Patch Notes: “The Gospel of the Comb Filter”

Let’s take a look at the patch behind my latest track, “The Gospel of the Comb Filter,” from my album Metaphysical Shitposting. This track started with the desire and intent to process some speech using some of the vast quantities of crap I have accumulated over the years. That means I started with this weird little box you see at the bottom right of the photo — the Panasonic RR-830 Cassette Transcriber/Recorder.

I picked this up a few years ago — got it for free, actually, because the woman selling it couldn’t figure out how to operate it, so she just gave it to me since we both thought it was broken. Turns out it was a simple operator error and it’s mostly operational (the auto backspace function seems broken, but whatever) but I haven’t done shit with it except use it as part of the photo illustration on the cover of my Psychedelic Ghost Stories album.

Similarly, I have stacks of weird Christian apocalyptica, would-be prophecy and general lunacy on cassette tape (I grew up in an environment steeped in the stuff, and tho I have long since broken away the crazy fringe of it remains fascinating to me to this day) and I decided it was damn well time I did something with some of it. I picked out a tape on signs of the apocalypse from Bible history, popped it into the RR-830 and I was off!

I knew I wanted to play with some different ways of processing the speech in modular so the first thing I did is run the output of the transcriber into Ears to get it up to a level that the modular could deal with. Alas there was some kind of impedance mismatch going on, so I also got a pretty nasty and prevalent buzz as a bonus, but this is modular, and we can deal with any problem (even if we create three more in the process). A quick pass thru the Doepfer SEM filter cleaned up the buzz without fucking up the voice, and we were off.

I multed both outputs of the SEM filter, with two copies coming out of the lowpass output; one going to the QPAS and one straight to my mixer, the Planar 2 (more on this later). The bandpass output fed the Rossum Panharmonium and the FX Aid XL, which was running a comb filter program (hence the track title). The outputs of all of these modules, plus the direct output from the SEM filter I mentioned earlier, fed the Planar 2’s four VCA inputs, allowing me to control the relative volumes via the joystick. Cool!

The four direct outputs feed into the Worng Sound Stage at different points, mostly concentrated on the lower half inputs — this is voice from a tape, so it’s nearly all midrange, after all. In effect, this allowed me to move the joystick around to feed differing amounts of each affected vocal to the Sound Stage, generating a weird, dynamically shifting soundscape of lightly to heavily altered vocals that moved around the stereo field.

The Sound Stage feeds the Mimeophon (delay effects), which in turn goes to the Desmodus Versio (reverb) and then to your ears. Pretty simple really — just a bit of parallel processing controlled with a dynamic mix, but pretty effective!

Modulation wise, I had a Make Noise Function cycling to provide modulation for the QPAS cutoff and the repeats of the Mimeophon. The Sloths up top is modulating the feedback and frequency of the comb filter effect as well as the CV control of the filters in the Sound Stage (keeps that stereo field lively, in a subtle way). Finally, the Lightstrip controller was patched to the Zone control of Mimeophon, so I could glide between ultrashort, chorus-type delay effects and longer dubby delays.

Performing the patch had a number of elements that kept me busy. The transcriber is controlled via a footpedal and a few sliders on its face. With those I could play, rewind and alter the playback speed and pitch (semi separately, through some weird analog granular effect), and I did all that at different times. The joystick allowed me to choose which elements were prominent in the mix and the Lightstrip gave me realtime control over the delay times. I also hand tweaked a few knobs on the Panharmonium and QPAS here and there as well.

I spent a day practicing all of that and then recorded a take I really liked thru my AKAI cassette deck, and that’s it: that’s the version you hear on the album, minus about 30 seconds I clipped off the front and some minimal post processing (hipass filter to remove any subaudible junk, light compression, normalization).