Quick question about impedances in frequ response plots

[Rodeodave]

Hi,

In some datasheets (namely Fostex) there is more than one curve in the frequency response plots. I'm not talking about the impedance-curve, the ones with the numbers 4,8 and 16 are puzzling me. Does a deliberate output impedance/speaker impedance mismatch help to gain sensitivity? Is this only possible with tubeamps then (given the OPT has appropriate taps)?

[claudio]

Hi,
let's take a Fostex datasheet for example: http://www.fostexinternational.com/d.../pdf/fx120.pdf

The 3 frequency response curves are taken with the mic on axis, 30° off axis, 60° off axis.
Are these you are questioning?

[Rodeodave]

Oh, that makes sense. Never seen a datasheet with such annotations before. The first time I came across such a plot was http://www.fostexinternational.com/d...df/fe108ez.pdf and it has "4, 8, 16" next to the three lines, which made me think...
Thanks for clearing this up!

But still, I've heard rumors that a deliberate impedance mismatch can have positive efects. How? Why?
Anyone?

[rjbond3rd]

4, 8 and 16 on the right-hand side are ohms, for the impedance plot.