Is a Lower Noise Floor, are "Blacker" Blacks, Consonant With the Music?

[Atmasphere]

The only time I'm aware of such a black background and potential exhileration is when it's the very first note at the very start of the music piece.

Yes.

IME when you have a blacker background (less intermodulations in the noise floor) instruments are simply more real.

 

[Blackmorec]

I see what you mean. However, there seem to be some-to-many who tout a "black background" as a badge of honor who don't stop there but go on to exclaim how exhilerating it is to hear an instrument say a trumpet rise up out of this black background.

The only time I'm aware of such a black background and potential exhileration is when it's the very first note at the very start of the music piece.

So to the best of my knowledge, the term black background remains a very misleading term at least for some.

But thanks for clarifying.

Walk outside on a city street at night and look up at the stars. See any? A few maybe. Now do the same thing in the middle of the wilderness and you’ll immediately see any visible planets and the utter grandeur of the Milky Way. The stars haven’t changed...only the background light level is different. The same thing happens to audio detail when you get rid of background noise...there‘s just a lot more information becomes audible

 

[stehno]

Walk outside on a city street at night and look up at the stars. See any? A few maybe. Now do the same thing in the middle of the wilderness and you’ll immediately see any visible planets and the utter grandeur of the Milky Way. The stars haven’t changed...only the background light level is different. The same thing happens to audio detail when you get rid of background noise...there‘s just a lot more information becomes audible

Nice analogy. Except for one thing. It's not the background light level that has changed in the wilderness. It's the foreground light level between your eyes and the stars that allows them to become visible. Which in this case makes for a very good depiction of what happens to the music (stars) when a noise floor drastically lowered.

 

[GertSterlingSilver]

Get some Balanced amatures IEMS @ 8-16 ohms and you'll hear all the noise you want to.
Is it relevant at the listening experience? Tough call.
Imho not a single recently produced device can get the linear response and low output impedance than "ye olde" sansa clip+ can do.

 

[tima]

The same thing happens to audio detail when you get rid of background noise...there‘s just a lot more information becomes audible

Do inaudible frequencies affect how we perceive what we do? The tricky part is discriminating between what is background noise and what is not.

 

[Blackmorec]

Do inaudible frequencies affect how we perceive what we do? The tricky part is discriminating between what is background noise and what is not.

Absolutely they do. But things have changed in the last 50 years with digital technology. With analog, there was only one way to sort noise from signal and that was to avoid adding the noise in the first place. Once noise is added it is impossible to differentiate it from the signal. Pretty much the only noise reducing strategy we had was Dolby, which essentially reduced noise we KNEW we were going to add, by boosting the signal at the frequencies of the noise prior to its addition, then reducing their amplitude post noise addition, thereby reducing those frequencies back to their normal amplitude and with them reducing the amplitude of the added noise. But with digital, things are different . A digital signal has 2 elements....the digital pattern AND its analog voltage representation, used to move the stream along wires and through components And the two can be separated and the digital pattern used to resynthesize its analog representation. If you look at a network, most people see the means to transport a digital stream, but for the audiophile it can be much, much more. If you take a modem data stream and feed it straight into a server and DAC you are essentially also feeding whatever noise it arrives with and picks up from the modem. But we can improve that signal. Replace the modem’s SMPS with a low noise LPS and add a designed-for-hi-fi switch + LPS and SQ improves. Why? Because you’ve thrown out the original highly contaminated analog representation and resynthesized it twice (Modem and Switch) using much quieter electronics. So a network is not only a transport medium for a digital stream, its also the means to recondition and improve the stream. Now theoretically, bits should be bits and as long as the digital pattern reaches the DAC intact, we should be able to reproduce the music with its sonic attributes intact. Theoretically. Unfortunately for audiophiles a few experiments along the lines of the above will quickly show that removing noise and jitter from the digital stream indeed has a very positive effect on sound quality, leading to the counter-theoretical conclusion that noise content and timing quality of a bit-perfect bit-stream are indeed critical elements when judged in terms of the sound quality the bit-stream produces.

