Talk:Multi-dimensional array: Difference between revisions
→Column-major order in Fortran
Line 62:
:::Puns intended. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 12:35, 25 October 2016 (UTC)
::::Agreed. I changed the task to explicitly ask about this storage order: it is not specific to Fortran, and whatever the convention may be, it has to be known in order to use the language correctly, for cross-language calls, memory cache optimization and possibly for disk storage if the memory is mapped directly to a file.
::::Anyway, what was written was wrong, and I don't think it's a good thing to have wrong comments on a site supposed to be useful for students, in a task supposed to be tricky to newcomers. It's still far too lengthy to my state, by the way. Looks like an old-timer enjoys remembering the
:I agree that I should not have mumbled that mathematics uses "row major" since in the Platonic space the issue of storage doesn't arise, but I was trying to explain the difference between the expectations engendered by mathematical notation and layout, and the "formula translated" with an eye to the many who have been disconcerted by the results of WRITE(...) A after a READ(...) A followed by some calculations. And indeed there are computations that may be expressed row-wise or column-wise and there may be mathematical reasons to prefer one over the other or to regard both as equivalent. But when staring at the realisation of one in Fortran, where when the mathematics calls for (i,j) and (j,i) one may find (j,i) and (i,j) perhaps with DO J...; DO I... when DO I...; DO J... were expected, but that after some inspection seems to be consistent and indeed correctly implementing the algorithm, so, why? Aha, someone has discovered "thrashing" and been driven to taking steps... [[User:Dinosaur|Dinosaur]] ([[User talk:Dinosaur|talk]]) 01:19, 26 October 2016 (UTC)
| |||