Thanks. <span></span><br><br>On Sunday, June 9, 2013, Asumu Takikawa wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On 2013-06-09 21:15:31 -0500, Robby Findler wrote:<br>
> Sorry: I should have been clearer: I would only expect a difference when<br>
> the list is short (so your benchmark 2).<br>
<br>
Here're the numbers for that:<br>
<br>
#lang racket/base<br>
(define lst '(a b c))<br>
(time<br>
(for ([i (in-range 30000000)])<br>
(member 'b lst)))<br>
<br>
;;; NEW BRANCH<br>
$ racket member-benchmark-4.rkt<br>
cpu time: 1408 real time: 1414 gc time: 0<br>
$ racket --no-jit member-benchmark-4.rkt<br>
cpu time: 3724 real time: 3731 gc time: 0<br>
<br>
;;; MASTER<br>
$ racket member-benchmark-4.rkt<br>
cpu time: 1380 real time: 1384 gc time: 0<br>
$ racket --no-jit member-benchmark-4.rkt<br>
cpu time: 3616 real time: 3622 gc time: 0<br>
<br>
Cheers,<br>
Asumu<br>
</blockquote>