Computer Audio Asylum

Music servers and other computer based digital audio technologies.

Return to Computer Audio Asylum


Message Sort: Post Order or Asylum Reverse Threaded

People here keep acting as if it is simple to hear a -120dB down signal.

71.146.8.173

Posted on September 12, 2014 at 00:58:38
This is what it looks like when folks hear a +120db signal.

Does any one really think those folks are worried about an oscillation -120dB down?

+120db sound levels are not just an abstraction, they physically shove people around. How many posters here have stereos that actually push them around their listening room? And while flying from sofa cushion to sofa cushion, let them hear which DAC has lower jitter?

JE

 

Hide full thread outline!
    ...
RE: People here keep acting as if it is simple to hear a -120dB down signal., posted on September 12, 2014 at 05:00:16
fmak
Audiophile

Posts: 13158
Location: Kent
Joined: June 1, 2002
Read up on acoustics, the dynamic range of the ear is reckoned to be 130dB or even better for some

 

RE: People here keep acting as if it is simple to hear a -120dB down signal., posted on September 12, 2014 at 06:57:05
Ugly
Audiophile

Posts: 2912
Location: Des Moines, WA
Joined: August 22, 2006
I doubt even the best listening rooms of Asylum board users is better than around 20-30dB above absolute zero ambient acoustic noise levels. How are you going to hear that -120dB signal with a system set up for playing a typical bit of program even if you had one of the best rooms? The math apparently isn't supporting the idea. Sure you can gain up the noise enough to hear but if you throw a CD on when it's turned up that loud something will fail whether it's your system or the building supporting it. Not a practical situation at all. What's the point?

 

RE: People here keep acting as if it is simple to hear a -120dB down signal., posted on September 12, 2014 at 07:47:13
b.l.zeebub
Audiophile

Posts: 9361
Location: 52deg 28'N,1deg56'W
Joined: April 17, 2006
I assume that you use transistor amps because -120dB is between 1000 and 10 000 times lower than your typical valve amp distortion (0.1-1%).
If THD is that much higher there is no chance that anyone could possibly hear anything at -120dB as it would get completely drowned out by the distortions.

I suppose 0.0001% THD could be achieved by a SS amp if it uses vast amounts of negative feedback.

 

RE: People here keep acting as if it is simple to hear a -120dB down signal., posted on September 12, 2014 at 07:54:30
Tony Lauck
Audiophile

Posts: 13629
Location: Vermont
Joined: November 12, 2007
"I doubt even the best listening rooms of Asylum board users is better than around 20-30dB above absolute zero ambient acoustic noise levels."

The confusion comes between broadband noise levels (according to some weighted curve) and narrow band noise levels. If there is no operating HVAC system running and noisy appliances are not in use, rural settings will have little or no noise in the frequency range where human hearing is most sensitive. In addition, it is possible to hear single tones that are as much as 20 dB below the broadband noise floor. Except for low frequency noise or when appliances or HVAC systems are running, what one can hear is going to depend on volume control settings, the recordings on plays and the sensitivity of one's ears, not room noise.

As to those volume control settings, and so as to be accused of not cheating, peak sound pressure levels at live orchestral concerts have been measured at over 120 dB (e.g. Mahler symphonies, row 5). A dynamic range of this amount is required for realistic reproduction of this music. Some of us have systems with this peak capability throughout most of the frequency range applicable (except possibly for organ pedal notes). And some of us turn up our volume controls to achieve these realistic levels.

Some sounds, for example jangling keys, have peak to average sound levels that are greater than 30 dB. Dropping one's keys on a stone floor can produce peak SPLs of 140 dB, but there will be no building collapse. I have played such a recording at volume control settings that I normally use for playing orchestral music and the jangling keys were not even at natural volume levels.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

 

I know a troll when I smell it - and this one literally stinks. N/T, posted on September 12, 2014 at 09:33:35
carcass93
Audiophile

Posts: 7181
Location: NJ
Joined: September 20, 2006
N/T

 

I said hearing in SPL, posted on September 12, 2014 at 09:51:45
fmak
Audiophile

Posts: 13158
Location: Kent
Joined: June 1, 2002
nt

 

So what?, posted on September 12, 2014 at 10:48:51
b.l.zeebub
Audiophile

Posts: 9361
Location: 52deg 28'N,1deg56'W
Joined: April 17, 2006
If you get a tube amp to produce 130dBspl through your speakers you will not hear anything at -120dBspl because the distortion noise is 1000 to 10 000 times stronger.