 

[tima]

Absolutely they do. But things have changed in the last 50 years with digital technology. With analog, there was only one way to sort noise from signal and that was to avoid adding the noise in the first place.

I was thinking of power conditioning for electricity and the self-noise generated by components and put onto the signal and other components. In that realm there is impact on both analog and digital. External and internal power conditioning and filtering and noise reduction cabling and power cords as somewhat generic based on where (at what frequency) we think the noise can exist and if noise and signal can co-exist at certain near frequencies. Is that filtering sufficiently sophisticated to differentiate them? That's probably a rhetorical question.

 

[Atmasphere]

I was thinking of power conditioning for electricity and the self-noise generated by components and put onto the signal and other components. In that realm there is impact on both analog and digital. External and internal power conditioning and filtering and noise reduction cabling and power cords as somewhat generic based on where (at what frequency) we think the noise can exist and if noise and signal can co-exist at certain near frequencies. Is that filtering sufficiently sophisticated to differentiate them? That's probably a rhetorical question.

Good power conditioning can help out! Poor power conditioning IME often makes things worse- particularly for power amps.

A good conditioner will be able to filter out the 5th harmonic, which is arguably the worst offender when it comes to 'dirty power'. The 5th can cause diodes to become more noisy, power transformers to become more noisy (and less efficient) and these things have a direct effect on the performance of the power supply involved- which in turn can affect the audio equipment using it.

When the power supply has more noise, this noise can get intermodulated with the audio and becomes part of the noise floor.

Unfortunately I can't point to very many good power conditioners in high end audio; IMO very few exist. PSAudio probably makes the best. A company called Elgar which was/is not involved with audio at all made excellent conditioners for the commercial/industrial/medical market decades ago; if you find one of those it will likely need refurbishment. But they are the best I've seen. I was introduced to them by Michael Percy who was a dealer of ours before going into selling high end parts for audio use.

The Elgar has a massive transformer in it that is set up to be able to boost the AC power if needed. A power amplifier is used internally to amplify the feedback signal from the output of the conditioner; this feedback signal is compared to a sine wave from a low distortion oscillator that is synchronized to the AC line frequency. The feedback signal is applied to a feedback winding of the transformer and the output voltage adjusted to a regulated (adjustable) output. THD is 0.5% right to full output amperage. So the 5th as well as high frequency noise (which is usually less problematic in most power supplies) are absent at the output. Its an impressive bit of engineering and a pity that no high end manufacturer has seen fit to do anything of that scale. Customers of ours that have refurbished units report that the sound is improved across the board. Elgar made some pretty big units- the 3000 series could run 28Amps continuous! Usually the best installation is to have them wired into the AC line feeding the audio room at the breaker box.

 

[tima]

A good conditioner will be able to filter out the 5th harmonic, which is arguably the worst offender when it comes to 'dirty power'.

Thanks Ralph - your posts are always informative

I only knew Elgar from high school graduation.

 

[Kingrex]

Power filter is such a put your dukes up word. As is black background. Black background for such an odd reason. People we shout till blue in the face reinforcing what you just said, but 2 words are slightly different in the description and that ends their world.

I think everyone agrees we all attempt to get rid of equipment noise that gets into the playback signal and amplification chain. Who want's injected noise that was not in the source? I don't. Well, ok, lets fight now. Every single piece of electronics you own has a voice. Subtle as it may be. That is why there are so many brands and everyone finds a preference to one or another. Hopefully. And that voice, or noise, is what we decide upon in choosing the playback gear in out home, cars, office etc.

 

[stehno]

Atmosphere is correct in that there are very few superior line conditioners on the market. It's like looking for a needle in a haystack but they do exist.

I've heard some inferior LC's at customers / friends homes where the system actually sounded more musical once we removed the LC's and just plugged the gear straight into the wall. The only LC's I can recommend are Foundation Research (now defunct) and Jena Labs. Jena Labs has about 4 different model LC's which all I presume follow similar methodologies. I've auditiioned three of their models and currently own 3 models named THE TWO's which were slightly more musical than my Foundation Research units.

