How to read a speaker's Cumulative Spectral Decay plot

[Razz]

So this is interesting to me....

Above is a very well liked speaker around here. As you can see it is very well behaved and comes in at under $2K!
Great looking response! I don't think anyone will mind... I't the Paradigm Studio 60 v.5

And below is a far less attractive chart. It seams to look not as controlled right? Especially in the mid-range and a spike around 14.5Hz

Surprisingly, the speaker here costs $60K..... yes $60,000.
It is the Magico Q5. I have heard this a few times at audio shows and was in love with this thing! Such a dynamic and detailed speaker at loud levels. I would take the Magico over the Paradigm in a millisecond!

So I guess I might be back to the drawing board about these grafts as they don't always paint the right picture.
But at least I know how to read them and what they mean. Thanks for the class Flint.

 

[Razz]

...Or is there something I'm missing in reading the 2 charts I posted?

 

[Flint]

Spectral decay is just one measurement. It does not address step response performance, frequency response, dispersion, peak output, thermal compression, power response, or THD at listening levels.

What it does explain is resonances, decay, and measured empirical dynamics (which many people perceive inversely).

 

[malsackj]

These plots can also be done with Room Eq Wizard ??

But this includes the mic and the room acoustics

if this test is done in the anechoic chamber then that would be accurate for the speaker.

Doing this outside ?? not 100 percent and leave some room for error, but might be the best a home user could do.

 

[Flint]

These plots can also be done with Room Eq Wizard ??

But this includes the mic and the room acoustics

if this test is done in the anechoic chamber then that would be accurate for the speaker.

Doing this outside ?? not 100 percent and leave some room for error, but might be the best a home user could do.

I do not think REW will do this. It requires a gated system and a pulse - the pulse is instantanious and the mic has a gate on it to remove all acosutic reflections from it's capture window. I don't beleive REW will support decay response curves. I know TrueRTA cannot do it.

 

[Zing]

:text-link:

 

[Flint]

INteresting. I would love to see a comparison of this tool versus a PrecisionAudio output.

 

[malsackj]

That is why I was looking to try this outside and not have the room acoustics to consider.

It would mean the mic may have to be very close ???? Thinking the 1 foot off the baffle.

 

[Flint]

If the mic is too close, it is not going to measure the way the speaker performs in a proper space. With a multiway speaker, 1M is a the minimum distance for getting a realistic measurement from a speaker.

 

[TitaniumTroy]

Thanks for posting this Flint, especially the electrostatic speakers measurements. So how do True ribbon tweeters fare on this test?