Alternatively you could stand next to an air hammer and the chances are that you won't be able to hear the ticking of a watch you might be wearing.

While we can perceive noises or signals from a few dBspl to about 120+dBspl we can not perceive them if they are happening simultaneously. For simultaneous aural stimulation we can hear differences of about 40-50dB. Anything more than that and the lower one will be completely masked.
Similarly and using a Mahler crescendo of 120dB or more you won't be able to tell if the triangle is being played or not.

 

"differences of about 40-50dB" - couple of questions for you., posted on September 12, 2014 at 11:09:12
carcass93
Audiophile

Posts: 7181
Location: NJ
Joined: September 20, 2006
1. Where did you get those numbers? The same place where creators of MP3 codec got theirs, along with masking they love to talk about? Why did it fail so miserably then in their case, even at the highest bit rates, to sound exactly as CD, with its measly 96 dB of dynamic range?

2. Are you able to hear a difference between power cords, interconnects, USB cables, media players on PC, ASIO vs. KS vs. WASAPI etc.? If not, I respectfully suggest doing something about it, like ASAP. But, if yes - what would be your guess, how far down below (in dB) the measured differences would lie?

 

RE: People here keep acting as if it is simple to hear a -120dB down signal., posted on September 12, 2014 at 11:14:42
AbeCollins
Audiophile

Posts: 46196
Location: USA
Joined: June 22, 2001
Contributor
  Since:
February 2, 2002

"...it is possible to hear single tones that are as much as 20 dB below the broadband noise floor."

That's why I used Morse Code as a HAM Radio operator when I was a kid. One could audibly decode a series of weak dots and dashes (approx 1KHz tones) in the noise floor where normal voice transmissions would be unintelligible.

Additionally, the audio bandwidth required for voice transmission (SSB) was about 2KHz so 2KHz bandwidth filters were used. Dropping the filter bandwidth also improved signal-to-noise ratio, dropping broadband noise by nearly another 10dB by using narrow filters CW (Morse Code) vs ~2KHz for SSB voice.



 

RE: I know a troll when I smell it - and this one literally stinks. N/T, posted on September 12, 2014 at 12:14:18
Old SteveA
Audiophile

Posts: 648
Joined: March 27, 2011
The best tactic to dealing with a suspected troll is "!!!!!!!!!!!!!!!!!!!!"or "???????????????????????????????????????????????"

(I crack myself up sometimes; additionally I hope I occasionally entertain others)

 

RE: So what?This, posted on September 12, 2014 at 12:41:15
fmak
Audiophile

Posts: 13158
Location: Kent
Joined: June 1, 2002
is just argument for the sake of it - no response.

'-120dBspl because the distortion noise is 1000 to 10 000 times stronger'

 

If your..., posted on September 12, 2014 at 13:34:10
audioengr
Manufacturer

Posts: 6017
Location: Oregon
Joined: April 12, 2001
ears are young and not ruined yet by loud rock shows and your system is quiet and resolving enough, why not?

 

RE: If your..., posted on September 12, 2014 at 15:49:27
AbeCollins
Audiophile

Posts: 46196
Location: USA
Joined: June 22, 2001
Contributor
  Since:
February 2, 2002

Some legit questions for you and I'm not trolling....

Is it not true that most of what we consumers listen to came off master tapes? Is it also not true that these tapes have a dynamic range of about 70 or maybe 80dB at best? I would also assume that this is under ideal conditions that are rarely met.

