Skip to content
Karel Kubicek edited this page Feb 13, 2017 · 1 revision

Following is the list of bugs we know but probably won't fix:

  • Approximately 2 of 1000 runs of determinism/recommencing test fails due to a floating point rounding error.
  • The floating point comparison routine causes some processors to generate illegal instructions (happened on linux build, couldn't reproduce on Windows).
Clone this wiki locally