Like the Foundation Research units, Jena Labs THE TWO LC's are rather small passive, dedicated, and bi-directional filtering and most importantly, they actually do much to cleanse, purify, filter, and/or condition the noisy AC coming in from the street. In contrast, inferior LC's will either do little or nothiing or actually induce their own sonic harm.

I've been using what I consider superior line conditioners since 2000 and these or perhaps something else equivalent used to address the noisy AC are quite possibly one of the most musical gains one might ever obtain from a right-out-of-the-box perspective. Why some don't take AC treatments more seriously is beyond me. In fact, once conditioned to listening with superior LC's, my ability to tolerate other systems employing inferior or no AC treatment is about 10 min and I'm done. Life's too short.

 

[Cycling Nut]

I think lower noise and lower distortion are very beneficial to let the music thru as it is. Having less distortion and noise being called a "blacker background" is strange. "More music, less noise and distortion" is a mouthful. I try to not add noise or distortion, of course I havent fully succeeded yet!

 

[PeterA]

I think lower noise and lower distortion are very beneficial to let the music thru as it is. Having less distortion and noise being called a "blacker background" is strange. "More music, less noise and distortion" is a mouthful. I try to not add noise or distortion, of course I havent fully succeeded yet!

I agree. I have always found “blacker backgrounds“ associated with less information and less resolution. I consider it to be a coloration. I’ve never heard black backgrounds when listening to live music.

 

[Atmasphere]

Having less distortion and noise being called a "blacker background" is strange.

It might be strange, but from an engineering POV that is what is needed. The language is simply the idea that the music provides all the color and the electronics does not: it stays neutral.

I’ve never heard black backgrounds when listening to live music.

Of course not! That's because its the real thing.

Whatever the music is, it has to be reproduced by the electronics. the electronics needs to be low noise and low distortion to do that. The tricky bit is that some types of low level distortion are the noise floor in many systems, which causes them to have less low level detail. The ear can make out detail below the noise floor, to a limited extent, through natural hiss caused by wind, tube and transistor hiss. This is the one 'sort of' exception to the masking rule. But if the noise floor is composed of harmonic, inharmonic and IMD information then the ear cannot penetrate it at all; any low level detail below that kind of noise floor is lost.

If an audio circuit employs feedback but in insufficient quantity, then the noise floor of the circuit will contain these elements. You either need to run so much feedback that the circuit can clean this stuff up, or no feedback at all. That's how you get the blackest backgrounds.

 

[stehno]

It might be strange, but from an engineering POV that is what is needed. The language is simply the idea that the music provides all the color and the electronics does not: it stays neutral.

I'm guessing some of us are talking apples and oranges here.

Of course not! That's because its the real thing.

You may be a designer of components but as audio enthusiasts listening to live music where it's impossible for a black background to exist, enthusiasts also potentially listen to recorded music via their playback systems with the desire or hope to strive closer to live music. That same place where it's impossible for black backgrounds to exist. Are you saying because electronics are involved, it's impossible to achieve this lack of black background through a playback presentation?

Whatever the music is, it has to be reproduced by the electronics. the electronics needs to be low noise and low distortion to do that. The tricky bit is that some types of low level distortion are the noise floor in many systems, which causes them to have less low level detail. The ear can make out detail below the noise floor, to a limited extent, through natural hiss caused by wind, tube and transistor hiss. This is the one 'sort of' exception to the masking rule. But if the noise floor is composed of harmonic, inharmonic and IMD information then the ear cannot penetrate it at all; any low level detail below that kind of noise floor is lost.

The ear can make out detail below the noise floor? In another recent threat discussing a playback system's electronics-induced noise floor I kinda' reached the conclusion that there are multiple noise floors (NF), like maybe 4 or more that we're potentially dealing with.

