Better than the sum of their parts?

[LTig]

Trying to match different solid state equipment to improve SQ is futile. This is different with tube electronics due its higher output impedance which influences frequency response depending on what the output feeds. And we know that even slight changes in frequency response are audible.

Connect a longer cable to a tube preamp output and its capacity creates a low pass which may be perceived as "warm" sound.

Connect a speaker to a tube power amp and its impedance characteristics creates a frequency dependent voltage divider, often resulting in a kind of loudness curve (more bass, more trebles) for typical 2-way speakers which may be perceived as "fuller" sound. Actually the same is true for some esoteric solid state power amps sporting high output impedance due to low or no negative feedback.

Having both of those situation above the loss of treble between pre- and power amp can be compensated somewhat by the increase of treble due to the impedance of the tweeter rising with frequency, which may be perceived as more neutral sound.

Of course this is a rabbit hole and the best way to circumvent it is to use proper electronics with sufficiently low output impedance to get a neutral sound and then use tone controls to adjust frequency response to taste and/or to fix bad recordings.

 

[egellings]

Unless one of the devices has the exact same distortion profile as the the other, but 180 degrees out of phase, while keeping the actual music signal in phase, then no.

Not something that would ever happen in normal audio gear by chance. It would have to be very specifically tailored to get that effect.

Noise and distortion will only increase every time you add a link to the chain. The idea of "synergy" is either make-believe or nothing but simple frequency response changes aka. EQ done in a very cumbersome way.

Distortion is usually measured as a magnitude of the unwanted signal content. Phase angle is not included in that, so basically, distortion is an absolute value; that is, a positive number. I do not see how you would sum two distortion components to get a smaller number. The result would always be larger. You could have a frequency response error (a sort of distortion that does not result from unwanted components added to the measured signal), and correcting that would reduce the error. That's what equalizers do.

 

[LTig]

Distortion is usually measured as a magnitude of the unwanted signal content. Phase angle is not included in that, so basically, distortion is an absolute value; that is, a positive number. I do not see how you would sum two distortion components to get a smaller number. The result would always be larger.

Just because the phase angle of a distortion component is not measured/published doesn't imply it does not exist. So it is possible to add 2 distortion components and get a smaller residual distortion. It's pure chance though and IMV rather unlikely to occur.

 

[OldHvyMec]

Of course this is a rabbit hole and the best way to circumvent it is to use proper electronics with sufficiently low output impedance to get a neutral sound and then use tone controls to adjust frequency response to taste and/or to fix bad recordings.

Thus the used of SS for the bass response and valve amps for mids and highs. I see a pure harmony and once again the WHOLE is greater than its parts.
I'm not a SS fan for the most part when it comes to final amplification or pre amps either, but SS sure has its place too. The distortion measured is audible the same way 10% distortion in the bass region is UN AUDIBLE. People say they can hear the distortion in a valve amp. That is BS according to their own testimony, they can't hear 10%+ and the overall mess it leaves in its wake from 280-300hz and below. Once it's all cleaned up by EQing the room mechanically, and proper placement first, the use for equipment EQ (at all), is usually because of the crappy source recording.

It kind of reminds me of live performances, most suck. I prefer a great mix most of the time. There were exceptions like Carlos, and Prince. I never HEARD a bad show from either.

Regards

 

[Piere]

Harmonic distortion relates to harmonic signals. Good for mathematics, measurements and engineering. However music is more like a noise signal. That is: A temporarily signal.

 

[thor79]

No I'm not buying this. The stereo is a big series circuit. You can have feedback between parts, but parallel it is not nor do I think conceptually thinking of it as a parallel system is at all helpful or useful except as a marketing tool. The idea you add things on in parallel to improve results is great for marketing and not at all reality. If your source is bad, everything after might faithfully reproduce the badness. If your speakers are trash, everything fed into them will be compromised. Fortunately it does not cost much money for excellent performance up to the speaker inputs.

This thread could also have been called audio system synergy.

I certainly wouldn't force you to believe that an electrical circuit, in terms of signal, should be considered more as a parallel circuit (connection) than as a series connection.

You are right the description of the parallel circuit would be more logical and understandable explanation of what hi-fi enthusiasts experience of sound differences from power cables to chassis and everything between.

