Quantcast
bob ludwig and loudness - Page 2 - Gearspace.com
The No.1 Website for Pro Audio
bob ludwig and loudness
Old 27th January 2009 | Show parent
  #31
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
Quote:
Originally Posted by 24-96 Mastering ➑️
Reading this today seems like pretty basic research, only marginally better than static weighing curves. This was in 1936. What happened since?
Very little, if anything at all. Since phon and sone, whose origins are telephony, we as an industry have done virtually nothing to further define a unit of "loudness". It's such an elusive parameter, based on too many variables of sound power, intensity, bandwidth, crest factor, impulsive vs. steady state, etc.

It's truly a can of worms.

I'm not saying it wouldn't be a worthy cause, just that it's only those of us here in this special little corner of the audio world that care about quantifying it at all. For everyone else, the decibel, in all of its manifestations, works fine. But for quantifying loudness? Particularly for music? It fails miserably.
Old 28th January 2009 | Show parent
  #32
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by Superdisc ➑️
Some times(if not all times) 10 dB will sound much louder than twice as loud.
Agree!

Quote:
Useful Notes in using the Decibel Unit:
1. One dB represents approximately the smallest volume change which can be heard if listened for carefully. One dB represents an approximate change in pressure of 12%.
Actually level differences down to 0.1dB can be heard. In scientificall audio testing it is common to match to 0.05dB in order to avoid false results from a level difference between two audio samples.


Quote:
2. Three dB represents a small but noticeable change in volume. .
Actually 3dB is concidered being a to big step in hifi preamps so I wouldn't say small. I even know people that said a 3dB difference was experienced as a doubling of SPL in a blind test.

Quote:
3. Six dB represents a change in level of twice (or half) as much pressure.
Yes, pressure or level and one small blind study I partcicipated in (10-20 participants with interest in audio) showed that for midrange sines at medium level a 6dB difference was experienced as doubling of SPL. Most ended up at 4-7dB or thereabout. I felt 5dB was double as loud.

Quote:
For all practical purposes, yes. Although the ear is not exactly pressure sensitive, it is closer to being pressure sensitive than to being anything else.

If not pressure sensitive, what would it be? I think it's safe to consider the ear a pressure sensor.

Quote:
The following is presented as factors supporting this:

1. For centuries, composers and conductors have used a formula that it takes four times the musicians to get twice the volume. If a composer/conductor wanted the violins to be twice as loud, they would specify 4 times as many. This is four times the power or a 6 dB volume increase.
Now it's getting tricky. Four times as many identical sources in phase and close in distance relative the wavelength being produced will put out 12dB more SPL but since a group of muscians will never be totally in phase (such as an array of speakers could be) the poor correlation will make the result closer to +6dB (+3dB for every doubling of non-correlated sources like noise for example). BUT once in a while the sum will potentially add up
in phase and that's why choir for example is extremely hard to cope with for a stereo playback chain. You need some serious power amps not to clip non-compressed choir on medium sensitive speakers.

Quote:
Except for the final amplifier that drives a speaker almost all equipment used in recording and sound reproduction is voltage sensitive.

Well actually a power amp typically is a voltage driven voltage source as well.


/Peter
Old 28th January 2009 | Show parent
  #33
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
Quote:
Originally Posted by Audiop ➑️
Now it's getting tricky. Four times as many identical sources in phase and close in distance relative the wavelength being produced will put out 12dB more SPL
Sorry Peter, but the answer is 6dB. Since we're talking about acoustic sources.
Old 28th January 2009 | Show parent
  #34
Lives for gear
 
🎧 10 years
fader8,

why did you quote only half the sentence? If you read the whole sentence (and the rest beneath it) I think we agree! ;-)


/Peter
Old 28th January 2009 | Show parent
  #35
Gear Nut
 
🎧 10 years
Quote:
Originally Posted by Audiop ➑️
Agree!


