Open toliaferrode opened 8 years ago
Confirmed with LMMS 1.2.0-rc.4.
Could be that the control value is simply interpolated and it is initially at a cutoff value defined by the calf people and when you open it fresh it will then do a bit of a trip to whatever value is defined by the lmms ladspa settings?
If I read this correctly the cutoff value would start it's life here with a knob value of '0'. Then it would get set to it's proper value here and the value is interpolated here but I don't know if that interpolation could answer for a sweep that long. Maybe. I'm not concerned by that sweep as you don't usually record sound live when adding/removing pluggins. What I'm a bit worried about however is initializing the model to '0' as that assumes the pluggin isn't doing something silly as actually using a frequency of '0', which actually has been the case if I remember correctly. Some HP filter was trying to EQ '0' Hz resulting in something glitchy. This was pre 1.0 .
I wonder if #3929 has fixed this issue. Can anyone re-test this?
Tested. I see no difference.
This is also on the LV2 version of calf filter. The behavior is the same when I add a module in Ardour @JohannesLorenz You hack calf. This is an upstream issue right? Close-ish?
If you add a Calf Filter to an instrument or channel while the engine is running, or on it's first play after it's added, or on export, the cut off seems to sweep briefly. I have one project where it didn't do this (I think), but in all my subsequent tests it does it.