The suggestion Zebra and Matt gave me got rid of the error and is a lot closer to what I remember 1.0 sounding like for me
As far as the soft clipping goes I think I was running the signal in too hot. What I was hearing pass through was different than what it was recording at contrary to setting the monitor value to match the input level. (which is how I would set it in 1.0)
I just lowered the input value on the audio page down to around -6db under what the output and monitor are set to and its actually sounding pretty damn close to what’s going into the buffer.
From the very beginning of all of this I was trusting the graphical meters too much when setting the initial input levels. I’ve just ran 2 tests with different sources and its sounding pretty damn good now. Im so relieved. I really love Norns and didn’t want to have to abandon sampling functionality.
In the future then I would assume that I’ll need to modify scripts that use softcut with the filter functions to keep it as close to the original signal as possible yes?