Actually level differences down to 0.1dB can be heard. In scientificall audio testing it is common to match to 0.05dB in order to avoid false results from a level difference between two audio samples.
I wouldn't doubt it and perhaps I should have written this a little differently. I was speaking more from the viewpoint of hearing a level change made in a music recording while critically listening to the music. That's a lot different than the difference that could be heard under the best circumstances during a scientific experiment.

Quote:



Actually 3dB is concidered being a to big step in hifi preamps so I wouldn't say small. I even know people that said a 3dB difference was experienced as a doubling of SPL in a blind test.
What you call what you hear in tests is a subjective measure of "small" - but the 3 dB point is used in many places as the point of "noticeable" change. For instance cut-off frequencies of filters are specified as the 3 dB down point. I find the 3dB being a "small but noticable" change as about the first point someone listening causually (like to the TV) would say, Hey did the volume change? It's the smallest amount a husband can get away with turning down that loud NFL game on TV as requested by the wife. heh

So, the first thing I say to my second semester mastering students is Cut that "small but noticeable change" down to 1.5 dB in your mind.



Quote:


Yes, pressure or level and one small blind study I partcicipated in (10-20 participants with interest in audio) showed that for midrange sines at medium level a 6dB difference was experienced as doubling of SPL. Most ended up at 4-7dB or thereabout. I felt 5dB was double as loud.


If not pressure sensitive, what would it be? I think it's safe to consider the ear a pressure sensor.
Well it basically is, but the ear also compresses due to the mechanical coupling into the hairs in the inner ear (gotta go though bones and such). So this could alter the pressure sensitive response. If this has elasticity, the inner ear may be less responsive to pressure changes on the ear drum.

The results of your study though pretty much match my experience, except I'm too used to the 6 dB reading on VU meters.

Quote:


Now it's getting tricky. Four times as many identical sources in phase and close in distance relative the wavelength being produced will put out 12dB more SPL but since a group of muscians will never be totally in phase (such as an array of speakers could be) the poor correlation will make the result closer to +6dB (+3dB for every doubling of non-correlated sources like noise for example). BUT once in a while the sum will potentially add up
in phase and that's why choir for example is extremely hard to cope with for a stereo playback chain. You need some serious power amps not to clip non-compressed choir on medium sensitive speakers.
Certanly unision parts would tend to build up better than harmony or totally disrelated sounds, but even for these, the 10:1 ratio to be "twice as loud" is too much IMO.
Quote:


Well actually a power amp typically is a voltage driven voltage source as well.


/Peter
Old 28th January 2009 | Show parent
  #36
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
Quote:
Originally Posted by Audiop ➑️
why did you quote only half the sentence? If you read the whole sentence (and the rest beneath it) I think we agree! ;-)
I quoted what I deemed to be your premise, in that if you have 4 identical, correlated sources, that the addition of their power would equal 12 dB. If these were signal paths feeding a mix bus, then that's true. The voltage gain would be 4 and and you'd see a 12dB increase on your bus meter. 20Log.

But 4 identical acoustic sources add powers, which uses 10Log, and therefore gives a 6dB increase when you quadruple the power.

You're absolutely right that in real world violin section conditions, such correlation isn't possible. But, for example, if you're stacking horns in a big concert sound system, it's very real and you'll come very close to these numbers.
Old 28th January 2009
  #37
Lives for gear
 
theblotted's Avatar
 
🎧 15 years
Quote:
Originally Posted by wado1942 ➑️
...and now there's CDs averaging -3.5dBfs....completely unlistenable mush. .
what song is averaging -3.5dBFS??
Old 28th January 2009 | Show parent
  #38
Lives for gear
 
Russell Dawkins's Avatar
 
🎧 15 years
Quote:
Originally Posted by Audiop ➑️
Now it's getting tricky. Four times as many identical sources in phase and close in distance relative the wavelength being produced will put out 12dB more SPL but since a group of muscians will never be totally in phase (such as an array of speakers could be) the poor correlation will make the result closer to +6dB (+3dB for every doubling of non-correlated sources like noise for example). BUT once in a while the sum will potentially add up in phase and that's why choir for example is extremely hard to cope with for a stereo playback chain. You need some serious power amps not to clip non-compressed choir on medium sensitive speakers.
/Peter
Ain't that the truth! Choir is a challenge to the playback equipment, and precious few speaker/amp combos don't mangle it. My recent decision to buy a set of K+H 0300s was clinched by their performance on a small (but loud) vocal ensemble of 12 voices, singing Balkan style. I like them so much, I have developed a taste for more of the same and may move up to the 0410s.

