Scriptnode container.split functionality

[JL.LV]

There's something odd going on with the split node....Split node example.mov.zip
The node is supposed to split the signal and then sum at the end for processing, although it seems like there is some additional leftover signal that is being fed through

In the following example I have a bandpass filter set in the WET chain of the split node, while the unaffected signal going in the DRY chain of the split node. When I turn the crossfader to be 100% wet, the signal is indeed bandpassed, but If I turn the DRY gain off, the dry signal gets passed through.

Ive tried this out with other types of effects and the same thing happens. Its as if the dry signal always gets passed through either way

[ustk]

@JL-LV That is a normal behaviour. You bypass the Gain that is set to zero, meaning the signal goes through unaffected by its gain setting like in any bypass, be it software or hardware...

[JL.LV]

@ustk thanks for the quick reply

I want it to be a smooth dry/wet though, not just a hard bypass. it seems when using the split node this does not work

[ustk]

@JL-LV I don't understand, that is what you have already, it smoothly goes from dry to wet using the xfader knob. Your graph looks good

[Matt_SF]

@JL-LV I didn't understand at first but :

In your video, when you're bypassing the nodes :
The first time : the DRY signal is set to -100 (silent) and you're bypassing it, meaning that the signal becomes == 0dB.
The second time : the DRY signal is set to 0dB and you're bypassing it, meaning that the signal becomes == 0dB.

Maybe that's the strange behaviour you're talking about, which is actually the expected behaviour of the signal.
For a smooth DRY/WET, just use the xFader node - don't bypass anything.

[JL.LV]

ive found the problem....its because i was using multi mic samples, and scriptnode seems to have problems processing all the channels. this explains why sound was still coming out when bypassed