For example. For the high-end audio enthusiast, we have a playback system's electronics induced NF, an environmental NF (kids playing, dogs barking, lawnmowers, etc), a bass NF induced by a speaker's mechanical energy / acoustic reaction to a given room's borders, etc. I think there were a few more NF considerations as well. Photography has at least one NF and I'm sure other industries have varied noise floors as well.

Noise floors do not have to include audible noise. As you might agree, noise is just another word for distortions and distortions can include audible and/or inaudible noise. I'm of the camp that the most serious distortions are of an inaudible nature. However, in my limited endeavors, I'm simply unaware of a time witnessing anybody claiming they could hear or see below a noise floor. To me, it's an oxymoron.

I think a good simple example is the mechanical energy / acoustic noise floor having to do with quality of bass reproduction. From time to time some of us play around with speaker positioning within our rooms, sometimes for aesthetic reasons and other times for improved bass response, etc.

When moving a speaker around and listening for before / after differences it's not uncommon to hear differences in the bass reproduction. If the difference is genuinely better we hear tighter, deeper, more well-defined bass, etc, etc, and guess what, we hear bass notes that previously where hidden or inaudible. If a speaker position is genuinely worse, we hear the opposite, including hearing fewer bass notes where some notes actually become inaudible. But there's more benefits to this bass-related noise floor. For example. Another ancillary benefit is that when the bass is dialed in, a previously lean sounding playback presentation can suddenly sound far more balanced. Moreover, some higher frequencies that were on the verge toward fatiguing (think a bit shrill or on the edge of ear bleed) suddenly are gone or greatly diminished when the bottom end is dialed in. Yet, nothing was added or substracted as it was just speaker positioning. So perhaps this thing I call bass noise floor actually includes a few ancillary noises floors within.

Anyway, I've yet to ever heard anybody claiming to hear or see below a noise floor. This is new to me. If one could hear / see below (and above) a noise floor, then what exactly is a noise floor? Is it truly a noise floor? Or might some interpret some form or type of distortion and just call that measurement the noise floor? I mean, high-end audio is perhaps 100% subjective with very few standards so we actually have the freedom to label or claim most anything we want, right? One can say the bass is sloppy like a rolling earthquake while another might say it's the most musical bass they ever heard and in the end neither is proven incorrect.

If an audio circuit employs feedback but in insufficient quantity, then the noise floor of the circuit will contain these elements. You either need to run so much feedback that the circuit can clean this stuff up, or no feedback at all. That's how you get the blackest backgrounds.

Hmmmmm. Maybe all designers think as you and perhaps use the same vocabulary or might you be a bit unique in this regard? I'm asking because maybe I'm the one who's unique in this regard. But I am curious what you think is going on when say, you replace a pair of ic's with a slightly superior / more musical set? If indeed you're playback presentation is genuinely a tad more musical now, is that not because you're actually hearing more music info or do you suppose you're already hearing all the music info you can ever hope for from a given recording and the ic upgrade just makes the music info already audible a bit cleaner?

Personally, I think both are quite possible, but going back to the speaker / bass and mechanical energy/acoustic noise floor example above, there were no upgrades, subtractives, nor additives. Just speaker movement and with those movements bass notes appear and disappear and the quality of bass deteriorates or improves.

Until now I've considered a playback system's noise floor to be a collective pool of electrical energy distortions whose sum total equates to the noise floor. But perhaps that's not quite so accurate. We can only listen to a playback sytem as a collective whole but if a pair of ic's swapped in improves the sound and we hear more, then maybe that implies every part of the playback vineyard has its own isolated noise floor with little / no overlap into other parts of the vineyard. But I digress.

Anyway, at your test bench, do you R&D with the assumption that your speaker output is already capturing 100% of the music info embedded in and retrieved/processed from a given recording? Or are you working from the premise that of the 100% music info embedded in and retrieved / processed from a recording, what you audibly hear at your speakers is less than 100% of the music info read in?

 

[Atmasphere]

The ear can make out detail below the noise floor?

