Hey guys I'm having a huge problem.
I've been running SB1.3 for many years (Still on Snow Leopard so I can't use v2a). I am working on a parallel compressor that can become quite usefull if you want to save on tracks in a session.
Thing is that he plugin works well, but I have a loud popping sound whenever I pass silence (muted audio) and whenever a signal goes through... meaning: If a run a song through it and it has 3 seconds of silence in the middle (or in any part), then I hear a popping sound two times: when the silence is on and when the sound returns... (happens in both AU and VST). It also seems to use A LOT of CPU when this happens (going even to 30%, when the average of use from this plugin goes near 3%)
-First I thought that it might be the amount of processing from the VU meters (I sectioned 4 meters in 10 parts each, so then I reduced it to 3 parts each)
-Then I thought it might have to do with the fact that I removed the Knee knob and replaced it with a static value...
- Maybe it was because I added a Makeup gain to the compressor.
- I thought it might had to do with the "on" buttons that I added on both channels...
None of the above... I tried fixing those parameters and nothing happened... still the same problem.
I attached the source file, so please check it out, and help me figure this out.... if you have any additional advice on how to improve this plugin, I'll be more that glad.
The other thing I want to know is if there's a way to have a little more control over the Display meters... they seem pretty erratic and unprecise... they react too slow and they don't mark every peak of the audio signal.
Last but not least: ¿¿There's no way sonicbirth 2 could be used on OSX Snow Leopard??
Thank you so much!!
VYERO
Taxicab Records
Loud popping sound when audio is muted, among other things
-
- KVRist
- 158 posts since 25 Mar, 2005
From a quick look: you divide by the output of the enveloppe follower. The output of the division gets crazy large as the output of the env fol goes to 0. Then you get NaN (not a number) when it gets to 0. Clamp the output of the env fol so it doesn't go below some sensible value (1e-6 or something).
-
- KVRist
- 158 posts since 25 Mar, 2005
I'll quickly add that a NaN value is contagious and will propagate through all operations (to simplify).
-
- KVRer
- Topic Starter
- 3 posts since 10 Dec, 2014
Thanks Makira!!! 
So, I should:
1.- Connect the LR input to the divider and the EQ to the envelope? (I got it the other way around)
2.- Set other values on the knobs around the divider (attack, release and Makeup gain)?
3.- Eliminate the divider and replace it with something else?
4.- ...none of the above?
How do I "clamp" the envelope's output?
So, I should:
1.- Connect the LR input to the divider and the EQ to the envelope? (I got it the other way around)
2.- Set other values on the knobs around the divider (attack, release and Makeup gain)?
3.- Eliminate the divider and replace it with something else?
4.- ...none of the above?
How do I "clamp" the envelope's output?
-
- KVRist
- 158 posts since 25 Mar, 2005
Well, I haven't looked enough at your patch to know what is your goal, and you should try to figure it out 
Use the debug element and/or the debug osc element, that should help!
I'll give you this: to clamp the output of env fol, you could simply use the 'maximum' element, with the output of env fol on one side and 0.00001 on the other. So the value will never go below 0.00001 (or whatever makes sense for what you are trying to do).
Use the debug element and/or the debug osc element, that should help!I'll give you this: to clamp the output of env fol, you could simply use the 'maximum' element, with the output of env fol on one side and 0.00001 on the other. So the value will never go below 0.00001 (or whatever makes sense for what you are trying to do).
-
- KVRer
- Topic Starter
- 3 posts since 10 Dec, 2014
Yes!!,... It seemed to solve the problem!... That's great!
The only thing is that the constant couldn't be set up to 0.00001 (it was automatically converted to NaN) I set it up at 0.0001 and it worked great.
I'll keep doing tests and I'll let you know if anything else happens.
Regarding the display meter question.. Any way to have more control over it?
The only thing is that the constant couldn't be set up to 0.00001 (it was automatically converted to NaN) I set it up at 0.0001 and it worked great.
I'll keep doing tests and I'll let you know if anything else happens.
Regarding the display meter question.. Any way to have more control over it?