I think of the effect of the sum, as you say, adding up in phase as equivalent to the "rogue wave" in the ocean which, it seems is nothing more than the occasional arrival at the same point in time and space of a large number of wave crests, resulting in monster peaks of as much as 60 feet. I think in a strictly analogous way, there are the occasional "rogue transients" of surprisingly high level in certain kinds of musical sounds - choir and piano being two that come to mind.
Choir still in some ways is the toughest challenge - partly because even the real thing, heard acoustically, can sound like it is distorting! I often hear a shredding/tearing type distortion with multiple loud voices on the loudest sections.
I once recorded a 1200 voice choir that was so big that they occupied the seating area including the balcony in a theatre and I recorded them from the stage. That had a sound all its own.
Old 28th January 2009 | Show parent
  #39
Gear Maniac
 
tonmeister's Avatar
 
1 Review written
🎧 15 years
good thread!
Quote:
Originally Posted by Superdisc ➑️
1. One dB represents approximately the smallest volume change which can be heard if listened for carefully.
When fine tuning mixes I find 0.5dB changes to be noticeable enough to be useable. (thinking of vocals here).
We have some pretty fancy speakers now compared with when they came up with this stuff I suppose.
Maybe it's quite different listening for level changes to:
a. a solo instrument/ tone/ complete mix
b. an instrument within a simple 8 track recording
c. an instrument within a densely layered multitrack mix
Old 28th January 2009 | Show parent
  #40
Gear Maniac
 
chopstickkk's Avatar
 
🎧 10 years
Quote:
Originally Posted by Dale ➑️
but the discussion is about human perception,
unfortunately wiki is not the definitive source for points of bickering!
damned if I know where to steer you to though.
who's bickering?
I was merely making the point that discussing perception of volume levels is merely to allow oneself to wallow in subjectivity rather than objectivity, as we've all perceived 10db of difference and we've all perceived it differently. Yes I was dismissing the thread and trying to steer it back to ludwig. But it's become an interesting historical sidenote now so i'll read on!
Old 28th January 2009 | Show parent
  #41
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by fader8 ➑️
I quoted what I deemed to be your premise, in that if you have 4 identical, correlated sources, that the addition of their power would equal 12 dB.
Ah, no, I wrote SPL not power.

Quote:
If these were signal paths feeding a mix bus, then that's true. The voltage gain would be 4 and and you'd see a 12dB increase on your bus meter. 20Log.
Still true in the acoustic world.

Quote:
But 4 identical acoustic sources add powers, which uses 10Log, and therefore gives a 6dB increase when you quadruple the power.
I think the mistake you do here is assuming some power level. When for example you add several loudpseakers you also add system efficiency (more power out for the same power in) and what is the summed power of the right arms of the violinists? ;-)

Quote:
You're absolutely right that in real world violin section conditions, such correlation isn't possible. But, for example, if you're stacking horns in a big concert sound system, it's very real and you'll come very close to these numbers.
From high freqeuncies where the distance between the sources is big in relation to the wavelength, yes. For low frequencies where the distance between the sources is short i relation to the wavelength, nope.

So, I think the math that I outlined is correct actually.


/Peter
Old 28th January 2009 | Show parent
  #42
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
Quote:
Originally Posted by Audiop ➑️
Posted by fader8:
I quoted what I deemed to be your premise, in that if you have 4 identical, correlated sources, that the addition of their power would equal 12 dB.

Posted by Audiop:
Ah, no, I wrote SPL not power.

