Apple M1 performance in standalone Reaktor is better with lower buffer size

I just discovered that Reaktor appears to perform significantly better using a lower buffer size (32 samples) than the default (512 samples) on M1 in rosetta mode. Can anyone out there with an M1 confirm this behavior?

For example when using Reaktor's standalone and CPU meter:

Polyplex:
32 samples buffer size = ~12.2%
(35% when using Activity Monitor)

512 samples buffer size = ~25-30%
(59% when using Activity Monitor)

Razor:
32 samples buffer size = ~6%
(35% when using Activity Monitor)

512 samples buffer size= 21%
(50% when using Activity Monitor)

In any case it's a good problem to have, but the factory default setting of 512 leads to an underwhelming impression on first try.

 

I have nothing to add to add to this thread because I can't even get Reaktor to launch on my M1 (Code Signature Invalid exception is thrown). Can I ask what version of Reaktor is working for you?

 

Strange, sorry you've had issues. I'm running Reaktor 6.4. No issues at all with it and Big Sur 11.1