Yes, but only if its certain forms of hiss as I mentioned. But also as I mentioned there are certain forms of what sounds like hiss that block the ear's ability to hear into it. As previously mentioned that latter form contains harmonic, inharmonic and intermodulation components.

I'm guessing some of us are talking apples and oranges here.

Yes- and I was careful to point that out although not in so many words.

Are you saying because electronics are involved, it's impossible to achieve this lack of black background through a playback presentation?

This is an issue of perfection , which does not exist in this world. But we can strive to get the electronics to have such a black background that the music is pretty well represented. IME we do get glimmers of that from time to time. When the system does something that makes me spin about and look to where that sound came from, IOW when it actually fools me (which is what its supposed to do) its a bit spooky and kinda cool.

I am curious what you think is going on when say, you replace a pair of ic's with a slightly superior / more musical set?

With regards to interconnects, I use balanced lines and gear that supports the balanced line standard, which eliminates hearing any significant differences in cables. The more variables eliminated the better.

at your test bench, do you R&D with the assumption that your speaker output is already capturing 100% of the music info embedded in and retrieved/processed from a given recording? Or are you working from the premise that of the 100% music info embedded in and retrieved / processed from a recording, what you audibly hear at your speakers is less than 100% of the music info read in?

I don't worry about what the speakers are going to do, just how well our gear will present the signal to the speaker. We can't control what the customer buys. Its assumed that we're not getting everything out of our recordings- and that is why we try to do better. I think everyone sees it that way but I could be wrong and it wouldn't be the first time!

 

[stehno]

Yes, but only if its certain forms of hiss as I mentioned. But also as I mentioned there are certain forms of what sounds like hiss that block the ear's ability to hear into it. As previously mentioned that latter form contains harmonic, inharmonic and intermodulation components.

I'm curious why you associate audible noises / sounds with a playback system's electronically-induced noise floor threshold?

Yes- and I was careful to point that out although not in so many words.

This is an issue of perfection , which does not exist in this world. But we can strive to get the electronics to have such a black background that the music is pretty well represented. IME we do get glimmers of that from time to time. When the system does something that makes me spin about and look to where that sound came from, IOW when it actually fools me (which is what its supposed to do) its a bit spooky and kinda cool.

Sure, it's an imperfect world and high-end audio is no exception to that but I don't think it's relevant. There is also a noise floor not yet mentioned and that's the noise floor of the playback presentation overshadowing certain audible noises / sounds / imperfections we prefer not to hear anyway. Unless of course we prefer listening at elevator music volume levels.

I could be wrong but this I think is what the thread was intended to be about. Not necessarily what a component designer does at the test bench.

You said above that there is no black background with live music because it's a live performance to which I agree. So if my pursuit for my playback systems is to strive closer toward the live performance, why should I expect black backgrounds there? Regardless of genre and venue for the live event, why should anybody think a complete lack of a black background is unachievable in our playback presentations?

Tape or component hiss or ticks and pops from a format should be irrelevant to a system's noise floor, except that they too are audible frequencies and hence, they too should be impacted the same as music info is impacted. Especially since a well-enough designed component ought not or perhaps cannot discriminate between sound types. Sound is sound, right? Unless perhaps a designer were to suppress certain offending frequencies and thereby suppress the music info in those same frequencies? As I recall, the Ayre Conditioners did exactly that. But I certainly hope that's not being done on a wide scale.

With regards to interconnects, I use balanced lines and gear that supports the balanced line standard, which eliminates hearing any significant differences in cables. The more variables eliminated the better.

That's debatable. I've heard some poor sounding balanced cables in time past but that's another topic and irrelevant to this topic except as an illustration I pulled from a hat.

I don't worry about what the speakers are going to do, just how well our gear will present the signal to the speaker. We can't control what the customer buys. Its assumed that we're not getting everything out of our recordings- and that is why we try to do better. I think everyone sees it that way but I could be wrong and it wouldn't be the first time!

