Super-knock has been a significant obstacle for the development of highly turbocharged (downsized) gasoline engines with spark ignition, due to the catastrophic damage super-knock can cause to the engine. According to previous research by the authors, one combustion process leading to super-knock may be described as hot-spot induced pre-ignition followed by deflagration which can induce detonation from another hot spot followed by high pressure oscillation. The sources of the hot spots which lead to pre-ignition (including oil films, deposits, gas-dynamics, etc.) may occur sporadically, which leads to super-knock occurring randomly at practical engine operating conditions. In this study, a spark plasma was used to induce preignition and the correlation between super-knock combustion and the thermodynamic state of the reactant mixture was investigated in a four-cylinder production gasoline engine. The engine experiments were complemented by rapid compression machine (RCM) experiments of iso-octane and air which also used a spark plasma to investigate the fundamental physical and chemical mechanisms of super-knock. For the engine experiments, at low-speed high-load conditions, early spark timing was used to systematically induce preignition in the range of spark timing from 8 to -50 °CA ATDC in increments of 3 °CA. The intake pressure of the fresh air charge was set at 1, 1.3, 1.6 and 1.9 bar, and for each intake air pressure, the intake temperature was set at 20 and 60 °C. The results show early spark ignition could be used to trigger pre-ignition and to induce super-knock when the mixture thermodynamic state was above a critical condition. At a constant air temperature and high intake pressure, advancing the spark timing caused transitions from normal combustion to knocking conditions, to super-knocking conditions and then back to knocking conditions and to normal combustion. The transition between knocking conditions was well correlated with the thermodynamic state at the start of the in-cylinder pressure oscillations associated with knock. At a constant temperature and naturally aspirated conditions, no matter how advanced the spark timing was, only slight knock was observed, which indicates the thermodynamic state dominates the determination of knocking or non-knocking conditions. An energy density-pressure-temperature (E-P-T) diagram was developed to define super-knock, knock and normal combustion criteria. For the RCM experiments using isooctane at stoichiometric conditions with air dilution, a spark plasma was used to simulate pre-ignition. Three compression ratios were used to create different end of compression (EOC) temperatures. For each TEOC, with increasing EOC pressure, the combustion process also transitioned from normal combustion with flame propagation, to sequential end-gas auto-ignition with pressure oscillations and then to detonation with severe pressure oscillations. The combustion processes observed in the RCM experiments and the thermodynamic states were similar to the state conditions at the start of the pressure oscillations of the constant temperature and high intake pressure engine experiments. The results indicate RCM studies can be used to investigate the physical and chemical mechanisms of super-knock and E-P-T diagrams can be used to predict and distinguish super-knock from other combustion modes.