There's a skilled mathematician by the name of Martin Vicanek who is often sharing his latest ideas and mathematical skill without cost, and I recently looked over the ideas Martin has regarding inter sample peak detection.
MV employs a carefully designed lowpass and curve after amplifying the mono signal * 4. You would of course need to first change your stereo into mono. I would recommend to that end that you avoid mean averaging and stereo width as a means of deciding loudness.
Instead I would just create a switch that detects the loudest of left or right and offers that signal so to ensure that the loudest peak is captured. You could also have a mean average employed furthermore, but only after some subtraction and division, ensuring that your averaged loudnes is not less than the actual loudness. Otherwise you've missed the mark. It's only after you've captured the tru peak after all that you can then port it to rms or to a nice peaking db meter with enevlope control.
Download: https://dspplug.com/?download=1&kccpid=4110&kcccount=https://dspplug.com/wp-content/true-peak-detector-optimized.fsm
Edit: 27/10/2021 7pm updated the file reflecting any affected sample rate (other than 22050) within a few points.
Edit: 28/10/2021 1:44pm reduced fsm filesize by 50%, reduced cpu usage by 1% (roughly using 1.3% of a 4ghz 8core amd processor)
Intersample Peak detector
Return to “FL SynthMaker / FlowStone - Presets & Programming”
[You can only see part of this thread as you are not logged in to the forums]
Intersample Peak detector
You do not have the required permissions to view the files attached to this post.
Return to “FL SynthMaker / FlowStone - Presets & Programming”