Anonymous user
Talk:Sorting algorithms/Cycle sort: Difference between revisions
m
→4 REXXen: added a comment.
m (→number of writes: added an clarifying adjective to a comment.) |
m (→4 REXXen: added a comment.) |
||
Line 25:
::: No thanks, I've beat that dead horse before. I assume the aforementioned REXX program works for ooRexx (since it uses a construct that is only valid in ooRexx); although it was my version that was translated to ooRexx, not version 1.
::: On the by and by, I timed the two REXX versions with identical data (suppressing the displaying of the input and output lists), and for a larger sized list of 1,000 times the size used in version 1 (same data, just replicated a thousand times), REXX version 2 is over '''20%''' faster than version 1 --- not that much of a speed-up, but enough to note the difference. I try to write compact code when speed is warranted (I've always assumed that speed is important when sorting). -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 18:50, 21 May 2020 (UTC)
[Note that ooRexx and NetRexx are considered different languages than REXX, but it is thought that it might be possible to (re-)write most REXX programs so that they will execute under ooRexx and NetRexx if certain restrictions are followed --- this is an on-going ··· discussion about this very matter --- does a programmer restrict their programming of (Classic) REXX so as to also execute under a different language such as object-oriented REXX?]
| |||