I've read multiple times (for example in some of the answers to this question https://stackoverflow.com/questions/8869563/how-much-memory-can-be-accessed-by-a-32-bit-machine) that a CPU with 32 bit address bus can support at most 4GB worth of memory (2^32) excluding any extensions such as PAE. But doesn't that depend on whether memory is byte addressable or word addressable? Assuming it is word addressable, wouldn't the maximum memory that can be supported be 2^32 * WordSize? Granted such a CPU wouldn't be efficient due to the extra shifting that would be required to access byte level data but theoretically the memory capacity can exceed 4GB, right?


closed as off-topic by Patrick87 May 6 '14 at 16:22

  • This question does not appear to be about computer science within the scope defined in the help center.
If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ That's right, it's $2^{32}$ of the basic addressable units. You seem to already know the answer, so suggest closing the question. $\endgroup$ – Yuval Filmus May 5 '14 at 21:23
  • $\begingroup$ It's also not really a computer science question. $\endgroup$ – Raphael May 5 '14 at 21:42
  • $\begingroup$ I fail to see how this is not a compute science question. Is not computer architecture part of any Computer Science curriculum?! $\endgroup$ – Ashaman Kingpin May 6 '14 at 0:32
  • $\begingroup$ +1 @AshamanKingpin Computer Architecture is certainly part of Computer Science and such questions are welcome here. I have no idea why this question was closed as off-topic. It might have been closed since you essentially already had the right answer, but the reason given for closing the question is incorrect. Take comfort in knowing that your thinking is correct on this issue, and thanks for contributing. $\endgroup$ – Patrick87 May 6 '14 at 16:18
  • 1
    $\begingroup$ This question appears to be off-topic because it is about confirming an already correct answer to a question. The question can be made suitable by removing an attempt at an answer and posting the answer separately. (For now, the close reason defaults to the message seen below, although that message is inaccurate and does not describe the problem with this question. Thank you for your understanding.) $\endgroup$ – Patrick87 May 6 '14 at 16:22