Transparency and the sound of a system

[stehno]

What is interesting here, to stretch the visual analogy, is that something could be colored and STILL have a high degree of transparency. You would only lose transparency in the event that something you are looking at was MASKED by the color of the glass you are looking through. I have heard situations where the overall presentation was highly transparent and all images, space etc. were well resolved but there was a pervasive coloration to the tonality. If that tonality is caused by simple linear distortion (i.e. frequency response errors) then it may not have any impact on transparency whatsover (like the AudioStatic ES100s or STAX ELS F-81s, which were über transparent...but tonal balance was not neutral) but if it is from cabinet resonance, driver break up etc. then there is a situation where the distortion is masking the intended frequencies and this may impact the ability to hear "into" the music. Noise-floor of a speaker will also impact this "hear into" quality greatly.

For electronics, it has a lot to do with noise floor and signal CORREALTED noise floor. If the noise floor is being impacted by distortion components that cause the rise and fall with level, this will negatively impact transparency.

Interesting. Can you list even one type of system-related distortion that does not impact the noise floor?

And if as you say the noise floor negatively impacts transparency, isn't every other aspect of the presentation likewise negatively impacted?

 

[morricab]

Interesting. Can you list even one type of system-related distortion that does not impact the noise floor?

And if as you say the noise floor negatively impacts transparency, isn't every other aspect of the presentation likewise negatively impacted?

Frequency response and phase shift.

I don’t necessarily think true noise has to impact transparency but signal correlated noise and distortion for sure do. Tonal balance might not be affected. True noise (like tape hiss or white noise) should have little impact unless severe.

 

[Bodhi]

I think most people who focus on "transparency" really mean detail.

Yes, but i'd add to that a kind of 'see through' quality which comes primarly from accuracy and dynamic range. I still have vivid memories of the first time I heard the Infinity IRS-V's. Their ability to accurately reproduce complex orchestral scores with realistic scale and impact, and ethereal 'reach out and touch it' imaging was phenomenal.

 

[stehno]

Frequency response and phase shift.

I don’t necessarily think true noise has to impact transparency but signal correlated noise and distortion for sure do. Tonal balance might not be affected. True noise (like tape hiss or white noise) should have little impact unless severe.

So a playback system's limitations e.g. frequency response now equates to a distortion? I must have missed this white paper. Please explain.

Are you also saying "true noise" does not impact a noise floor? Interesting. I would have thought that all noise (audible and inaudible distortions) impact the noise floor whether we hear it or not. But your response begs the questions... Which noises impact the noise floor and which do not? How are you able to determine whether a noise floor has been impacted?

Are you also saying this "true noise" discriminates between which characteristics e.g. transparency are impacted? Can you explain how this discrimination works?

What's the threshold that determines whether this "true noise" is severe or less than severe?

 

[stehno]

Yes, but i'd add to that a kind of 'see through' quality which comes primarly from accuracy and dynamic range. I still have vivid memories of the first time I heard the Infinity IRS-V's. Their ability to accurately reproduce complex orchestral scores with realistic scale and impact, and ethereal 'reach out and touch it' imaging was phenomenal.

Yes, KeithR is correct as more transparency ultimately implies more detail. Then again, every system-related improved characteristic and every system-related improvement is the result of greater resolution or detail.

I appreciate your fond memories of the Infinity's but it's important to realize that no speaker is an island. The speaker is simply the last in a long line of signal processing of which every component and cable and whatever else lies upstream has IMO more to do with what we hear at the speaker than the speaker itself.

 

[morricab]

So a playback system's limitations e.g. frequency response now equates to a distortion? I must have missed this white paper. Please explain.

Are you also saying "true noise" does not impact a noise floor? Interesting. I would have thought that all noise (audible and inaudible distortions) impact the noise floor whether we hear it or not. But your response begs the questions... Which noises impact the noise floor and which do not? How are you able to determine whether a noise floor has been impacted?

Are you also saying this "true noise" discriminates between which characteristics e.g. transparency are impacted? Can you explain how this discrimination works?

What's the threshold that determines whether this "true noise" is severe or less than severe?

