After they removal of lead in gasoline the rule of thumb became to move the decimal over one place to the right and that was the octane the car needed. Say a 9.34:1 would need 93 octane.
This is a little bit old school with the advances in engineering. A lot of car manufactures have adopted the use of aluminum heads. This allows them to run higher compression ratios, more advanced timing curves to achieve higher MPG and performance from cars. Not to mention it is dirt cheap and very light weight. The aluminum heads shed off heat at such a faster rate that the fear of detonation is greatly reduced. Even cars that still use iron heads have became cutting edge in there design to handle the heat.
Most OBD-II cars can sense what type of octane the gas is in the tank. With the higher octanes the timing will advance. With the lower octane it will retard. Any new school car can run any octane. Before the EPA got rid of MBTE in gas for ethanol you could see quite a differance between 87 and 93 octane in terms of MPG.
So with hydrogen you would think that the lower octane would play to your advantage because of the more retarded timing curves. Each car is different in terms of how sensatvie they are so I would suggest some experimentation to see if you see and feel what you want to out of the reduced octane.
Look, I'm all for the HHO technology, but the biggest problem I see with everyone's setup is the complete lack of metering the HHO gas. Although you may have enough HHO to augement the lean mixture at a cruising speed, engine's don't knock at cruising speeds. they knock under load.. Now for those who have modified their map/maf for a leaner burn, you have more chance of detonation. throw less octane into the mix, and the problem is exaggerated. Those who run just an o2 mod, you'll have 'normal' fueling conditions under load which would be safer.
But my point.. still have a problem with 2lpm of HHO mixing with anywhere from 500lpm to 5000lpm of a/f through all different size motors, running lean, with zero metering. with that being said, how can anyone expect HHO to make up for reduced octane if it's percentage varies throughout the RPM band?