The Missing Measurements..., Coming to a Magazine Near You..., Possibly

Some people like to play their music loud. https://www.audiokarma.org/forums/st...es/cuernos.gif

So, how would one know how loud a speaker will play, without distorting, if the magazines don't test for this? Well, it just may happen: https://www.audiokarma.org/forums/st...s/scratch2.gif

My letter to Stereophile Magazine:

The Missing Measurements

As a long-time reader of Stereophile, I very much enjoy the speaker reviews, especially in how the measurements translate to the perceived experience of the listener. But I feel that there is something missing in the measurements; that being the maximum volume that the speaker can play too while “holding it all together.”

I see reviews of speakers with great curves (both cabinetry and frequency response), and some not so great, but then look at the compliment of components and think to myself, “There is no way those speakers could play something like a Van Halen guitar solo at concert levels, let alone hold it together when the band kicks in.” (I do have a pair of original Chartwell LS3/5As, and this is their major shortcoming.)

And while Stereophile does publish the SPL at 1 watt input, and the maximum input rating for the speaker is often provided by the manufacturer, one could come to the wrong conclusion as to how loud the speaker can play, again, keeping it together, because of such factors as dynamic compression, which need not be the same for all of the included drivers.

I would suggest the inclusion of a “waterfall” plot to look at frequency response across a wide variety of power levels (e.g., 1, 2, 4, 8, etc. watts) up to some level of distortion, (e.g., 10 percent) illustrating dynamic compression across the frequency band. A simple line plot, similar to that used for amplifier power ratings, could examine the total harmonic distortion as the input level is increased up to this level of distortion.

It’s time to let the “Big Boys” show what they can do relative to what’s out there.

Thanks,

Todd Brody


And the response from John Atkinson:

Thanks very much for writing, Todd. This subject has been on my mind for a while, but is difficult without access to an anechoic chamber to get a proper measurement of dynamic compression. Don Keele sent me an AES paper he wrote in 2018 on how to do this. Though his method is lengthy and time-consuming, it does appear to produce repeatable results. Trying his test is on my to-do test.


Stay well. - John


So we'll see where this goes. https://www.audiokarma.org/forums/st...es/naughty.gif

I'd be very interested to hear of any audio publication hiring an anechoic chamber with the necessary professional backup to perform meaningful measurements. How distorted any transducer sounds varies not only only the signal level, but also the programme material. In my view the best adivice is go and listen to a speaker with your preferred music and preferred listening level and hear for yourself how it sounds. Whatever speakers you own unless your listening environment is accoustically treated chances are your room is contributing as much to the listening experience as the speakers within it.

Hello Todd;

Here is a link to m-noise. If you haven't already, give this a read.

https://m-noise.org/

I have begun using this as a test metric because I also want to quantify sound quality i.e. linearity at higher SPL.

I realize that this is something that you may not be interested in doing yourself but if you do, the distortion characteristics of the measurement mic's become a significant factor. Most real measurement mic's SPL limits are stated at 3% distortion and many reach this well before they flat line at maximum rail voltage.

I ended up with these. http://www.larsondavis.com/contentSt...D0501.0014.pdf and the required power supply and pre-amps.

If 172dB SPL seem ridiculous, realize that you can put a very high quality Earthworks M30 on the rails (≈130dB) with only 350mV on a large format driver on a plane wave tube.

This I believe is exactly your point. ? Quantification at real use levels?


Barry.

Quote:

---

Originally Posted by 1audiohack

Hello Todd;

Here is a link to m-noise. If you haven't already, give this a read.

https://m-noise.org/

I have begun using this as a test metric because I also want to quantify sound quality i.e. linearity at higher SPL.

I realize that this is something that you may not be interested in doing yourself but if you do, the distortion characteristics of the measurement mic's become a significant factor. Most real measurement mic's SPL limits are stated at 3% distortion and many reach this well before they flat line at maximum rail voltage.

I ended up with these. http://www.larsondavis.com/contentSt...D0501.0014.pdf and the required power supply and pre-amps.

If 172dB SPL seem ridiculous, realize that you can put a very high quality Earthworks M30 on the rails (≈130dB) with only 350mV on a large format driver on a plane wave tube.

This I believe is exactly your point. ? Quantification at real use levels?


Barry.

---

Thanks, I will give it a read. I am well familiar with Larson-Davis and have used their sound level meters on several occasions dating back to the early 90s actually purchasing one for the company I worked for. Did the job and much cheaper than the B&K.