CINEBENCH, wPrime, and AIDA64
These benchmarks are run using default settings and configuration.
AIDA64 AES and HASH
In this case margin of error is quite important because we are testing the same machine, just with different software modes. We see a slight bump in CINEBENCH, which is similar to the few percent increase GIGABYTE told us we would see, but it's also interesting to note how it depends on the mode of operation of the AI. At the edge was a bit slower than both disabled and cloud. In wPrime we see perfect scaling with cloud being best followed by edge and disabled. Once again in AIDA64 SHA-1 Hash test we see the same thing we saw in CINEBENCH, while in the AES encryption test it scaled much like in wPrime. In our AIDA64 memory test we are seeing basically the same results just with deviation due to margin of error.