Posted by fader8:
If these were signal paths feeding a mix bus, then that's true. The voltage gain would be 4 and and you'd see a 12dB increase on your bus meter. 20Log.

Posted by Audiop:
Still true in the acoustic world.
OK, this is one of the most common misconceptions in audio engineering and with all due respect to the many fine authors who write about this subject, this aspect remains one that is often under-explained.

Yes, pressure and voltage are fundamentally equivalent mathematically if we're discussing electricity. However in acoustics, to double the "pressure" wave it takes four times as much power, not twice as much. Power is proportional to the square of the pressure, so if we double the pressure, power quadruples. Triple the pressure and we increase the power nine times, etc.

I know this is a bit tough to grok because everyone is taught in audio 101 that if you double the amplifier power, you net a 3dB increase and if you double the voltage you net a 6dB increase. This is simple as we're talking about electrons confined in a wire, like water in a pipe. But it's a different thing if we're talking about a wavefront that is expanding outward from a source.

Two equal and frequency correlated sound pressure levels will sum in air with a net increase of 3dB and four will sum to 6dB.
Old 28th January 2009 | Show parent
  #43
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by fader8 ➑️

Two equal and frequency correlated sound pressure levels will sum in air with a net increase of 3dB and four will sum to 6dB.
Hi!

I am aware of the things you bring up (the correct parts :-) but what I wrote is still correct and I think you are missing some things.

What you write above is not correct. Two correlated sources with equal sound pressure levels sum to +6dB SPL.


/Peter
Old 28th January 2009 | Show parent
  #44
Lives for gear
 
wado1942's Avatar
 
🎧 10 years
Quote:
I often hear a shredding/tearing type distortion with multiple loud voices on the loudest sections.
What's really freaky is when you have 2 people singing different notes and you hear a 3rd note come from the rafters. Intermodulation exists every bit as much in the acoustic world as it does the electronic world.



Quote:
what song is averaging -3.5dBFS
Angels & Airwaves "I-Empire" gets up there. Iggy Pop has a record that often gets around -4dB. Metallica's new record gets up there too.
Old 28th January 2009 | Show parent
  #45
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by wado1942 ➑️
What's really freaky is when you have 2 people singing different notes and you hear a 3rd note come from the rafters. Intermodulation exists every bit as much in the acoustic world as it does the electronic world.
While it is clear that any two sounds modulate the waveform there is a difference between a new spectral component being created (as when two signals is mixed via a nonlinearity) and beating.

We hear beating but also spectral components can be created (and are)
in the ear.

There is no significant nonlinearity of the air that produce sum and difference tones or side bands (IMD) acoustically from audio frequencies.

If there is any work done on this that I'm unaware of, of course I would bee glad to learn more though.


/Peter
Old 28th January 2009 | Show parent
  #46
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
Quote:
Originally Posted by Audiop ➑️
I am aware of the things you bring up (the correct parts :-) but what I wrote is still correct and I think you are missing some things.

What you write above is not correct. Two correlated sources with equal sound pressure levels sum to +6dB SPL.
Peter, with all due respect, and I mean that, I do understand and agree with your other statements, but this aspect is not correct. Before you respond, I would implore you to research this yourself. Although as I said, few, even authoritative, sites bother to explain this distinction or do so clearly.

But you don't have to take my word for it. Take a couple of monitors out on the lawn and set up an SPL meter. Set up each one for equal level and then measure the result with both active with the exact same signal.
Old 28th January 2009 | Show parent
  #47
Lives for gear
 
wado1942's Avatar
 
🎧 10 years
No, you're right about that. However, I was stating that constructive/destructive interference can create the ILLUSION that more tones are present than there really are.
Old 28th January 2009 | Show parent
  #48
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by wado1942 ➑️
No, you're right about that. However, I was stating that constructive/destructive interference can create the ILLUSION that more tones are present than there really are.
Absolutely!


