Frequencies are not equal for many reasons in a variety of factors; I'll touch on one, or two.
Different frequencies often have differening levels of transmit power; at the higher frequencies it is easier to fill the Q of the loop because they contain more energy. A lower frequency
like 5kHz (vs 19kHz) requires more battery battery power to raise the energy level, yet they are often set, milder (to conserve reserve). And even a higher frequency (30kHz for example)
is still very sensitive to iron. (some units have a fixed transmit level, others may be variable--check your user's guide) So a sensitivity number across several frequencies will have varying performance.
And for some reason at some point the sensitivity level will false on rejected targets that at a lower setting you do not hear. On a Teknetics Mk-1 @ 6.592kHz I found that an extremely high sensitivity
setting would cause iron to appear that was completely rejected at normal sensitivity. (also recall that a conductor weakens a detector's field, and iron strengthens it,
so iron depending on alloys, shape (BC's), size/halo can have some odd effects--and the field alone suddenly being stronger might cause a frequency shift which is detected in the transmitter and unbalances
the receiver--causing an audio (and these units are not exactly off resonance discriminators!)

Make sure you rebalance the ground when changing frequencies (very important) and experiment with the sensitivity settings. And often rejected targets near a notch edge may come
through at a higher setting)
IF you can vary the transmit setting, a higher setting can often be used to overpower emi, and even allow the sensitivity level set lower if needed--with an increase in depth.)
