from Wittwer CT and Farrar JS (2011) in PCR Troubleshooting and Optimization

Terms such as "rapid" or "fast" are relative and vague. A 1 hour PCR is fast compared to 4 hours, but slow compared to 10 min. Furthermore, faster PCR is possible if you start with a higher template concentration or use fewer cycles. It is better to define the time required for each cycle and rapid-cycle PCR has been defined as 30 cycles in 10-30 min (See: Wittwer and Farrar, 2011 in PCR Troubleshooting and Optimization) so that each cycle is 20-60 s each. The actual time of each cycle is longer than the sum of the times programmed for denaturation, annealing and extension. Indeed, during rapid PCR the temperature is usually changing. This challenges the "equilibrium paradigm" of PCR, where 3 reactions (denaturation, annealing and extension) occur at 3 temperatures over 3 time periods each cycle (Wittwer and Farrar, 2011 in PCR Troubleshooting and Optimization). Although intuitive, the equilibrium paradigm does not fit well with physical reality. Instantaneous temperature changes do not occur and reactions occur over a range of temperatures at different rates. More accurate is a kinetic paradigm for PCR where reaction rates and the temperature are always changing. Under the equilibrium paradigm, a cycle is defined by three temperatures each held for a time period, whereas the kinetic paradigm requires transition rates and target temperatures(Wittwer and Farrar, 2011 in PCR Troubleshooting and Optimization).

Paradigms are not right or wrong and should be judged by their usefulness. The equilibrium paradigm is simple to understand and lends itself well to the engineering mindset. The kinetic paradigm is more relevant to biochemistry, rapid PCR and melting curve analysis. Although most commercial instruments still follow equilibrium protocols, rapid protocols are a nice match for microsystems, where small volumes and rapid PCR are natural (Wittwer and Farrar, 2011 in PCR Troubleshooting and Optimization). Rapid-cycle PCR is used in real-time instruments such as the Roche carousel LightCycler and Cepheid's SmartCycler. Other companies now promote "Fast" protocols on more conventional thermal cyclers. Few instruments based on microtiter plates and heat blocks can approach rapid-cycling speeds and rapid PCR does not require special reagents.