/Peter
Old 28th January 2009 | Show parent
  #49
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by Audiop
I am aware of the things you bring up (the correct parts :-) but what I wrote is still correct and I think you are missing some things.

What you write above is not correct. Two correlated sources with equal sound pressure levels sum to +6dB SPL.
Quote:
Originally Posted by fader8 ➑️
Peter, with all due respect, and I mean that, I do understand and agree with your other statements, but this aspect is not correct. Before you respond, I would implore you to research this yourself. Although as I said, few, even authoritative, sites bother to explain this distinction or do so clearly.

But you don't have to take my word for it. Take a couple of monitors out on the lawn and set up an SPL meter. Set up each one for equal level and then measure the result with both active with the exact same signal.

With all due respect fader8, but I must ask you to check this up. :-)

I don't have an SPL meter but I do have an Earthworks mic and sophisticated measurement software I use when I design speakers and measure room acoustics. If you add one source in phase and so on as outlined earlier you will see 6dB gain.

If you measure a small speaker in its low range in free air, high up in the air, and then place it on the ground you will see a 6dB gain.

If you use a fourth order Linkwitz-Riley crossover (acoustic slopes, not electrical) in a two-way loudspeaker the summed response will be 0dB when both legs are down 6dB at Fc.

When placing a source at the ground you go from 4pi to 2pi steradian radiation and adding a second source in free air gives the same results.

A corner placement of a subwoofer gives 18dB gain as compared to mounted in free air.

I think you are mixing up power with intensity my friend.


/Peter
Old 28th January 2009 | Show parent
  #50
kjg
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by fader8 ➑️

But you don't have to take my word for it. Take a couple of monitors out on the lawn and set up an SPL meter. Set up each one for equal level and then measure the result with both active with the exact same signal.
If I recall correctly they would indeed sum to 3dBSPL more, not 6.
Old 28th January 2009 | Show parent
  #51
Lives for gear
 
🎧 10 years
You got it all backwards fellows!


/Peter
Old 28th January 2009 | Show parent
  #52
Lives for gear
 
🎧 10 years
http://www.sengpielaudio.com/calculator-soundlevel.htm

I think Sengpiel suffice as an authorative source.

Type 100Pa in the middle box (sound pressure) at the top of the side and press calculate.

Then type 200Pa and press calculate and see what you get.

Two sources will pump the double amount of air for an increase of 6dB SPL.


/Peter
Old 28th January 2009 | Show parent
  #53
Gear Addict
 
Dale's Avatar
 
🎧 15 years
Quote:
Originally Posted by chopstic*** ➑️
who's bickering?
I was merely making the point that discussing perception of volume levels is merely to allow oneself to wallow in subjectivity rather than objectivity, as we've all perceived 10db of difference and we've all perceived it differently. Yes I was dismissing the thread and trying to steer it back to ludwig. But it's become an interesting historical sidenote now so i'll read on!
so sorry, I did not mean to make your comment the blunt of my "bickering" comment.
the debate on the facts and the history of the facts
were quoting wikipedia as the source for this "debate".
it seems to me that they were pissing on the fire hydrant to prove who's was biggest!
Old 28th January 2009 | Show parent
  #54
Lives for gear
 
🎧 10 years
If you have a phase relationship of 90 degrees between two sources equal in level you will get a 3dB increase in SPL.

This is the typicall case on axis with a speaker using first order butterworth slopes. The lowpass and highpass leg will be 3dB down at Fc and 90 degrees apart.


/Peter
Old 28th January 2009 | Show parent
  #55
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
Quote:
Originally Posted by Audiop ➑️
I think you are mixing up power with intensity my friend.
That would be difficult since Intensity is a measure of power, ie watts per unit area, but that aside . . .

I'll agree to disagree with my esteemed colleague here. But suffice it to say that it would be impossible to convince me of something that goes against what I consider to be one of the basic tenets of sound system engineering. But hey, I can still keep an open mind. What's left of it, anyway!
Old 28th January 2009 | Show parent
  #56
Lives for gear
 
MAzevedo's Avatar
 
🎧 10 years
Well, sound power is energy output by a source in all directions.

