Idea I had after watching this video on Serum resampling:
https://www.youtube.com/watch?v=5FLNn_0U4E0&t=583s
If you render a waveform, say, a simple saw, with 16 voices, as a new wavetable, you should be able to make those phase-y supersaw plucks without high CPU usage simply by rendering with Unison=16 and then importing the wavetable again.
Just an idea if anyone has CPU issues with Serum while using many voices. I believe this would help drop CPU. You can also resample with FX but personally I would rather keep all FX out of the wavetable and do them in post.
It's somewhat like bouncing to audio to drop CPU except you can just render the unison'ed version as a wavetable and you still retain live control over every parameter (except unison of course).
Submitted May 02, 2018 at 10:59AM by whatthefuckmane https://www.reddit.com/r/edmproduction/comments/8ghlpm/render_to_osc_a_b_in_serum_reduce_cpu_usage_by/?utm_source=ifttt