You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When cycling filter modes through LP/BP/HP, there is a harsh popping sound as the filter adjusts to a new state.
Ideas:
Consider a fast ramp down/up every time the filter mode is changed. When the mode is changed, delay the actual mode change for a brief time period, during which we fade out the input. This might be on the order of 0.01 seconds. Reset the filter to zero, then change the filter mode, then start fading in the input audio again for another brief time period. I believe the brief dip in audio level will be much more pleasant than the current popping sound.
Because the filter calculations themselves are so lightweight, it may even make sense to have two simultaneous filters operating during a brief cross-fade period. This would briefly double the CPU usage, but the CPU usage should be tiny anyway. The cross-fade period may need to be a good fraction of a second in order to sound smooth enough, but only experimentation will tell.
Put this in class GravyEngine, to unify behavior across future platforms.
The text was updated successfully, but these errors were encountered:
Turns out I didn't need any cross-fading. Maintain separate values of y1 and y2 (the previous two outputs) for LP, BP, and HP. Calculate all 3 filters simultaneously because the overhead is negligible. No more popping sound!
When cycling filter modes through LP/BP/HP, there is a harsh popping sound as the filter adjusts to a new state.
Ideas:
Put this in class GravyEngine, to unify behavior across future platforms.
The text was updated successfully, but these errors were encountered: