This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Instruction execution time at 16MHz in ADuCM360

I am writing some part of ADuCM360 program in assembly. I am running at full 16 MHz internal clock. My question is what will be time taken by chip to execute a instruction in assembly for example for a Nop() instruction.

I will appreciate early reply

Thanks and Regards

  • On average an arm can process a instruction in one clock cycle. This is because it
    will pipeline instructions so that it can fetch decode and execute all of its instructions
    in parallel and will be able to process a instruction every clock cycle.

    16 X 10**6 hertz clock frequency has a period of (take the reciprocal of frequency)
    62.5 x 10**-9 or 62.5 nano seconds.

    it you were to only write one instruction though it would take three times as long since
    the pipeline of the processor can not be used for juat one instruction.

    It would take 1 cycle to fetch, 1 cycle to decode and 1 cycle to execute just one instruction

  • many have got in trouble with calculating times with pipelined architectures, Gary mentions some cases and every cache miss, for whatever reason, will make your calculations invalid

    Erik

  • The best you can possibly do is state the minimum time of execution.

    Can be acceptable for very short and unimportant delays, but probably not much else.