MMultiBandWaveShaper question
-
- KVRian
- 1102 posts since 30 Oct, 2005
I dont think so...but it would be nice to have such possibility!borkman wrote:Can MMultiBandWaveShaper be used for asymmetric distortion, as in, the positive part of the wave has a different transfer shape than the negative?
/so far Im using MMSaturator for assymetric distortion.../
-
- KVRAF
- 1924 posts since 15 Oct, 2008 from Germany
Just today I was thinking the exact same thing (while demoing Saturn and Trash). An asymmetric mode would be awesome.
Until then here's a work around:
- add an MUtility. Gain -6db. Math: Add 0.5. This moves the whole wave into the positive range [0, +1].
- add MWaver/MMultiBandWaveshaper. Put in linear mode and set to 100% wet. Put a node at -12db/-12db. Now use the range between -12db - 0db to shape the positive part of the wave and the range -12 - -inf for the negative part.
- add MUtility. Gain +6db, Math Add -1. This stretches the shaped wave back to [-1, +1]
- optionally: add MUtility, use Math Add to adjust DC offset.
And there you have it - asymmetric wave shaping for your personal tube distortion needs
And yes, this can be neatly encapsulated in MXXX. It's 100% correct because the lower part has a different distribution of the values, but you can get some nice results anyway.
Until then here's a work around:
- add an MUtility. Gain -6db. Math: Add 0.5. This moves the whole wave into the positive range [0, +1].
- add MWaver/MMultiBandWaveshaper. Put in linear mode and set to 100% wet. Put a node at -12db/-12db. Now use the range between -12db - 0db to shape the positive part of the wave and the range -12 - -inf for the negative part.
- add MUtility. Gain +6db, Math Add -1. This stretches the shaped wave back to [-1, +1]
- optionally: add MUtility, use Math Add to adjust DC offset.
And there you have it - asymmetric wave shaping for your personal tube distortion needs
And yes, this can be neatly encapsulated in MXXX. It's 100% correct because the lower part has a different distribution of the values, but you can get some nice results anyway.-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122 - KVRAF
- 14325 posts since 15 Mar, 2008 from Czech republic
paterpeter: Yeah! That's called creative thinking! 
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14325 posts since 15 Mar, 2008 from Czech republic
Ok, so new version will have assymetric option for all waveshapers
-
- KVRAF
- 1924 posts since 15 Oct, 2008 from Germany
Awesome! Thanks Vojtech!
BTW, if somebody uses the workaround I mentioned, make sure the input signal doesn't clip, otherwise you'll get very nasty artifacts.
BTW, if somebody uses the workaround I mentioned, make sure the input signal doesn't clip, otherwise you'll get very nasty artifacts.
-
- KVRAF
- 1924 posts since 15 Oct, 2008 from Germany
Feature request for the asymmetric mode: optional automatic DC offset compensation. IIRC a single pole non-resonant highpass filter at 20Hz pretty much does the trick.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14325 posts since 15 Mar, 2008 from Czech republic
Ok, good point.paterpeter wrote:Feature request for the asymmetric mode: optional automatic DC offset compensation. IIRC a single pole non-resonant highpass filter at 20Hz pretty much does the trick.
-
- KVRAF
- 1924 posts since 15 Oct, 2008 from Germany
Another feature request: a "wiggle" or "sine" curvature would be cool - like pulse, but smooth. And likewise a saw-tooth curvature.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14325 posts since 15 Mar, 2008 from Czech republic
Okey dokey folks, wait for an update
-
- KVRAF
- 1924 posts since 15 Oct, 2008 from Germany
Hi Vojtech. I just checked the latest version which includes the asymmetric mode. Great stuff! But I have a FR: the DC offset filter seems to work at a rather high frequency, 100 Hz I suppose. Enabling it reduces the low end of the signal very audibly. Could you please reduce the cutoff frequency to e.g. 20 Hz? It should be enough to get rid of the DC offset (which is basically "0Hz") without affecting the low-end to much.
E.g.: here's pink noise:
The noise running through MWaveShaper, 100% wet, no change of the curve, DC blocker active:
Note the drop below 200Hz. That's very audible.
Here's the noise with just a non-resonant 6db/oct 20Hz HP filter (using MEqualizer):
Obviously, the changes are almost inaudible.
But this filter still works fine as a DC offset filter. Consider this wave-shaping (asymmetric, totally flatten upper side of the signal):
Without a DC filter, a sine wave looks like:
With a simple MEqualizer in front of it...
... the DC offset is taken care of rather well:
Here's the sine wave with the DC blocker enabled:
It's not much better or worse than the filter @20Hz.
E.g.: here's pink noise:
The noise running through MWaveShaper, 100% wet, no change of the curve, DC blocker active:
Note the drop below 200Hz. That's very audible.
Here's the noise with just a non-resonant 6db/oct 20Hz HP filter (using MEqualizer):
Obviously, the changes are almost inaudible.
But this filter still works fine as a DC offset filter. Consider this wave-shaping (asymmetric, totally flatten upper side of the signal):
Without a DC filter, a sine wave looks like:
With a simple MEqualizer in front of it...
... the DC offset is taken care of rather well:
Here's the sine wave with the DC blocker enabled:
It's not much better or worse than the filter @20Hz.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14325 posts since 15 Mar, 2008 from Czech republic
Well, ok, I'll check it out, but it's weird, I was checking the filter response and I thought it looked different. We'll see .
.