power the center speaker. The volume level for these three speaker was about -12 to -20 Db relative to the two center-flanking speakers.
    The effect was to create a wide expansive sound stage that made a lot of big orchestral and psychedelic music, Mahler and Pink Floyd, come alive. (Drug users loved the sound.) Turning off the extra speakers resulted in the most profoundly sad collapsing of the sound field. An added benefit was the perceived magnifying of the power output of the amplifiers. The downside was the unnatural sonic perspective for small jazz ensemble and single voices and instruments.
    If I were to retry this experiment today, I would apply a low-pass (1 kHz) filter to the outside speakers and a high-pass (1 kHz) filter to the center speaker.


Subject:
An Alternative Look at RIAA EQ
The march 2000 issue of the Tube CAD Journal inspired me to write to You about an idea I have been working on for some time.
   The insertion loss of an (anti) RIAA circuit have probably bothered many designers of phono stages.
  It would be interesting if there existed a method to at least partly eliminate this loss of gain.
  My idea is to design a summing filter instead of the more common attenuating filter. To do that we must first look at the general transfer function H(s) for an RIAA filter:

   H(s) ~ 1/(1+sT3) * (1+sT2)/(1+sT1)

where T1 = 3180us, T2 = 318us, T3 = 75us by substituting s with j*2*pi*f it is possible to calculate amplitude and phase as a function of frequency.
  Now observing that T2 = T1/10 we can rewrite the equation to:

   H(s) ~ 1/(1+sT3) * (1+(9/(1+sT1)))

   This equation describes a summing filter and a practical phono amplifier could consist of a linear input amplifier followed by the 2122 Hz low pass pole. Then the signal would split into two. One goes into a amplifier with exactly nine times gain followed by the 50 Hz pole. The bass boosted signal would then be combined again with the second (unprocessed) signal in an adding circuit.
 
-Bengt

    Believe or not, and I do have witnesses, in the mid-80s, I built a hybrid preamp somewhat like what you describe. It used one 6DJ8 and an AD712. A single 6DJ8  triode stage handled all the frequencies above 500 Hz and one half of the AD712 cascaded into the inverting other half, yielding a gain 10 times (a total of about 280) that of the 6DJ8's, at least, I remember it being 10 times greater. Both stages fed a 6DJ8 based mixer, which set the 2100 Hz pole. It worked fairly well, but it was unfortunately cannibalized for the next preamp idea I had, the phono stage qua bi-amp crossover. Since I did not own a computer then, I do not have a schematic on my hard drive. Shown below is my memory of what I did. Maybe by next issue, I will be able to find the paper copy.