User:Tarphon21/sandbox

<!--I would like to comment on the introductory paragraphs of the article on 'Flip-flop (electronics)'.

a) The introductory sentence 'In electronics, a flip-flop or latch is a circuit that has two stable states ...' gives the impression that flip-flops and a latches are one and the same thing whereas most authors distinguish between these two concepts: A latch (as it consists of a single memory element) has 2^1, i.e., two states, a flip-flop (as it consists at least of two memory element) has at least 2^2=4 states. Furthermore, in the above introductory sentence there should already be the reference to 'State (computer science)' to document that 'state' is a technical term, and to explain what it is. Furthermore, flip-flops or latches need not be 'circuits'. There are multiple examples of pneumatic, hydraulic, electric, and electronic 'devices' that behave as latches, some, even as flip-flops. This makes it obvious that the concepts of a flip-flop or a latch are not restricted to electronics or even to switching circuits.

b) Consider the given definition 'A flip-flop is a bistable multivibrator'. If one follows the link to bistable multivibrator one does not find a general definition of a multivibrator (as would be needed to abstract from it the term latch), one only finds an example circuit which, moreover, is exactly the same circuit as in the first figure in the flip-flop article. The definition of a latch thus boils down to 'a latch is the following circuit'. This is clearly not general enough.  I would prefer a stand-allown definition as given in S.P. Vingron 'Logic Circuit Design', Springer Heidelberg 2012 (here, put forth colloquially together with a brief amendment):

A latch, in the field of sequential logic circuits, is a switching device or circuit with two or more binary inputs and a single binary output, the latch's primary property being that there exists at least one (present) combination of input values which retains or memorises the output value caused by the previous combination of input values; these (present) combinations of input values are called memorising. The remaining combinations of input values are called either setting or resetting. A setting input combination causes the output to be 1 irrespective of the previous output value. A resetting input combination causes the output to be 0 irrespective of the previous output value. Any combination of input values is either memorising, setting, or resetting.

Latches with only two inputs are called elementary latches, of which there are 36 (as opposed to 10 combinational functions with two inputs and one output). Latches come in all kinds of technologies (e.g., mechanical, fluidic, pneumatic, electric, magnetic, electronic). Many of these memorising devices (i.e., latches) work without feedback. Nevertheless, feedback plays a dominant role in developing latches because any desired latch and its relevant properties can be calculated using switching algebra, the result of the calculation then being interpreted as a logic circuit with feedback which in turn can be realised, for instance, electronically. All latches, irrespective of the number of their inputs, can be realised by a circuit with a single feedback.

c) Consider the leading sentence in the third paragraph of the article : 'Flip-flops can be either simple (transparent or opaque) or clocked (synchronous or edge-triggered); the simple ones are commonly called latches.' Neither of the terms transparent, opaque, synchronous, or edge-triggered are explained. Transparency, for instance, is not an all together self-explaining concept, as Edward J. McCluskey in 'Logic Design Principles', Prentice Hall, 1986 (ISBN 0-13-539784-7) points out on pages 277 and 278:

'A very important characteristic of a latch is that when an input value is changed, any effect on the output appears right after the new input appears. The new output is delayed only by the propagation time of the devices in the circuit. All latches have this property, but it is sometimes emphasized by calling them transparent latches [Texas Instruments, 1981 Supplement to the TTL Data Book, 2nd ed., Texas Instruments, Inc., Dallas, 1981]. A word of warning is appropriate here. The terminology used in connection with bistable circuits such as the set-reset latch can be very confusing since there is a lack of consistency and care in naming such circuits. The biggest problem with naming bistable circuits is the failure to identify carefully those circuits with the transparency property. Often the term flip-flop is used indiscriminately for bistable circuits. In this book the term latch will be used for bistable circuits with the transparency property and the term flip-flop will be reserved for circuits without this property.'

On the other hand, I was not able to find a reference to 'opaque' in connection with latches. So what property characterizes an opaque latch? The terms 'transparent' and 'opaque' would seem to refer to opposite meanings. Next, look at the second half of the above mentioned sentence 'Flip-flops can be ... clocked (synchronous or edge-triggered) ': Is 'synchronous' the same as 'edge-triggered'? Does this mean that a flip-flop must be clocked? I, for one, wouldn't agree.~ -->