Why would one need 120dB of dynamic range in a system when the source material will never take advantage of it? At 120dB down aren't we also well into the noise floor of most audio systems?


 

Inmates are resistant to the idea of audio masking, posted on September 12, 2014 at 16:23:52
since of course if humans can only hear about -50dB into a song or signal that is playing then all the claims of hearing unmeasurable differences between power cables and USB cables fly out the window.

This explains why the Audio Diffmaker is such a bogeyman here. Inmates are afraid to even try it as it might upset their view of audio on two points: one, their ears aren't as superhuman as inmates think they are and two, the dratted machine can detect audio information the inmate could not.

Here's a thought experiment. How loud can a healthy NFL quarterback shout? I'm guessing after a lifetime of calling signals as a class they've got pretty strong voices. However, there are NFL stadiums that routinely have crowd noise measured at 130db when the visiting team is on offense. In those situations it's next to impossible for the quarterback to make himself heard by his teammates so clubs go to the so called "silent count" where visual signals are used instead. Pro football clubs certainly have no problem believing in audio masking.

JE

 

RE: People here keep acting as if it is simple to hear a -120dB down signal., posted on September 12, 2014 at 19:49:53
Ugly
Audiophile

Posts: 2912
Location: Des Moines, WA
Joined: August 22, 2006
" If there is no operating HVAC system running and noisy appliances are not in use, rural settings will have little or no noise in the frequency range where human hearing is most sensitive. In addition, it is possible to hear single tones that are as much as 20 dB below the broadband noise floor."

I would have guessed most Asylumers are urban to suburban but don't know. I never get near complete silence around here that I can remember. When it gets really quiet my ears agc starts cranking up the gain and I still hear noise.

Are you expecting your noise floor to be single tones, or that people want to try and listen to program material below the noise floor? Neither situation seems likely.

"A dynamic range of this amount is required for realistic reproduction of this music."

I don't really know what your listening to but none of the recordings I have use anywhere near 120dB dynamic range as part of the presentation. Even if you were super exceptional and managed a normal listening unfiltered ambient listening position room noise level of +10dB of whatever band you like, which there is very little likelihood for most I'd guess, you'd still need 130dB output at the listening seat to be able to to be guaranteed to hear it all of a 120dB dynamic range show. Sounds a little brutal to me.

"Dropping one's keys on a stone floor can produce peak SPLs of 140 dB, but there will be no building collapse."

I'd imagine if you could sustain say 140- 150dB unfiltered whiteish, around 10-10kHz, noise with fairly even distribution throughout most structures you'd start having problems. Again, this is just a guess but it sure sounds like a fun experiment I may never get to try.

 

RE: "differences of about 40-50dB" - couple of questions for you., posted on September 13, 2014 at 07:44:54
b.l.zeebub
Audiophile

Posts: 9361
Location: 52deg 28'N,1deg56'W
Joined: April 17, 2006
Klippel test of added distortion. Easily heard through my system down to 0.5%THD (-46dB).
Which is why I would not ever employ a valve amp of any kind, I want my electronics to be at least one order of magnitude below my speakers when it comes to adding distortions of any kind.

I do not use mp3s since they sound horrible. I don't even use lossless codecs.
I also do not use USB except to connect my keyboard and mouse to the computer.
USB is a deeply flawed and inelegant way to stream anything.
And while I fairly easily hear differences in amps or convertors I have never heard any between standard, decent interconnects or speaker cables and more expensive ones.

I have no idea about ASIO, KS or WASAPI since none of that stuff exists on my computer.
However I do remember ASIO from a couple of decades ago and it was shit and lacking in elegance in design, very much like USB.

 

RE: People here keep acting as if it is simple to hear a -120dB down signal., posted on September 13, 2014 at 09:02:04
Tony Lauck
Audiophile

Posts: 13629
Location: Vermont
Joined: November 12, 2007
I'm sorry for you if you never get silence. Not only is it bad for listening to music, it is bad for health. Where I live if I shut off my appliances I get silence, as there are few neighbors around most days. On a non-windy day the only outside noise I hear are roosters, ravens and geese, plus seasonal lawn mowers and snow plows. The "main" rood is 200 yards away and it has little traffic outside of ski season. Last week was a bit noisy, as the road was repaved.

