I can get more consistent results when running the examples in Ubuntu.
On Monday, November 5, 2012 6:09:25 AM UTC+8, Marek Šrank wrote:
>
> These are my results using (dotimes (time (dotimes ...))) combo to get
> consistent results:
>
> - on sets:
>
> (dotimes [i 10] (time (dotimes [j 100] (into #{} (range 1e4)))))
> "Elapsed time: 388.595226 msecs"
> "Elapsed time: 406.666801 msecs"
> "Elapsed time: 421.559753 msecs"
> "Elapsed time: 444.719263 msecs"
> "Elapsed time: 431.319374 msecs"
> "Elapsed time: 392.3779 msecs"
> "Elapsed time: 392.382365 msecs"
> "Elapsed time: 408.71734 msecs"
> "Elapsed time: 444.203973 msecs"
> "Elapsed time: 442.351363 msecs"
>
> (dotimes [i 10] (time (dotimes [j 100] (naive-into #{} (range 1e4)))))
> "Elapsed time: 1287.412806 msecs"
> "Elapsed time: 1318.931199 msecs"
> "Elapsed time: 1314.773064 msecs"
> "Elapsed time: 1292.664029 msecs"
> "Elapsed time: 1277.616099 msecs"
> "Elapsed time: 1289.514678 msecs"
> "Elapsed time: 1700.74875 msecs"
> "Elapsed time: 1352.734662 msecs"
> "Elapsed time: 1464.09882 msecs"
> "Elapsed time: 1449.704738 msecs"
>
> - on vectors:
>
> (dotimes [i 10] (time (dotimes [j 100] (into [] (range 1e4)))))
> "Elapsed time: 307.838501 msecs"
> "Elapsed time: 243.426624 msecs"
> "Elapsed time: 234.013566 msecs"
> "Elapsed time: 247.711805 msecs"
> "Elapsed time: 253.154264 msecs"
> "Elapsed time: 284.375476 msecs"
> "Elapsed time: 227.773865 msecs"
> "Elapsed time: 247.807286 msecs"
> "Elapsed time: 240.959477 msecs"
> "Elapsed time: 230.627513 msecs"
>
> (dotimes [i 10] (time (dotimes [j 100] (naive-into [] (range 1e4)))))
> "Elapsed time: 406.55752 msecs"
> "Elapsed time: 419.365134 msecs"
> "Elapsed time: 413.544069 msecs"
> "Elapsed time: 410.892756 msecs"
> "Elapsed time: 394.827165 msecs"
> "Elapsed time: 407.338251 msecs"
> "Elapsed time: 441.60916 msecs"
> "Elapsed time: 401.721826 msecs"
> "Elapsed time: 412.7506 msecs"
> "Elapsed time: 399.972445 msecs"
>
>
> Clojure 1.4.0
>
> java version "1.6.0_26"
> Java(TM) SE Runtime Environment (build 1.6.0_26-b03)
> Java HotSpot(TM) Server VM (build 20.1-b02, mixed mode)
>
>
> Cheers.
>
> Marek.
>
> On Sunday, November 4, 2012 3:41:14 AM UTC+1, Satoru Logic wrote:
>>
>> Hi, all.
>>
>> I am following an example demonstrating that `transient` can help
>> optimize mass updates to data structures:
>>
>> First, a function is defined, which doesn't use transient collection:
>>
>> (defn naive-into
>>> [coll source]
>>> (reduce conj coll source))
>>
>>
>> This is supposed to run slower than the built-in `into`, because, as the
>> book said, `into` uses transient collections whenever possible.
>>
>> But the result shows up in my `repl` is just the reverse.
>>
>>
>>> user=> (time (do (into #{} (range 1e6)) nil))
>>> "Elapsed time: 4882.176 msecs"
>>> nil
>>> user=> (time (do (naive-into #{} (range 1e6)) nil))
>>> "Elapsed time: 3745.707 msecs"
>>> nil
>>
>>
>> What could be the reason of this?
>>
>
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