Yes, speaker choice is irrelavant to this topic as well. My earlier point was music info is read in, processed, and due to a playback system's established noise floor level will determine what is audible / inaudible at the speaker - any given reasonable speaker.

In conclusion, I get the impression you're saying that because we employ electronics in an imperfect world, our hopes of ever achieving a reasonable counterfeit of a live performance are pie-in-the-sky dreams. Simply because our electronics guarantee we'll continue to experience black backgrounds and as a result volumes of low-level detail from the live performance captured at the recording mic's and embedded in the recording will remain inaudible and replaced with a black background?

 

[Atmasphere]

I'm curious why you associate audible noises / sounds with a playback system's electronically-induced noise floor threshold?

Experience.

There is also a noise floor not yet mentioned and that's the noise floor of the playback presentation overshadowing certain audible noises / sounds / imperfections we prefer not to hear anyway.

Actually that is precisely what I've been talking about in my last few posts here.

I could be wrong but this I think is what the thread was intended to be about. Not necessarily what a component designer does at the test bench.

I'm going with 'wrong on this' but only because of the thread title, which encompasses design aspects in addition to other issues.

if my pursuit for my playback systems is to strive closer toward the live performance, why should I expect black backgrounds there? Regardless of genre and venue for the live event, why should anybody think a complete lack of a black background is unachievable in our playback presentations?

Well, I don't expect it, but I won't buy the product it if its not there. With regards to the latter question, because perfection is not attainable in this world. But we can get pretty close.

Tape or component hiss or ticks and pops from a format should be irrelevant to a system's noise floor,

Actually with regards to LP playback this statement is false. Phono preamps can generate ticks and pops entirely on their own, incited by an electrical resonance caused by the inductance of the cartridge and the capacitance of the tonearm cable. If that peak goes into excitation, and if the phono section's input circuit has poor high frequency overload margins (which is a common problem) then it can generate a tick or pop when it overloads. To prevent this, the peak has to be in consideration by the designer, and you certainly won't get there with a poor noise floor.

That's debatable. I've heard some poor sounding balanced cables in time past but that's another topic and irrelevant to this topic except as an illustration I pulled from a hat.

Yup! That is entirely because high end audio does not recognize nor adhere to the balanced line standard a.k.a. AES48. The equipment I play in my system does and so makes the cables neutral and transparent, and as a nice side benefit, keeps cable costs down too. I've shown at audio shows with 50 feet between the preamp and amp of inexpensive studio cable and gotten 'best sound at show' a number of times. And you might keep in mind that most of your favorite recordings employ balanced cables, but done according to AES48. This ignoring of the standard in high end, whether willful or out of ignorance, is rampant!

In conclusion, I get the impression you're saying that because we employ electronics in an imperfect world, our hopes of ever achieving a reasonable counterfeit of a live performance are pie-in-the-sky dreams.

Not exactly. If you use headphones and listen to microphone feeds, you find that electronics as well as microphone and headphone transducers have been where they needed to go for a very long time: decades. They can be very convincing! I've experienced this many times in the studio, and I've seen jaundiced audiophiles turn as white as a ghost when doing exactly that- the mics and 'phones convinced them that what they heard was real and happening in their presence when it was happening in another room. So it seems the big challenges are with the amps, speakers, and media.

 

[Kingrex]

What the science minds forgets to do is experiment. To much time talking theory and numbers. To much time postulating theory. If it could even be called that.

Just plug it in and listen. But be aware. If your going to listen through a piece, you need a flow plan on how to evaluate. There are threads on this. There are process that help get your subjective opinion and beliefs out of the listening test.

 

[stehno]

What the science minds forgets to do is experiment. To much time talking theory and numbers. To much time postulating theory. If it could even be called that.

....

Very well said, Kingrex. Tesla has a nice quote about this.

"Today's scientists have substituted mathematics for experiments, and they wander off through equation after equation, and eventually build a structure which has no relation."

If it was true 100 years ago, I can only imagine how much worse it might be today.