Nezbleu,
JimB posted an elegant solution to the "33 nF problem" a while back, here:
http://orion.quicksytes.com/viewtopic.php?f=12&t=1018&start=105I'm sure that the idea would also work for many of the other capacitor woes ailing us these days. In Jim's solution, he changes the 33 nF capacitor to a 30 nF part, of the same high quality, and of which there are plenty in stock at DigiKey. Then, he "re-scaled" the circuit that used this capacitor, by changing resistor values appropriately, and the result is a circuit essentially equivalent, but using readily available parts. The point is that 1% resistors are commonly available, but the fancy 1% or 2% poly caps are becoming less so.
One way to see how this works is to study SL's spreadsheet pz-eql.xls (see the Pluto+ page, in the "Frequency response equalization section"). The magic number is the value of "k". Changing this changes the impedance scaling of the filter stage. Small changes in k result in identical circuit performance, but different values of the resistors and capacitors. Jim did the same thing with some of the Pluto circuit (specifically, the 1000 Hz crossover section). He changed a capacitor value, and then brought everything back into focus by changing the associated resistors.
Steve