I believe what you observed. For those who talk only 9 HP lost, it is related to much more than that. The entire power band is affected. And typically much more than a published spec. It has to respond instantaneous to the knocking, hard core. Especially if conditions of the air are not perfect for published spec measuring ( sea level, 29.98, 58 degrees, and 40% humidity or so). In some weather conditions, power across the curve can be increased beyond published spec by the programming (or forced to detune much more). The affect on the overall power band is much more than 9 HP at peak. You will feel the difference, as you have observed. And it will take some time to get the system set up back to the condition because it is set to adapt back slower.
As for the comments for those sniping from under the bridge about torque steer, I understand what you are saying. I assume you meant having to correct the
wheel to keep it straight. Enough breakaway power in the power curve to cause the tires to supersede for a short time the nanny ESC. And kinda funny, because of the bushings and no
rigid collars in the subframe/suspension mounts, the rear suspension May in fact "steer" a bit from intended line under certain torque related conditions. [emoji41]
In the old days, a engineer acquaintance at Ford had print outs of different graphs and curves for power at different conditions and fuel. He studied them for his autocross weekends we both competed. He would use them in a kinda secret way to understand how much toluene to sneak into the gas. He claimed toluene was like hooking up a O2 tank to the car. Giving it stronger O2 in combustion process. Generating more potential power with right CPU settings. And he would reset/program the CPU in the stock class car. Plus crank up the boost in a custom chassis modified class turbo car. He wanted to help me use toluene to pick up some extra power. I was too chicken to try.. Through him I learned about how the mapping works in general, and how different quality fuels affect how the CPU can help create power with higher octane (and in his case adding things that created stronger explosion in the cylinder, and the effects of higher lead free 100 aviation gas to those additives.. Verses lower octane).
That was in a time that compared to today, the computers and sensors were much more limited. With a high compression engine, adjustable cam timing that have aggressive limits, spark, GDI, and much more, available on the TAU ( and much better/faster CPUs and programming), what is possible today to detune (87 octane) high performance engines (conversely increase performance with optimal fuel) would be so much better.
It is asked above if the V6 is set up where 91-93 under 5% ethanol rating would help the HP curves. That is a question to Hyundai. In general, the V6 GDI has some nice and high compression. They would have to answer if the CPU, cam systems, and other things could take advantage of higher octane. In general publications I have not seen where they say it is optimal or designed to run with premium. I bet in some weather conditions, it could help IF the CPU mappings were designed to react quick enough to capture a advantage of some temporary weather condition. (Low altitude Ultra cold and dry as one example).