I just wanted to make sure if I'm reasoning correctly. So, if two computers are solving the same mathematical puzzle i.e. SHA-256 function of Bitcoin (finding nonce that satisfies difficulty target), the computing power used by each of them can be compared in units of Gigahashes, Terahashes etc. and we know which one used more computing power in time. But what if we wanted to compare computing power (herein called "CP") used by two computers in time t if those two computers were solving different hash functions (e.g. one SHA-256 and one SHA 3-512) or different mathematical functions altogether? Could we then denominate computation power used by Computer 1 and 2 to reach a conclusion that in time t: Computer's 1 CP > Computer's 2 CP and by how much? And what would be that proper and best denominator (FLOPS, MIPS, IPS..)?

  • $\begingroup$ It depends on lots of factors. It seems that your question will not have a definitive answer. Are you actually interested in a more restricted question? Say, how are the most powerful supercomputers are ranked? $\endgroup$ – Apass.Jack Dec 29 '18 at 13:04
  • $\begingroup$ Question is definitive. What factors, in your opinion, should be taken into perspective? Could you elaborate? Question is about unit of measurement for general comparison of computing power used to solve a given mathematical puzzle, where the puzzle is referent, but the set of puzzles and their kind is not determined in the question. $\endgroup$ – Pppidoo Dec 29 '18 at 13:32
  • 1
    $\begingroup$ A math puzzle can be a problem about 3D drawing. A math puzzle could be about symbolic computation. Almost any computational problem can be considered as a part of a math puzzle. A computer with some GPU might be better for graphic problems. $\endgroup$ – Apass.Jack Dec 29 '18 at 13:41
  • $\begingroup$ For many problems, there are computers built specifically for them. People are building specific computers to compute neural network/artificial intelligence. Specific computers have been built for Bitcoin farming. $\endgroup$ – Apass.Jack Dec 29 '18 at 13:45
  • $\begingroup$ Solving some problems needs enormous space inherently. The memory size of a computer should be considered as an important part of the power for those problem. $\endgroup$ – Apass.Jack Dec 29 '18 at 13:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.