I have observed that a particular question has been asked, and separately answered, multiple times.

In chronological order:

There are some unique variations among these but there has been a great deal of redundant work in creating these answers. I am seeking a way to make the existing solutions more evident and if possible group them in a single place for easy reference.

The numeric methods in the first Q&A are highly efficient and directly applicable in many of these cases (as C. E. has been posting as answers) but I do not think they are easy to find. Whatever "advanced indexing" is and whether or not Mathematica has it sounds unrelated to many of these, yet many of these questions could be closed as a duplicate of the 2821 if it were written differently. Could we perhaps turn this into the canonical Q&A with a more transparent title and content, without losing the meaning of the original?

In a sort of tangential way How can I make threading more flexible? is also related, and has its own cloud of duplicate and near-duplicate questions. It might be nice to acknowledge one Q&A cluster from within the other.

  • This is giving me a lot of trouble: mathematica.stackexchange.com/q/120522/12
    – Szabolcs
    Commented Jul 12, 2016 at 13:12
  • @Szabolcs GreaterEqual is not the only one.New functions added in version 10.0 or later are usually very slow
    – yode
    Commented Jul 12, 2016 at 14:05

1 Answer 1


After the better part of a year with no bites I decided to attempt a canonical reference to the best of my ability. That now resides at:

As referenced for a related class of problems requiring speed on numeric arrays there is:

And also as referenced in the question one should view:

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .