wonder why we bother discussing it TBH. Perhaps it's just because we're all perplexed by it.

It takes a certain mindset to even be interested in it.

Agreed. I can't help but wonder about the motivation for his cable-debunking crusade; he's been at it an awfully long time. Is it genuinely of academic interest, or is there some financial reason?

If he determines that he can in fact measure a difference between cables, I wonder if he will release the results as such or skulk away.

 

[RCHeliGuy]

Agreed. I can't help but wonder about the motivation for his cable-debunking crusade; he's been at it an awfully long time. Is it genuinely of academic interest, or is there some financial reason?

If he determines that he can in fact measure a difference between cables, I wonder if he will release the results as such or skulk away.

Ethan is just a retired engineer who is into audio, musical instruments and anything that makes sound.

I think he has just debated this topic so many times that he finally thought he could put this subject to rest.

 

[Bernard]

Ethan is just a retired engineer who is into audio, musical instruments and anything that makes sound.

I think he has just debated this topic so many times that he finally thought he could put this subject to rest.

I thought it was curious when he was here that he would never list the components in his system. Most people who are into audio do not hesitate to list their reference components. I had the distinct impression at one point that they were low-fi.

 

[Gordon Gray]

He did reveal awhile ago of the WBF Forum before he was banned.

Suffice to say it was very mid-fi (trying to be polite) based on a satellite / home theater type system with a subwoofer.

Think Best Buy.