When I drive Bazille's filters hard, the high-pass output lets through lots of bass, no matter where I set the filter's cutoff. The cutoff stops sounding like a filter cutoff at all, and changing the resonance barely changes things either.
This effect occurs on all four of Bazille's filters, and occurs when any loud signal goes into the filter, whether it is loudened with the filter's gain knob or otherwise.
I have noticed that the manual describes the high-pass filters as follows: "12dB per octave (2-pole) highpass with low shelf via Gain." So, it seems that this effect may be intentional. Personally, I would like the option to drive a signal hard while also having a true high-pass filter output. Maybe there are uses for a high-pass filter with a low shelf; this seems plausible. However, to control this shelf using the signal's gain? I don't understand this design choice. It seems to force unwanted bass on those who want a dirty output signal, while forcing unwanted dirt and drive on those who want a bassy high-pass output.
Have you noticed this effect, and if so, what do you make of it? Is there a reason you might want to control a high-pass filter's low shelf using the gain of the input signal?
(As it's my first post, I should add, Bazille is my favourite synth ever. For me, nothing competes.)
Bazille's HP filters put out low frequencies when driven hard
-
- u-he
- 30215 posts since 8 Aug, 2002 from Berlin
Hehehe, easy answer:
It's a 4-pole lowpass filter. The non-lowpass outputs are generated by adding or subtracting filter stages and input in certain ratios. This works fine in the linear case, i.e. when there is no distortion. But as overdriving the filter adds distortion, the maths don't add up anymore and the filter types get blurred.
(you might want to try the highpass mode in a certain low cost Minimoog clone... not a highpass at all even when driven lightly)
It's a 4-pole lowpass filter. The non-lowpass outputs are generated by adding or subtracting filter stages and input in certain ratios. This works fine in the linear case, i.e. when there is no distortion. But as overdriving the filter adds distortion, the maths don't add up anymore and the filter types get blurred.
(you might want to try the highpass mode in a certain low cost Minimoog clone... not a highpass at all even when driven lightly)
-
- KVRAF
- 2856 posts since 10 Jul, 2008 from Orbit SW US
? How can it be a MM clone if it has a HP filter?Urs wrote: Tue Apr 07, 2020 1:26 pm ...
(you might want to try the highpass mode in a certain low cost Minimoog clone... not a highpass at all even when driven lightly)
gadgets an gizmos..make noise~crystalawareness.bandcamp.com/ soundcloud.com/crystalawareness Restocked: 5/2026
if this post is edited -it was for punctuation, grammar, or to make it coherent (or make me seem coherent).
if this post is edited -it was for punctuation, grammar, or to make it coherent (or make me seem coherent).
-
- u-he
- 30215 posts since 8 Aug, 2002 from Berlin
I was wondering that myself. Even the keyboard is missing.CrystalWizard wrote: Tue Apr 07, 2020 9:07 pm? How can it be a MM clone if it has a HP filter?Urs wrote: Tue Apr 07, 2020 1:26 pm ...
(you might want to try the highpass mode in a certain low cost Minimoog clone... not a highpass at all even when driven lightly)
-
- KVRAF
- 2856 posts since 10 Jul, 2008 from Orbit SW US
gadgets an gizmos..make noise~crystalawareness.bandcamp.com/ soundcloud.com/crystalawareness Restocked: 5/2026
if this post is edited -it was for punctuation, grammar, or to make it coherent (or make me seem coherent).
if this post is edited -it was for punctuation, grammar, or to make it coherent (or make me seem coherent).
-
- KVRist
- Topic Starter
- 43 posts since 6 Apr, 2020
Interesting! Thanks for the quick reply. What are your thoughts on this characteristic of the Bazille filters; do you see it as a limitation? I suppose it is possible to drive a signal through filter 3 or 4 to dirty it up, then attenuate it going into filter 1 or 2 to achieve a true highpass filter effect. I'm not sure if that would achieve the same sound as an overdriven highpass filter output, though.Urs wrote: Tue Apr 07, 2020 1:26 pm Hehehe, easy answer:
It's a 4-pole lowpass filter. The non-lowpass outputs are generated by adding or subtracting filter stages and input in certain ratios. This works fine in the linear case, i.e. when there is no distortion. But as overdriving the filter adds distortion, the maths don't add up anymore and the filter types get blurred.
(you might want to try the highpass mode in a certain low cost Minimoog clone... not a highpass at all even when driven lightly)
In another post, you mentioned you were considering adding switchable filters to a future version of Bazille, as in Diva. I find this idea very attractive. Would you consider adding a filter option that included dedicated (or 'true') highpass and bandpass filter outputs? I imagine the existing filter system is conveniently light on processing requirements, and I haven't noticed anyone else pointing out this characteristic of the HP filter, so maybe there is no demand for dedicated highpass filter output with its added processing cost.