How much difference is 30Hz from 29Hz?

[Coytee]

I understand the relationship of having to double your power to get a noticable gain in loudness... BUT, how discernable is it (presuming all else equal) if you have two identical speakers side by side, one having 110db output with a -3db point at say 30Hz, and the other with 110db at 29Hz, 28Hz...or 20Hz?

in other words, having a -3db point at say 45Hz, and the other one at 10Hz, would be substantially noticable. How sensative is the average person from 1Hz to another?

How "off" can they be from one to another such that the average Joe, couldn't tell the difference?

[sfogg]

Richard,

That isn't something that can be easily answered really.

First off manufacturers specs may vary quite widely from reality.

But lets assume for a moment the two specs are accurate. The 1hz difference alone is no big deal and would quite likely be inaudible on music. What might trip you up however is even though the -3dB point may be 1hz off the -10dB points on the speakers could vary by quite a bit. If on is a sealed speaker and the other is vented for example the sealed system will have a lower -10dB point since it will roll off slower then the vented unit.

Also there could be a very large difference in the amount of distortion each speaker is producing at that level and frequency. That would be very audible as well.

Shawn

[jdm56]

If truly all else was equal, the difference would be practically non-existant because there is such little music with fundamentals below even 40hz. (btw, app. 40hz is the lowest tone on a 4-string electric bass.)

A doubling of power will yield a 3db loudness increase, while 1db is considered to be about the least detectible change, if I am not mistaken.

[Coytee]

Although I realize it might not be answered easily, the reason I'm oddball enough to even ask is, I have a pair of LaScallas that I think we all/most know have a nice solid punch. I also have a pair of EV's that go (per spec sheet) to 28Hz @ 106db.

I don't know what specs are right blah blah blah...but the real world feeling is, the LaScallas will punch you in the gut with their bass output, whereas, the EV's will literally shake the foundation. There is a substantial difference between the two.

I was just wondering how easy it is to tell (the difference) when the specs (accurate or not) are closer than these two are.

[lynnm]

All else being equal,the difference would be inaudible unless one had a virtually perfect listening room ( and perfect hearing ). Given that there is no such thing as a "perfect listening room and perfect hearing ability": Any perceptible audible difference would be a function of the speaker's performance regarding the midrange and/or tweeter response.

In any case a difference of 1db. is as far as I know undetectable except with better instruments than what is attached to our heads.

In the final analysis What sounds best to any given listener is the best.

[sfogg]

"In any case a difference of 1db. is as far as I know undetectable"

Depends on how wide the frequency range the 1db difference covers. For wide range sounds human hearing can detect down to about 0.3dB. The smaller the frequency range of the level difference the larger it needs to be to be detectable.

1dB over the entire frequency range is *easily* detectable in double blind ABX listening tests. I've scored 25 of 25 on such tests.

Shawn

[lynnm]

Let's discuss this like mature adults Mister Poopy-Pants

No duel invite intended here. I accept that my opinion based as it is on mere conjecture may well be incorrect.

[WMcD]

I don't have my musical scale versus frequency right here. I did post it a long time ago.

As far as "can you hear a difference of 29 versus 30 Hz."

I'm quite sure not, as far as frequency response as a whole.

Remember that if you're using 30 Hz as a reference, the octave above is a 60 Hz and the one below is at 15 Hz. You can put 16 half notes in either of those. So going down one Hertz there is roughly a half note on the musical scale.

Now if you shift back and forth between the two, you should be able to discern a change of a half note. But it is like having a piano with one extra key at the bottom.

There are lots of other things. Roy Heyser in his review of the K-Horn noted the same sort of thing as you have. A box type speaker can shake the house. But listen to a bass horn and it sounds natural.

Another issue is that the 3 dB down point really doesn't tell us what is going on farther below. Horns unload rapidly below cut off. Vented speakers do so to a lesser extent. Sealed boxes even less. So you may see some perceptable bass below the 3 dB point.

Another odd thing is that a peak at 120 Hz sounds like strong bass more than a peak at 60 Hz. The EV "Bible" says if you want to hike up perceived bass, go for 120 Hz. Some of it is that our ears are not very sensitive to low bass. Hence you'd really have to goose it.

I agree that the low E on the bass guitar, and probably the double bass violin is at 41 Hz. On the other hand, a very big drum (the marching band type), sound effects, pipe organs, do go lower.

Be careful. I killed a very high power EV woofer by hitting at the system resonance with a small amp and a Telarc CD (one with warnings). I think that also accounts for some home system failures.

Gil

[sunnysal]

this should help..

tony

[sunnysal]

and another (perhaps better for this discussion)and don't forget to visits art's pages, especially the one on the ear and hearing

http://www.silcom.com/~aludwig/

tony

[WMcD]

Thank you so very much SunnySal. This was exactly what I was talking about. A better diagram than I had in mind.

Wow, I was right about something.

We see that A4 is at 440 Hz. That is the modern tuning standard. And you can see the E which is the open string on the bass guitar at 41 Hz.

Look at the notes at the very bottom. You are extending the range by about half a note at the bottom by shifting from 30 to 29 Hz.

Can you hear that shift. I'd think so, if you play one note on the keyboard and then the other. Violin players who have to hold a finger on the string have to be able to hear a quarter note.

Still, as far as frequency response from a speaker, you are talking about a piano with just one more key.

Gil

[John Warren]

Above Fs, cone excursion is inversely proportional to the square of the frequency for a fixed SPL.

So the differecence betwwen 30 and 29Hz is 30²/29² = 1.07. In other words, to reproduce a 29Hz fequency at the same SPL as a 30Hz frequency, the cone has to move 1.07X more, not negligible.

The same analysis comparing 25 and 30Hz results in a 1.44X factor.

[Daddy Dee]

Sunnysal,

Thanks for posting these resources on instrument and voice ranges. Most helpful for my "wonderings" about a DIY speaker project.

[colterphoto1]

How much difference between 29 and 30 Hz?

Not enough to care about. Remember these are only specs. Everything changes in your listening room. Even if the -3dB down point is ever so slightly different, what also matters is the smoothness of the response curve, the response in YOUR room, and most importantly, how YOUR ears perceive the sounds coming out of the speaker system.

Let's just enjoy the music people. (my .02)

Michael

[sunnysal]

I wish I could find a chart that showed cymbals and other percussion instruments, these only show tympanies. if anyone finds one please post it here, warm regards, tony

[D-MAN]

My understanding is that it takes a change of +/- 3db to be noticable by the human ear. A single change of 1 db is reportedly not going to be noticably louder than the original value.

Electrically speaking the change is logarythmic; amplifier has to increase wattage by a factor of 10 to achieve a 3db change, for instance 1 watt +3db = 10 watts. 10 watts +3db change = 100watts. 100watts +3db = 1000 watts, etc.

So a +/- 1 db change is relatively meaningless to the human ear as far as detecting it, but makes a demand on the electronics involved.

DM

[LarryC]

----------------

On 8/2/2004 9:45:46 AM sunnysal wrote:

I wish I could find a chart that showed cymbals and other percussion instruments, these only show tympanies. if anyone finds one please post it here, warm regards, tony

----------------

Tony, some percussion instruments are "pitched," i.e., have finite pitches like typmani, while some, like cymbals and triangles, are not. The latter do not have identifiable fundamental tones. Some of them have very wide frequency ranges and are made up of a great number of different, dissonant frequencies, a sort of an aggregate clang. I suspect that this extremely complex mish-mosh of a huge number of tones is the reason that triangles and, especially, cymbals, are among the most difficult instruments to reproduce cleanly, and for some an ultimate test of their equipment. Anyway, that's probably why you don't see them on charts like the ones above.

FWIW, the difference between 29 and 30 is less than a half step. The ratio of frequencies that are a half-step apart is 15/16; this would amount to a 2 Hz difference at around 30 Hz, and a single Hz would be a QUARTER step. That might be hard to hear -- it's certainly outside our usual musical experience. A half step at 16 Hz is only 1 Hz, but is 4 Hz at 64, etc.

Larry

[sfogg]

"My understanding is that it takes a change of +/- 3db to be noticable by the human ear. A single change of 1 db is reportedly not going to be noticably louder than the original value."

I've done 1dB tests myself and passed them. I'm not even remotely alone in this.

There are AES papers talking about double blind listening that spec human hearing as being able to hear down to about 0.3dB differences in SPL if it is over a wide range. If the SPL change is in a narrower range it needs to be a larger difference to hear it.

http://www.pcavtech.com/abx/abx_crit.htm

As far as a 0.3dB difference over a wide range you won't perceive it as a volume difference, you will hear it as a difference in clarity, focus..etc..etc... but it is perceivable and identifiable. This is exactly why it is very important to level match to components to 0.1dB when doing a comparison between them. Otherwise SPL differences, that aren't preceived as SPL differences, will influence your perception of the sound quality. Very very few people do this though.

If you don't believe me that 1dB is easily audible try it yourself.

Download the PC ABX program from here:

http://pcabx.com/

And using whatever music or material your want rip it to your computer. Make a duplicate of that file and then use a digital audio editing program to lower the level of one of the files by 1dB. Load each file into the ABX program and take the test. I've done this and scored 25 out of 25 correct.

Shawn

[ygmn]

I always thought 1 dB is the average difference (for humans) in SPL that a human can distinguish....??

3db is the average difference in SPL that appears to be a double in the loudness....??

[Ray Garrison]

These numbers are largely apocraphyl.

Common wisdom (?) has it that a 1dB difference is "barely noticeable", a 3dB difference is "clearly" or "easily" detectable, and a 10dB difference is a subjective "twice as loud".

However, this is pretty much garbage.

What frequency? How wide ranging is the difference? If you take pink noise from two sources, and bump one of them in a two or three octave range in the midrange where the ear is most sensitive, most folks can pick up a 0.1dB difference in blind testing. If you take a signal that's restricted to the deepest bass, and apply a very narrow band boost in the deepest frequencies, it takes a *LOT* more difference before you can pick it up. Several dB, maybe more.