You are here
Home > VLSI Design >

The difference between latch and flipflop VLSI

The difference between latch and flipflop VLSI
The difference or the comparison between the latch-based design and the flip-flop-based design is explained briefly. Let us compare the two types of design a Latch-based design and a Flipflop or Register-based design. These also help to understand the advantages and disadvantages of the latch and flip-flop.

A Latch is a transparent device and is level-sensitive or pulse-sensitive. Either it can be a positive level triggered or negative level triggered.
A flip-flop is also a transparent device, which is made up of a pair of level-sensitive latches, and the flip-flop is edge sensitive.
The latch is very less sensitive to the pulse duration i.e. soft barrier.
The flip-flop is more sensitive to the pulse transition i.e. hard barrier.
The enable pin EN in the latch is less sensitive to glitches caused by the data signal.
The flipflop is more immune to glitches when compared to latches on the EN enable pin.
The latches take a lesser number of gates so, a Latch takes less area compared to a flipflop. And hence the less power.
Since the flip-flops are made of a pair of latches, a flip-flop takes more are than a latch and hence more power consumption.
The Latches are faster than the flip-flops. That is by using the latches the longer combinational paths can be compensated by using the shorter path delays in the upcoming logic stages.
Compared to latches the flip-flops are slower. Here in the flop-based design the longest path of the design limits the timing performance of the circuit. i.e. Tcombinational < Tclock_period
It is complex to verify the timing in latch-based design since they require more tool manipulation and human calculation.
The static timing analysis STA method can check for timing performance and verify it very fast.
Here in the latch based design if there is a noise in the input signal it gets propagated to latch output easily.
The flip-flop based design is more robust.
Last but not least the cycle borrowing or the time borrowing in the latch to retain more setup time in the next logic stage, until one cycle completes. So most of the timing designers consider latches  with the purpose of adjusting the mismatch in the timing.
The flip-flop is more strict, if the data triggered a rising edge it must setup to a stable state before the next rising edge coming up. If the data arrives late the system may fail or go to a metastable state and will produce irrelevant output.

Leave a Reply