Sound intensity is the energy passing through an area.

Sound pressure is the physical force applied over an area.

Sound power (dBSWL) and sound intensity (dBSIL) are both 10log functions, and double every 3dB.

Sound pressure (dBSPL) and voltage (dBV) are both 20log functions, and double every 6dB.
Old 28th January 2009 | Show parent
  #57
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by fader8 ➑️
That would be difficult since Intensity is a measure of power, ie watts per unit area, but that aside . . .

I'll agree to disagree with my esteemed colleague here. But suffice it to say that it would be impossible to convince me of something that goes against what I consider to be one of the basic tenets of sound system engineering. But hey, I can still keep an open mind. What's left of it, anyway!
Yea I meant that I thougth that you mixed up power and intensity with sound pressure level, my bad, I'm not usually thinking in terms of acoustic power since it makes little sense in every day audio to me.

You better have an open mind becasue I'm 100% sure you have it wrong.
I can dig out zillions of referenses that suport what I say and only a couple of times during the years have I stumbled across the erroneus idea that a 3dB SPL increase for correlated sources or the support of move from 4pi to 2pi steradianes.

I'm constantly discussing audio with some of the foremost experts in my country and the 6dB thing we argue about here is "first week in school" material.


/Peter
Old 28th January 2009 | Show parent
  #58
kjg
Lives for gear
 
🎧 10 years
So to recap, SPL doubles every 6 dB. Nothing new there (20 log)

and furthermore, Peter, you could probably also subscribe to the hypothesis that the "lawn test setup" as suggested by fader8 (a doubling of acoustic power), would result in measured SPL increase of approximately 3dB, right?
Or am I not getting what this is about?

Quote:
Originally Posted by fader8 ➑️
Take a couple of monitors out on the lawn and set up an SPL meter. Set up each one for equal level and then measure the result with both active with the exact same signal.
Thank you!

regards,
Klaas-Jan Govaart
Old 28th January 2009 | Show parent
  #59
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by kjg ➑️
So to recap, SPL doubles every 6 dB. Nothing new there (20 log)
Yes and adding two identical sources will result in the same +6dB SPL.

Quote:
and furthermore, Peter, you could probably also subscribe to the hypothesis that the "lawn test setup" as suggested by fader8 (a doubling of acoustic power), would result in measured SPL increase of approximately 3dB, right?
Or am I not getting what this is about?
No for heavens, that's what I'm objecting to. heh

If you take two identical sources, say speakers, and add them as proposed you WILL get +6dB SPL and also quadrupled (sp?) power.

If you have one speaker and double the power fed to the voice coil (assuming a hypothetical driver witout thermal compression and mechanical losses) THEN you would see double acoustic power and +3dB SPL.

But if you add another speaker connected to the same amp (assuming a perfect voltage source) the current draw will double and delivered power also doubles BUT acoustic power will quadruple and SPL will be +6dB.

In this last example we see a +3dB efficiency increase in form of one extra motor or Bl factor which explains these extra 3dB for free.

This is really basic physics that anyone involved with audio engineering should understand IMO.


/Peter
Old 28th January 2009 | Show parent
  #60
kjg
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by Audiop ➑️
This is really basic physics that anyone involved with audio engineering should understand IMO.
yes, well, there is a lot of stuff that is relevant to creatives in music
if only everyone would make good music once they remember/understand this basic physics... that would be the day.

i recall now, uncorrelated signals would add approximately 3 dB in the suggested setup, not 100% correlated as you were discussing here.

thank you.
πŸ“ Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 79 views: 33212
Avatar for rids
rids 1st July 2009
replies: 65 views: 9840
Avatar for RonaldDumsfeld
RonaldDumsfeld 26th March 2010
replies: 428 views: 13351
Avatar for grumphh
grumphh 1st October 2016
replies: 374 views: 66414
Avatar for JfromRVA
JfromRVA 13th June 2021
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump