Freq Response at Rock Concerts

[Dan McDonald]

I know this would vary considerably, but I'm involved in a discussion of this issue elsewhere and thought we might have some people here with experience who could shed a little light.

What would you guess, frequency response would you expect with a typical "national act" rock concert where the band brings its own sound system, etc.?

Someone made a recording and the freq response was up to about 15kHz; I thought that might be pretty close to being what the sound system actually put out, but I didn't know. Others have said they get frequencies around 20kHz from audience recordings, but I was tempted to suggest that may be noise output rather than actual music that they are picking up.

Any thoughts?

Dan

[DougMac]

Re: Freq Response at Rock Concerts

I can't resist. As long as it's up to 10k, what difference does it make? If you go to rock concerts, you probably can't hear freqs much above that anyway ;-). Forgive my levity.

On a more serious note, I would imagine there may be high frequencies picked up by the mike, but I doubt it's coming from the sound reinforcement system. If somehow you're picking up the cymbals directly, I could believe there's some high freq info out there.

[Dan McDonald]

Re: Freq Response at Rock Concerts

Thanks.

Craig - when you say 30+ feet from the loudspeakers, is that just a guess or is there some sort of formula that can be used to calculate this kind of information?

It would be very interesting to me to figure out how much of a signal is noise coming from nearby audience members, etc. and how much is music from the speakers. There's really no way to separate that out, as far as I can tell, unless you can compare an audience recording with a soundboard recording, but find some way to calculate distance from speakers and how that plays into it.

Dan

[Craig Maier]

Re: Freq Response at Rock Concerts

The figure that I gave of 8 KHz at 30 feet is based more on experience than mathematics. However, the response could be calculated based on acoustical modeling techniques. There are too many variables to perform such a calculation based on the limited information that we have on any specific rock concernt environment. The high frequency response generally drops off very quickly with distance no matter how good the loudspeakers or sound re-inforcement equipment happens to be. I hope that this helps.

[Dan McDonald]

Re: Freq Response at Rock Concerts

Thanks, Craig. That helps a lot.

Dan

[Craig Maier]

Re: Freq Response at Rock Concerts

What is actually being reproduced by the loudspeakers and measured 30 or more feet from same would have no frequencies greater than about 8 KHz. If the recording was make in the audience, the frequencies above that value may be local audience noise which will have higher freq content. Also, tape hiss will be behond 8 KHz.

[Rick]

Re: Freq Response at Rock Concerts

I found this site that has a bunch of calculators for acoustics things. It has one for the frequency dependent absorption of air. It shows an attenuation of about 5dB/100ft at 10kHz.
Of course this does not take into account many other factors such as the audience, which are great absorbers of high frequencies. The off axis response of the high frequency drivers if worse the higher you get in frequency.

Another thing to consider is that even if the sound system is 10dB down at 10kHz, relative to the midband response, and the band is pumping out 110dB SPL thats still 100dB of 10kHz.

The site is: http://www.doctorproaudio.com/doctor...empo-distancia

Rick

[Craig Maier]

Re: Freq Response at Rock Concerts

I think what Rick is trying to say is that the corner frequency (-3dB point) would be at around 10 KHz at something less than his stated distance. Correct me if I am wrong, Rick.

[Dan McDonald]

Re: Freq Response at Rock Concerts

Thanks, Rick. This is interesting, although it makes me wonder more about how all of this plays out.

At first I was thinking that if a band is producing 110 dB of sound, wouldn't that be the total spl, which, if you divided it up into frequencies, mean that you would get maybe 40dB of bass, 40 dB of midrange and 30dB of high end?

But then that I thought that can't be right, because it's on the log scale so it isn't really additive, in the sense of being able to partial it all out, is it? I guess I mean that if the band is putting out 110 dB, and all of a sudden all of the amplification associated with the bass and midrange was cut out, you'd still have a pretty loud high-end sound - not anywhere near 110dB, but certainly louder than 40dB.

I guess you'd need to work with a different measure of spl and then partial it out for the various frequencies, then convert back to dB to see what the effect is.

Maybe this is too specific a question for the DCArt website, but understanding how this plays out also helps understand what to do in restoring audio as well.

Dan

[Rick]

Re: Freq Response at Rock Concerts

Yes, SPL is the sum of all the frequencies, but music has a far from flat distribution. Most of the energy comes from low frequencies, so if you were to cut out everything but the highs you would be left with something less, maybe 80dB.

If you take a typical pop song, the high frequencies say at 10kHz, may be down 30dB from the low to midband at say at 100Hz. You can see this for yourself, just put up the spectrum analyzer on DCart and put it in averaging. Do this for an entire song and you will see the typical distribution for this song.

Back to dB, Here is an example. Say the total SPL is 110dB, now SPL is power so the formula is 10*log(p1/p2), where p1 and p2 are to relative powers. We dont have to worry about absolute levels here because its all ratios.

Lets say that this music is made up of three tones, all at the same level (very boring music). If you take away 2 of the tones, leaving one tone at its original power, you would still have 105dB SPL just from that single tone. A rule of thumb is that doubling of power give you 3dB more output, changing the power by a factor of 10 gives you 10dB more output.


Rick

[Dan McDonald]

Re: Freq Response at Rock Concerts

Excellent.

Thanks, Craig and Rick.

Dan

[Dan McDonald]

Re: Freq Response at Rock Concerts

I did some analysis of some recordings today and found there was some material up to about 15-18kHz - It appeared to be sibilance from the vocalist, cymbals and cowbell, followed by crowd noises - whistles, etc.

But most of the music was gone by 13kHz. These were recordings done with a DAT machine and some very nice microphones.

I used a combination of low pass and high pass filters to see where the sound actually cut out.
And the Spectogram to see which sounds were producing the high frequencies.

Dan