I’ve made a low-pass filter test of sorts. Very unscientific though.

I ran an old monosynth through three popular modern synths: Arturia MiniBrute, Moog Music Minitaur and Elektron Analog Four.

The sound source is the Roland SH-101 with both waveforms plus the sub-oscillator simultaneous triggered on C1. The gate signal controls the VCA, envelope and modulation are turned off, and key follow knob is set to 100%.

The audio signal is sent to the filter circuits of the different synths. The test pattern is defined as follows:

  1. A “clean” tone feeds into the filter for a few seconds, with frequency cutoff set to 100% and resonance at 0%.
  2. Manual filter cutoff sweep with resonance at 0%.
  3. Manual filter cutoff sweep with resonance at 50%.
  4. Manual filter cutoff sweep with resonance at 100%.

First off is the SH-101 internal 4-pole 24 dB per octave low-pass filter.

Next up is this unmodulated audio signal going into the MiniBrute and processed with the multimode Steiner-Parker filter, set to 12 dB per octave slope in low-pass.

On the higher resonance level, self-oscillation occurs and the volume peaks. (I had to put a limiter at -6 dB threshold to keep the volume at bay). The filter sweeps follow the same order as earlier.

The second external filter in this test is the Moog ladder filter (4-pole 24 dB per octave low-pass) of the Minitaur. Again the filter sweep follows the same order as above. The volume level drops noticeably on the higher resonance settings.

The last filter is the Analog Four’s ladder filter, another 4-pole 24 dB per octave low-pass.

The audio signal pass through the synth’s second filter unaffected (the 2-pole multimode filter is set to 12 dB per octave high-pass with minimum frequency and no resonance).

Maybe I’m doing something wrong here, because the audio signal seems to loose some decibels both in the treble and in the bass. Anyway, the flattest response is obtained when resonance is somewhere around 25, not 0.

On the last filter sweep, the resonance is set to 85 (out of 127), because the self-oscillation is going crazy over 85, and only the filter sounds. And the volume raises dramatically.

That’s the test for what it’s worth, I know the execution leaves a lot to be desired. The input level of the synths differ some, which makes a comparison hard. I should probably mention that the audio signal is running through a Mackie 1202VLZ4 mixer before being recorded using high-grade Cirrus Logic converters. This might have some affection and coloring on the signal.