Flip flop state storage

Simple digital systems are assembled from well-known libraries of parts combinational logic gates (and, or, inverters, tri-state buffers) storage elements (flip-flops, registers, RAM) synchronization parts (clocks, dividers). From these parts are assembled a huge variety of systems, but all of them can be understood in terms of these basic parts. 

Flip-flops are state storage elements which have provisions for programming and, in many cases, clocking. One architecture which is useful for making clock dividers is known as the T flip-flop, whose design is shown in Fig. D.6. The T flip flop stores state when the input is steady, but inverts its output state on a rising input edge. 

In each of these two cases, the assignments made within the Else clause create feedback loops around the flip flops. This allows the present states of the object(s) to be preserved. In the left statement, the variable has been synthesized as a signal on a wire, not as the output of a storage element. On the other hand, in the right statement the variable is read before being assigned. It is therefore synthesized as the output of a flip flop with feedback. 

There are two general types of RAM, static RAM (SRAM), and dynamic RAM (DRAM). SRAM is based on the flip-flop circuit discussed previously. Two cross-coupled transistors comprise the basic storage unit, a logical "1" if one transistor is "on" and a logical "0," if the other is "on." Two more transistors sense the state, and two more transistors control the access to the storage location, so six transistors are required to store one bit of information. Access time ranges from 10 to 30 ns. The term "static" refers to the fact that the information remains stored as long as the system is powered. When power is removed, the state of the flip-flops becomes indeterminate hence, data is lost unless stored on the hard drive. Such a memory is called a volatile memory. 

---