 

[Sokel]

Distortion is usually measured as a magnitude of the unwanted signal content. Phase angle is not included in that, so basically, distortion is an absolute value; that is, a positive number. I do not see how you would sum two distortion components to get a smaller number. The result would always be larger. You could have a frequency response error (a sort of distortion that does not result from unwanted components added to the measured signal), and correcting that would reduce the error. That's what equalizers do.

The chart I posted earlier shows exactly that.One can play and see what happens (Multitone analyzer has the option to measure a sine including specifying the phase angle except the feature of the transfer function showed).
There are some occasions where some combination reduce distortion but generally that's not the case.
It's fun to try nevertheless!

 

[Killingbeans]

You are right the description of the parallel circuit would be more logical and understandable explanation of what hi-fi enthusiasts experience of sound differences from power cables to chassis and everything between.

Well, the brain does parallel computing, so in that sense placebo could be seen as a "parallel circuit"

Besides that, you are seriously grasping at straws.

 

[Sokel]

The chart I posted earlier shows exactly that.One can play and see what happens (Multitone analyzer has the option to measure a sine including specifying the phase angle except the feature of the transfer function showed).
There are some occasions where some combination reduce distortion but generally that's not the case.
It's fun to try nevertheless!

As a simple test,see what happens when only reversing the phase of the worst harmonic,everything else identical:

Normal

reverse

Third harmonic is -3db down.

(don't ask me to interpate it,WAY beyond my pay grade).

 

[Piere]

With an amp driving a speaker cable and finally a speaker the laws of Kirchoff and Thevenin are still valid (despite all the Cryo and Quantum alchemical efforts). So whether it is a series circuit or a parallel circuit doesn;t matter much. As already noted, the capability of an amp to unconditionally drive the loudspeakers is usually the weakest link.

 

[Killingbeans]

I think one of the things that also confuse the non-technical minded, is that although noise and distortion doesn't "stack like lego bricks" it doesn't "mix like paint" either.

Out of two components, the biggest contributor of distortion will always dominate. There's no magical scource component that will "fix" a boutique brand speaker designed to be an unapologetic distortion factory.

 

[Joseph Crowe]

Trying to match different solid state equipment to improve SQ is futile. This is different with tube electronics due its higher output impedance which influences frequency response depending on what the output feeds. And we know that even slight changes in frequency response are audible.

Connect a longer cable to a tube preamp output and its capacity creates a low pass which may be perceived as "warm" sound.

Connect a speaker to a tube power amp and its impedance characteristics creates a frequency dependent voltage divider, often resulting in a kind of loudness curve (more bass, more trebles) for typical 2-way speakers which may be perceived as "fuller" sound. Actually the same is true for some esoteric solid state power amps sporting high output impedance due to low or no negative feedback.

Having both of those situation above the loss of treble between pre- and power amp can be compensated somewhat by the increase of treble due to the impedance of the tweeter rising with frequency, which may be perceived as more neutral sound.

Of course this is a rabbit hole and the best way to circumvent it is to use proper electronics with sufficiently low output impedance to get a neutral sound and then use tone controls to adjust frequency response to taste and/or to fix bad recordings.

I never said anything about linear types of distortion (frequency response errors etc). I'm specifically referring to nonlinear behavior such as IMD. Of course linear distortion can be corrected.

 

[Joseph Crowe]

As a simple test,see what happens when only reversing the phase of the worst harmonic,everything else identical:


View attachment 263551

Normal

View attachment 263552

reverse

Third harmonic is -3db down.

(don't ask me to interpate it,WAY beyond my pay grade).

Yes, it seems conceivable that a reverse nonlinear function can cancel out harmonic distortion generated by the offending upstream device assuming reverse phase. It's also interesting that a harmonic distortion test is phase blind. It could be completely out of phase with the fundamental tone and the test metric value would remain the same. What is more offensive to the ear? In phase or reverse phase for K2 and K3? Does anybody know?

 

[RayDunzl]

Distortion is usually measured as a magnitude of the unwanted signal content. Phase angle is not included in that, so basically, distortion is an absolute value; that is, a positive number. I do not see how you would sum two distortion components to get a smaller number. The result would always be larger.

I think I managed to do it for a simple case..

