Midband dip, presence dip, Gundry dip .... or not?
I have said before and here again for clarity - the dip you mention does not exist in a current Harbeth speaker for these reasons ....
Originally Posted by sglim
1. The dip you allude to is one of more dB's and was never applied to the midband as you suggest. The 'BBC dip', much misunderstood, was applied to the presence region - much higher up the audio band. If the midband was depressed, you would comment that the sound was thin and weightless: cello would sound like a violin. A depressed midband is an entirely different phenomena to pushing the presence (the immediacy) backwards by adjusting the loudness of the presence region.
2. The presence dip was introduced in old, long obsolete, second generation BBC speakers by their BBC designers to mask colouration - or so Dudley Harwood told me when I specifically quizzed him about this. It had a secondary (unexpected?) consequence of pushing the image away from the listener. Since we don't have any colouration in the presence region of a RADIAL driver we don't need to deceive the ear by fiddling with the energy levels in critical parts of the audio spectrum.
3. No Harbeth designed or manufactured by me in the past 21 years has deliberately featured a dip in the midband region. Such a speaker simply would not sell these days.
4. It is true that the ear (or at least my ear) is hypersensitive in the presence/low treble region. Extreme care has to be applied in that region during design to bring the speaker's bass-mid-top into a contiguous, seamless harmony. Fractions of a dB will make a difference to ones perception. In the crossover region, more midband and less tweeter sounds completely different to less midband and more tweet even though, measurably, they add to a flat line response at/around crossover frequency. 80% of the design effort is resolving this area.
Hope this helps.
Alan A. Shaw
Harbeth Audio UK