The dynamic range created by musicians in a symphony orchestra (loudest tutti attack vs. quietest solo sustained note) is about 55 dB. However, in the pauses between notes there is hall decay down to the threshold of audibility. The actual dynamic range experienced by the listener will depend on how quiet the hall is, where the listener is sitting, etc. but can be 120 dB. Hearing all the decaying ambiance may not be necessary to enjoy the music, but it's there and if it can't be heard in a recording this aspect of realism will be lost. It is possible to achieve this level of dynamic range using high resolution digital formats and suitable low noise microphones. You should be able to hear residual hall noise from the recording venue on the recording, not hum or hiss or other recording or system artifacts.

I checked and I can reliably hear a 1000 Hz tone recorded at a level of -100 dB (96/24) using volume settings that I customarily use to playback large scale orchestral recordings. I didn't shut off any appliances, but I did close the door to my listening room. My computer is not completely silent, so I couldn't do much better than this due to fan noise and possibly disk noise. I will try this again on another occasion, this time using a silent computer and powering down everything in my home other than the computer and audio equipment.

We don't have to "try" to listen to sounds below the noise floor. If we didn't have that ability we wouldn't be here today, as our ancestors would have been eaten before they reproduced.

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

 

RE: If your..., posted on September 13, 2014 at 09:29:28
audioengr
Manufacturer

Posts: 6017
Location: Oregon
Joined: April 12, 2001
Abe - because the S/N still means noise level, and noise added is noise added, no matter what the noise level is in the recording.

 

RE: If your..., posted on September 13, 2014 at 09:34:32
Tony Lauck
Audiophile

Posts: 13629
Location: Vermont
Joined: November 12, 2007
"Is it not true that most of what we consumers listen to came off master tapes? Is it also not true that these tapes have a dynamic range of about 70 or maybe 80dB at best? I would also assume that this is under ideal conditions that are rarely met."

Most consumers listen to MP3s or equivalent. These people and their tastes in music and listening are not relevant to audiophile forums.

If high resolution digital formats (24 bit PCM or DSD) are used the recoding medium and converters will have no problem achieving 120 dB dynamic range. The weakest link is going to be the microphone and microphone preamplifier which are limited by random motion of air molecules and Johnson noise. State of the Art microphones used to record classical music have a dynamic range of approximately 120 dB. The best recording labels use minimal post production, so one hears the first generation output of the microphones. (Example are the Channel Classics recordings, most of which are pure DSD with no post processing.)

"At 120dB down aren't we also well into the noise floor of most audio systems?

Not good ones.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

 

RE: If your..., posted on September 13, 2014 at 13:46:33
AbeCollins
Audiophile

Posts: 46196
Location: USA
Joined: June 22, 2001
Contributor
  Since:
February 2, 2002
"If high resolution digital formats (24 bit PCM or DSD) are used the recoding medium and converters will have no problem achieving 120 dB dynamic range."

I know that most consumers today probably listen to MP3. When I said 'consumers' I was talking about the medium from which recording engineers and such work from before the final product is sold and delivered to the consumer.

Assuming the working medium is a master tape, are you saying we will still be able to achieve 120dB dynamic range to and from the tape? If memory serves me, I thought high-end pro tape machines and their tapes are limited to less than 80dB.


 

RE: If your..., posted on September 13, 2014 at 13:48:19
AbeCollins
Audiophile

Posts: 46196
Location: USA
Joined: June 22, 2001
Contributor
  Since:
February 2, 2002

I don't think that answered my questions but thanks for the attempt.

