Quote:
Originally Posted by sai'ke
Ah whoops. Good catch! I was handling them at a per block level. D:
It should be fixed now. I also added a tiny lowpass filter for the dynamic variable, since sometimes, the extremely rapid transitions could cause some clicks.
Changelog 1.42
- Increased precision MIDI handling.
- Interpolate dynamic variable.
Changelog 1.43
- Display interpolated dynamic variable rather than raw.
- Minor bugfix MIDI handling (could be delayed by 1 sample before, which caused issues if MIDI was exactly at t=0).
Changelog 1.44
- Bugfix diode ladder. Thing should be reliable now.
|
You are just crazy!
Here is how it sounds after fixing. This is exactly as I'd imagined it before. You wouldn't believe that this filter is added after the synth. Amazing!
https://instaud.io/3aw0
Here another example. The source is only synth1 without its own filter.
https://instaud.io/3aw4
And in the next example I noticed two problems. First: The amount of filters isn't correlating to the automation input. So if you want to change the filter type thru automation (as I wanted for my examples) you only get half of the available filters to choose from. (The vertical course of the envelope lane does not cover the amount of filters.)
Second: If I make a hard jump from one cutoff sweep to the next, Filther needs a lot of time to get the cutoff at the right start point. You can hear a hard short sweep every time. Not sure if this is dependent on Filther's (or any filter's) structure but I thought that it has a look ahead function for that.
In this example the first loop is the clean synth1 output. Afterwards I changed various filter types with a filter cutoff sweep. (clean input, not optimized, just plain filter sweeps) It shows how much character a Filter adds to the sound. Just stunning! -`ღ´- (✿ ♥‿♥)
(But notice the weird behavior when the cutoff sweep starts again.)
https://instaud.io/3aw7