Benchmark

Hash routine benchmarks screen

The Benchmark function will give you an idea of the computational speed of the checksum routines (hash methods) supported by ExactFile.

Select the in-memory buffer size for the data to be hashed.  The larger the size, the more accurate the benchmark.

Click Go to run the benchmark.

Notes about this function:

  • Unlike the rest of ExactFile, Benchmarking of hash routines is not multi-threaded. ExactFile will be unresponsive until the process is complete.
  • The Adler32 and CRC32 hash methods are small and fast enough to make the application overhead a larger issue. There is little point in comparing these to methods to each other.
  • The data buffer generated for the hash routines is created with a psuedo-random number generator seeded with the same value each time it is run. This means that for any given version of ExactFile, every data buffer used in the benchmark is the same, so the data itself is never a factor in differences between benchmark results. Note that the random number generator code may change between ExactFile versions.
  • The lower the time listed in the report, the faster the checksum method.

5 thoughts on “Benchmark

  1. Brian

    In a nutshell, one of best(if not the best) hashers I’ve ever used. Hats off to you sir and keep up the great work.

  2. Steve

    Both ExactFile and exf are awesome!!! summing tools. It’s a first time then I see so powerful tool and for free. Yeap. Full Unicode support and a lot of summing formats are unbeatable, but calculation speed is not as fast as it could be. For example, there is another good summing free alternative: “RapidCRC” and it’s ~40% faster than ExactFile. Anyway, I choose ExactFile for operations with unicode. Cheers!

  3. Brandon Post author

    @Steve
    Curious how you are comparing — ExactFile’s benchmark output is not designed to be compared against other tools, as overhead issues, etc, could affect the comparison. Are you actually timing ExactFile and RapidCRC? Also, which hash methods?

Comments are closed.