 

RE: If your..., posted on September 13, 2014 at 14:11:06
Tony Lauck
Audiophile

Posts: 13629
Location: Vermont
Joined: November 12, 2007
The best medium from which the best recording engineers presently work today is a live microphone feed. This is captured directly by the ADC and distributed unchanged (except for "razor blade" editing) directly to the consumers. This is how Channel Classics works and why their sound quality is better than any other classical music label issuing recordings today. They also happen to use the best available DSD ADC.

There are first rate engineers working with tape at high speeds and width, presumably because they like the slight dynamic compression tape provides. This is suitable for some styles of recording and some musical genres. This is what Blue Coast Records does. They then mix the tape down to DSD and other digital formats.

With the arrival of state of the art DSD converters such as the Horus that work up to DSD256, expect to see other production processes come into play. At present, recording directly to DSD has not been practical for musical genres that depend on heavy post production, because of the generation loss converting in and out of DSD64. (This is apparent in a few of the Channel Classics tracks I have that are not pure DSD because post production was necessary for artistic reasons. I have both the original and remixed files of one of these tracks and can hear the sonic degradation from the DSD - DXD - DSD processing.)

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

 

Audiophiles prefer Formula E (nt), posted on September 14, 2014 at 09:42:08
Scrith
Audiophile

Posts: 1169
Location: Los Angeles
Joined: July 19, 2005
:-P

 

RE: If your..., posted on September 15, 2014 at 08:31:42
AbeCollins
Audiophile

Posts: 46196
Location: USA
Joined: June 22, 2001
Contributor
  Since:
February 2, 2002
Thanks Tony, I understand what you're saying about direct from mic to high resolution digital recording but here's what I'm wondering about:

If the masters reside on tape as many albums do, and tape has a dynamic range of no more than say 70 to 80dB at best, and that tape is used to create 'high-res' digital files, what is the benefit of a DAC with 100dB+ dynamic range if the source material's range is significantly less?





 

RE: If your..., posted on September 15, 2014 at 09:47:12
Tony Lauck
Audiophile

Posts: 13629
Location: Vermont
Joined: November 12, 2007
"If the masters reside on tape as many albums do, and tape has a dynamic range of no more than say 70 to 80dB at best, and that tape is used to create 'high-res' digital files, what is the benefit of a DAC with 100dB+ dynamic range if the source material's range is significantly less?"


If the dynamic range of the DAC is limited by purely additive noise and the additive noise has the same spectrum as the tape noise, then, given that the tape noise was 20 dB louder, there would probably be no benefit to a DAC with such dynamic range, provided that the digital transfer used all of the available dynamic range of the digital media (give or take a few dB) and provided that no digital volume control was used. If one of these provisos does not apply then more dynamic range in the DAC would probably make a difference.

Personally, I could care less about purely additive white noise, so long as it is at a low enough level. The problem is that the DNR of the DAC may be limited for reasons other than additive analog noise. Even something as simple as TPDF dither does not result in purely additive noise, despite the lies Redbook apologists tell. So if the noise comes from any portion of the digital processing, then the music is likely to be distorted unless the designer really understands digital signal processing.

Perhaps the most significant reason for a low DNR in a DAC is that it can only be achieved with paranoid attention to detail on the part of the designer. However, if there are design reasons for the noise that have other benefits at the expense of benign noise, then such a DAC may be preferable, even though it measures poorly on DNR. Some DACs push their analog electronics to measure good and this leads to an unnecessarily harsh sound (even digital clipping) on loud portions of recordings. This is just one of many reasons to be distrustful of products that are designed to measure good, rather than sound good.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

 

I don't know..., posted on September 16, 2014 at 02:16:52
but I have this crazy idea that >120dB sound levels are not good for my ears. While I appreciate your business's need to have a stream of affluent but deafened customers to support your sales targets, I must, with all due respect, decline to join that stream.

JE

 

you dont get it, posted on September 16, 2014 at 10:20:55
audioengr
Manufacturer

Posts: 6017
Location: Oregon
Joined: April 12, 2001
its -120dB, not +120dB

 

Ah! My mistake! Sorry about that! nt, posted on September 16, 2014 at 10:43:14
nt

 

Page processed in 0.042 seconds.