chicken-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Chicken-users] [ANN] fannkuch chicken benchmark!


From: Elf
Subject: [Chicken-users] [ANN] fannkuch chicken benchmark!
Date: Tue, 4 Mar 2008 08:12:27 -0800 (PST)


attached is new code for the fannkuch benchmark. it has been tested on openbsd (as thats the only machine i have access to with any memory or
cpu).

Old program:
run1:
    real   1m7.544s
    user   1m6.602s
    sys    0m0.492s
run2:
    real   1m8.324s
    user   1m7.250s
    sys    0m0.563s

New program:
run1:
    real   0m25.048s
    user   0m24.500s
    sys    0m0.422s
run2:
    real   0m24.835s
    user   0m24.227s
    sys    0m0.406s
run3:
    real   0m24.824s
    user   0m24.391s
    sys    0m0.328s
run4:
    real   0m24.999s
    user   0m24.469s
    sys    0m0.320s

C benchmark (for comparison):
run1:
    real   0m11.717s
    user   0m11.656s
    sys    0m0.000s
run2:
    real   0m11.814s
    user   0m11.602s
    sys    0m0.008s
run3:
    real   0m11.719s
    user   0m11.688s
    sys    0m0.008s


Note: Only the first two runs are consecutive; all following runs are staggered
      randomly along with running other programs in order to try to incur max
      inefficiency.

At the moment, it's looking like we're at twice the speed of C in this
benchmark. Other people, please test it on linux boxen, and post some figures... if they all look good, with your permission, i'd like to submit
this to the shootout.

-elf

Attachment: fannkuch_chicken.scm
Description: Text document


reply via email to

[Prev in Thread] Current Thread [Next in Thread]