If anyone ever runs out of precision using IIR filters: convert to second-order section (SOS). I stumbled upon this trick a few years ago and it's a game changer. No need for FIR.
I found the same. I once computed a high-order (something like 100) low-pass Butterworth using SOS sections as a test and did not notice any instability in the filtered signal. The key was to directly derive the SOS coefficients using bilinear transformation.
Yep. The other advantages are that now you can parallelize compute, and you get lower latency.