Yes, any deviation from what was recorded is distortion. FR and phase shift are what's known as linear distortion...this is well known...if you have some technical knowledge.

No, true noise doesn't have so much impact on how low level signals we can hear. For example tape hiss...even if it is pretty loud you can hear sounds well below this level as it is uncorrelated with the signal. Obviously, if it is loud enough it will impact the ability to hear and discern low level signals. However, noise that is correlated with the signal is far more damaging of the ability to hear low level information.

"Are you also saying this "true noise" discriminates between which characteristics e.g. transparency are impacted? Can you explain how this discrimination works?"

No, I am not say this.

 

[morricab]

Yes, KeithR is correct as more transparency ultimately implies more detail. Then again, every system-related improved characteristic and every system-related improvement is the result of greater resolution or detail.

I appreciate your fond memories of the Infinity's but it's important to realize that no speaker is an island. The speaker is simply the last in a long line of signal processing of which every component and cable and whatever else lies upstream has IMO more to do with what we hear at the speaker than the speaker itself.

" appreciate your fond memories of the Infinity's but it's important to realize that no speaker is an island. The speaker is simply the last in a long line of signal processing of which every component and cable and whatever else lies upstream has IMO more to do with what we hear at the speaker than the speaker itself"

I actually agree with this.

 

[Folsom]

Yes, any deviation from what was recorded is distortion. FR and phase shift are what's known as linear distortion...this is well known...if you have some technical knowledge.

FR, or amplitude is linear distortion. Phase distortion is 100% not linear distortion, it's harmonic distortion.

 

[morricab]

FR, or amplitude is linear distortion. Phase distortion is 100% not linear distortion, it's harmonic distortion.

Sorry, phase distortion is harmonic distortion? Harmonic of what fundamental? It doesn't really make sense...

 

[Folsom]

Sorry, phase distortion is harmonic distortion? Harmonic of what fundamental? It doesn't really make sense...

It doesn't really matter what order it is (could be anything), it just isn't linear. If you think about a graph chart and you see a signal, in order for here to be phase changes whatever part you're looking at moves right or left to become distorted. The act of changing it has affects beyond the exact place you're looking at. You simply cannot get a 1+1=2 out of phase changes.

 

[morricab]

It doesn't really matter what order it is (could be anything), it just isn't linear. If you think about a graph chart and you see a signal, in order for here to be phase changes whatever part you're looking at moves right or left to become distorted. The act of changing it has affects beyond the exact place you're looking at. You simply cannot get a 1+1=2 out of phase changes.

Well, based on electronics texts you are not correct:

"There are two types of linear distortions: amplitude distortion and phase (or delay) distortion. If the amplitude response of the channel is not constant (or almost within ±1 dB) in the message band, the result is amplitude distortion and if the phase response of the channel is not linear (i.e., various frequency components of the message signal suffer different amounts of delay), the result is then phase (delay) distortion. Delay distortion is a critical problem in data and video transmission, but the human ear is surprisingly insensitive to it. Wired telephone channels introduce linear distortion, which can result in intersymbol interference. However, equalizers can significantly mitigate the impact of linear distortion. For instance, adaptive equalization is employed in all high-speed voice-band data modems."

 

[Folsom]

That actually sounds a bit confused. Equalizers can't help phase. They are saying you need linear phase, so anything that is not cumulatively linear is suddenly linear distortion? That is not true. Perhaps they're calling phase shifts (delays) linear because they generally can affect the entire spectrum, or large portions of it that are below a self resonate frequency. It's a possible byproduct in long transmissions. But at the same time you could have some frequencies affected and others not (above the self resonate frequency). Either way it's not linear distortion when it's cumulative, and irrelevant when it accounts for the whole spectrum (phase shift). I can only imagine where you dredged that up from. Every electronic designer and speaker manufacturer on this planet knows HD (phase) isn't linear. You should just say, "oh my bad, you're right." But like always you're confrontational.

