Super-threading (or time-slice multithreading) is a type of multithreading that enables different threads to be executed by a single processor without truly executing them at the same time.1 This qualifies it as time-sliced or temporal multithreading rather than simultaneous multithreading (SMT). It is motivated by the observation that the processor's functional units are occasionally left idle while executing instructions from one thread due to long-latency events. Super-threading seeks to make use of the otherwise unused processor cycles by executing instructions from another thread until the previous thread is ready to resume execution.
While this approach enables better use of the processor's resources, further improvements to resource utilization can be realized through SMT, which allows the execution of instructions from multiple threads at the same time. Consider a two-way super-threaded processor with four functional units. If thread one issues three instructions, one functional unit remains unused. In an SMT processor, it is possible for thread two to issue an instruction to the remaining unit, attaining full utilization of processor resources.