Talk:Last letter-first letter: Difference between revisions
→Timings
Line 82:
Could the timings on the D examples be removed or changed to just a statement of their relative run times under similar conditions, as we don't compare language run times on this site as it is [http://shootout.alioth.debian.org/dont-jump-to-conclusions.php too difficult a task to get right]. If a language has only one timing then I guess it should be removed altogether? Thanks. --[[User:Paddy3118|Paddy3118]] 18:36, 28 June 2011 (UTC)
: Those timings don't even list the CPU used on purpose, because they are meant to give just a general insight about the efficiency of the implementation. It's not so important to know that a better CPU runs the code a bit faster, but for the reader it's useful to know when a Python version takes about one minute to run, while the C code takes half a second. This site isn't a benchmark Shootout, but some information about the runtime efficiency of the code shown is interesting information for the readers. Languages and their implementations differ by syntax, semantics, efficiency, etc.
:: I am afraid I am going to have to disagree with you (whoever you are). You can easily get order-of-magnitude differences from cpu speeds alone (for example, if you compared an older mini-itx system with a high-end modern system), and other issues (like system load or memory architecture) can also be significant. (Not that imaginary timings were a particularly persuasive argument in the first place..) --[[User:Rdm|Rdm]] 00:58, 29 June 2011 (UTC)
| |||