 

[morricab]

That actually sounds a bit confused. Equalizers can't help phase. They are saying you need linear phase, so anything that is not cumulatively linear is suddenly linear distortion? That is not true. Perhaps they're calling phase shifts (delays) linear because they generally can affect the entire spectrum, or large portions of it that are below a self resonate frequency. It's a possible byproduct in long transmissions. But at the same time you could have some frequencies affected and others not (above the self resonate frequency). Either way it's not linear distortion when it's cumulative, and irrelevant when it accounts for the whole spectrum (phase shift). I can only imagine where you dredged that up from. Every electronic designer and speaker manufacturer on this planet knows HD (phase) isn't linear. You should just say, "oh my bad, you're right." But like always you're confrontational.

They are saying that equalizers will help with linear (amplitude distortions). Electronic equalizers though often introduce phase distortion and/or harmonic (i.e. non-linear distortion). Sorry you don't like accepted definitions. When someone is correcting me correctly then I say what you would like to hear...until then...

" Linear distortion does not introduce new frequency components to a signal but does alter the balance of existing ones."

"
Linear distortions


Any frequency spectrum that passes through a linear system can undergo only linear distortions. In practice these distortions may emerge by a change in signal amplitude and/or a change in signal phase for each frequency component. Consequently the distortions of linear systems are divided into two categories: amplitude and phase distortions. "

"Distortion is any change in the content of an electrical signal or the shape of a sound wave during its transmission. It could also be described as information either lost or added relative to the original signal. Audio distortion can be placed in two broad categories.

• Type 1 — linear, altered amplitude content.
• Type 2 — nonlinear, added frequency content.

Type 1 is also known as frequency response, which involves both amplitude (changes in the volume) and phase errors. These distortions can be corrected with signal processing.

Type 2, in its most common form, is harmonic distortion (added frequencies). Type 2 can not be corrected after the fact. Each of these types can be subdivided into more specific forms."

"There are linear and non-linear forms of distortion. Linear distortions affect the amplitude and phase of audio signals, but don't show up on harmonic distortion analyzers as added frequency components that weren't there in the first place."

I could go on...it seems quite well defined in a number sources...so you could just say "oh my bad, you're right"...but you're just as confrontational...

 

[Folsom]

Phase error is a cumulative response from phase shifts, which appears like distortion to the reciever but isn't an actual distorted wave - it's just a bunch of out of time waves. It has nothing to do with audio. You're quoting aeronautic problems or something.

 

[morricab]

Phase error is a cumulative response from phase shifts, which appears like distortion to the reciever but isn't an actual distorted wave - it's just a bunch of out of time waves. It has nothing to do with audio. You're quoting aeronautic problems or something.

All quotes come from audio discussions...nothing to do with aeronautics...you can take it up with Nelson Pass...the last quote was from him.

 

[Folsom]

Again, phase error isn’t something we deal with in audio except maybe in badly designed digital filters - whether Nelson Pass said it or not. And it will only confuse people to call it phase distortion because you have to preclude that it is about multiple arrivals. The signals are not distorted themselves - this is why the proper term is phase error. Pass is being a bit liberal with the word distortion when it is entirely context driven, and the subject is typically audio with him.

The closest thing would be reflections, but they are talked about differently because the solutions are not electrical.

 

[morricab]

Again, phase error isn’t something we deal with in audio except maybe in badly designed digital filters - whether Nelson Pass said it or not. And it will only confuse people to call it phase distortion because you have to preclude that it is about multiple arrivals. The signals are not distorted themselves - this is why the proper term is phase error. Pass is being a bit liberal with the word distortion when it is entirely context driven, and the subject is typically audio with him.

The closest thing would be reflections, but they are talked about differently because the solutions are not electrical.

Speakers are full of phase distortion (or error if you are unhappy with the semantics)!

 

[Folsom]

Speakers are full of varied phase because that is the nature of a driver moving. It is not phase error. But a ported speaker uses phase error to increase SPL.

 

[morricab]

Speakers are full of varied phase because that is the nature of a driver moving. It is not phase error. But a ported speaker uses phase error to increase SPL.

Yes, and compared to the signal its fed it is distortion.

 

[Folsom]

Yes, and compared to the signal its fed it is distortion.

But it isn't phase error (unless we're talking about port).