Discussion in 'General Electronics Chat' started by balamurali, Oct 28, 2006.
can anyone jus briefly explain me wat ia an register and its working and types????
Basically a register is a hardware component that stores some form of data. Several registers would be housed together in the form of a register block which can be accessed directly by the processor. Registers are a critical area of the hardware memory heirachy sitting between the cache memory (for most general purpose computing systems) or physical memory and processor. So in a rudimentary way data is placed in the register(s) from the cache or physical memory (depending on the memory design) and accessd by the CPU as required.
From the perspective of digital logic, shift-registers are a common type register and come in the following forms:
For more information check Volume 6: Chapter 12
thanks 4 ur answers,but i want to know its working clearly..can u plz brief in shorter way ie in crisper way...thank u
You need to ascertain whether you want to know the workings of a hardware register, which I detailed the basic workings of above. For more specific details on the component level you could look at D-type flip-flops or JK-type flip-flops. You could also check Wikipedia's article on hardware registers for a little more information.
Or are you wanting to know about logic registers. Other than replicating what is on the link above I can say no more. I would recommend you read the following topics to get a clear understanding of the workings of the 4 main shift register types:
If you're looking for a shortcut I think you may be out of luck.
ya exatly iam jus looking 4 logic registers something like which makes use of de flip flops...eg by using a d flipflop how is tat a shift register works i.e if u take it for 4 -bit number something like 1001 how does it work i.e how de bits get inserted into it and get saved in register ???
The short, crisp, brief, and easy answers are here:
That's as easy as it gets. It doesn't get any easier. If tat no e-z nuf 4 u den tat's 2 bad.
Firstly, I take it you are familiar with the workings of a D-type flip-flop? Otherwise you will need to review the rudimentary workings of such a device to appreciate how it works.
Secondly, you do understand the concepts or series input/output and parallel input/output?
So take an example: If you have a 4-bit serial-in/serial-out shift register initially at '0000'.
If your input bit sequence is '1001' then the dataword is clocked into the register in 4 seperate clock cycles:
The first bit is clocked into the first D-type flip-flop, and shift register is now '1000'.
The first bit in first D-type flip-flop is now clocked into the second D-type flip-flop, and the second bit is clocked into the first D-type flip-flop. The shift register is now '0100'.
The first bit in the second D-type flip-flop is now clocked into the third D-type flip-flop, the second bit in the first D-type flip-flop is now clocked into the second D-type flip-flop, and the third bit is clocked into the first D-type flip-flop. The shift register is now '0010'.
The first bit in the thrid D-type flip-flop is now clocked into the fourth D-type flip-flop, the second bit in the second D-type flip-flop is now clocked into the third D-type flip-flop, the third bit in the first D-type flip-flop is now clocked into the second D-type flip-flop, and the fourth bit is clocked into the first D-type flip-flop. The shift register is now '1001', i.e. your original dataword.
Because the register in this example is serial-out the dataword is read out serially, i.e. Cycle 5 will read in the next bit at the input but will also read out the first bit '1' that was read in in Cycle 1. If the register was parallel-out then during Cycle 5 the whole dataword would be read out from the shift register, i.e. the output of the register would be '1001'.
The above links will give you the correct operation of all the 4 main types of shift register in common use, to fully understand them you must read the appropriate section.
thanks for ur answers...satisfied to some extent
Have you read the suggested literature? If so and you are still having issues post back with what exactly is posing you a problem. Believe me all the answers you are looking for contained within the suggested links.
ya iam bit clear,but after reading it i got a simlple doubt ,the doubt is " a register is a collection of flipflops ..it acts a storage device r some thing like that .when that is the case .what is the purpose of removing all the binary bits again from the registers ..to be more clear when 1001 is stored in a register .why do u again take it out..thats ticking me and the next thing is why ia clock pulse applied to input it in a flip flop...hope i sound clear..
Some investigation on you part into digital processors would explain your questions. Registers are the internal structures that computers use to hold the number they are operating on. Since a computer is a numeric engine, some means of holding numbers is central to their ability to function (the programmer decides how to handle the numbers).
iam not just talking about the entire network by which the register is made used in a pc..i jus want to know why clock pulse should be given for every f/f ..in a register ..to be more clear simply you giv an input u can get the o/p..y do u want giv an clock pulse ,which just changes the present state of the i/p..really iam damn confussed with this topic
Since the flip-flops are all D type, the only way to get data into them is with a clock pulse. Look up how 74xx74's work.
The point to remember is that registers are sequential logic (reference the introductory notes on the links provided), in the sense that they are dependant on the current and previous input values to a stage (flip-flip), we can say that they exhibit memory (which should be clear from the discussions previously).
Sequential logic (generally) requires a clock, upon which the clock transistion instigates a change in the state of the stage, and the flip-flop provides the ideal hardware solution for these requirements.
very funny, love it!
Storage devices need to be told when to store data. The clock pulse locks the data in. The output remains constant until another clock pulse copies the input data to the output pins.