Summed speaker and signal distortion to eliminate a harmonic:

Distortion Cancellation Experiment

Using REW, send a 450Hz tone at -20dB to the system (JBL LSR 308 speakers), take an RTA with UMIK-1. Note high level of 3rd harmonic distortion in the speaker output. It is clearly audible. Using the Distortion controls, add third harmonic to the signal at -40 degree phase angle.

www.audiosciencereview.com

 

[Blumlein 88]

Yes, it seems conceivable that a reverse nonlinear function can cancel out harmonic distortion generated by the offending upstream device assuming reverse phase. It's also interesting that a harmonic distortion test is phase blind. It could be completely out of phase with the fundamental tone and the test metric value would remain the same. What is more offensive to the ear? In phase or reverse phase for K2 and K3? Does anybody know?

REW gives you phase of the harmonics.

 

[Joseph Crowe]

Distortion is usually measured as a magnitude of the unwanted signal content. Phase angle is not included in that, so basically, distortion is an absolute value; that is, a positive number. I do not see how you would sum two distortion components to get a smaller number. The result would always be larger. You could have a frequency response error (a sort of distortion that does not result from unwanted components added to the measured signal), and correcting that would reduce the error. That's what equalizers do.

It's rational to assume that distortion only adds, however if you study the topic further to understand how nonlinear functions work, it is conceivable that they could work in concert to reduce distortion. An example of this is found in this video, where he shows the nonlinear function on his oscilloscope.

If a downstream device had a tail swerving the right instead of the left as shown below, it would sum to a net zero distortion.

 

[HarmonicTHD]

I certainly wouldn't force you to believe that an electrical circuit, in terms of signal, should be considered more as a parallel circuit (connection) than as a series connection.

You are right the description of the parallel circuit would be more logical and understandable explanation of what hi-fi enthusiasts experience of sound differences from power cables to chassis and everything between.

Whatever it is …. So far we haven’t seen a single piece of evidence from you. Just word salad. So pardon us for not buying your story.

 

[Piere]

It's rational to assume that distortion only adds, however if you study the topic further to understand how nonlinear functions work, it is conceivable that they could work in concert to reduce distortion. An example of this is found in this video, where he shows the nonlinear function on his oscilloscope.
If a downstream device had a tail swerving the right instead of the left as shown below, it would sum to a net zero distortion.

Yes, this looks a special laboratory case without circuit detail. Looks like an additive approach. A signal distorted by one non-linear function is not "un-distorted" by feeding that signal to an apparatus that provides a reverse non-linear transfer function. That is a multiplicative mechanism anyway and will prohibit it anyway. And that is the question this topic started with and audiophiles do by trying to match equipment.

 

[thor79]

Whatever it is …. So far we haven’t seen a single piece of evidence from you. Just word salad. So pardon us for not buying your story.

It seems that some of the participants in this forum are not aware of the reason for this forum's existence and especially the reason for the forums which are in strong opposition to ASR.

The reason can be roughly described as many hi-fi enthusiasts hearing among other things sound differences on cables including power cables.
Many take these sound differences for granted, they are not!

They hear something that can neither be measured nor theoretically proven.

That is what ASR stands for, they tell the "subjective hi-fi enthusiasts" that what they hear is not possible, some describe it in a more direct form.

I hear sound differences myself, despite my insight into electronics telling me I shouldn't.

There is no doubt that ASR holds the good cards in this case, and I have a conundrum, a problem.
I try here to describe the conflict, and the problems of measurement and its limitations, and hope thereby to inspire myself or others to alternative methods of measurement, so that the two groups of hi-fi enthusiasts can be united.

 

[BDWoody]

The reason can be roughly described as many hi-fi enthusiasts hearing among other things sound differences on cables including power cables.

Yes, many hear lots of things.

They hear something that can neither be measured nor theoretically proven.

At least they believe they do. What they don't typically do at that point is set up a properly controlled test to see if they are kidding themselves or not. All 'we' need is a single example of someone backing whatever flavor of these claims they want to make with actual evidence, rather than something along the lines of "we can't all be wrong, can we?"

I hear sound differences myself, despite my insight into electronics telling me I shouldn't.

You believe you do, but you've also admitted that your listening tests don't meet basic thresholds for meaningful controls. Do you believe they don't really matter?