Frank Sheeran
2018-10-02 02:33:16 UTC
I guess it can be
proven that the final value is a monotonic function of delta, but
possibly with large multipliers the output value step (for the smallest
change of delta) will be quite large, so the search will not fully
converge.
Yes. For instance with Start = 0+-47dB, End =1+-47dB, Curve = 5, Time=2.5proven that the final value is a monotonic function of delta, but
possibly with large multipliers the output value step (for the smallest
change of delta) will be quite large, so the search will not fully
converge.
sec, SampleRate=44100, there is literally no double for Delta that gives
you 110249 samples. You can get 110248 or 110250!
When I started this project I spent a lot of time trying to be
sample-accurate, but actually I'm not trying to build lab instrumentation.
It's supposed to be a musical instrument. I think most musicians would
accept something that's 99.999% accurate. Or even 90% accurate...
(One place where cutting corners didn't work though is that I made a fast
LFO that approximates with straight lines of 32 samples. That affects the
range of frequencies you can set it to. You would think it would be fine
anyway... but it can't sync to external clocks whereas the old one with
full floating-point frequency could...)