[ https://issues.apache.org/jira/browse/MATH740?page=com.atlassian.jira.plugin.system.issuetabpanels:commenttabpanel&focusedCommentId=15955063#comment15955063
]
Michael Borcherds edited comment on MATH740 at 4/4/17 12:26 PM:

I've run the official tests on FastMath on GWT + Core i7 in Chrome
A lot slower:
abs 20
acos 14
atan2 21
cbrt 11
pow 12
asin 2.4
log 10
log10 12
log1p 13
hypot 20
expm1 2.0
No significant difference:
atan 1.2
cos 1.5
cosh 0.95
sin 1.4
sinh 0.84
tan 1.5
tanh 1.0
faster:
exp 0.64
sqrt 0.76
See MATH1406 for more GWT discussion
was (Author: murkle):
I've run the official tests on FastMath on GWT + Core i7 in Chrome so my patch deletes almost
all of FastMath and uses Math instead
A lot slower:
abs 20
acos 14
atan2 21
cbrt 11
pow 12
asin 2.4
log 10
log10 12
log1p 13
hypot 20
expm1 2.0
No significant difference:
atan 1.2
cos 1.5
cosh 0.95
sin 1.4
sinh 0.84
tan 1.5
tanh 1.0
faster:
exp 0.64
sqrt 0.76
See MATH1406 for more GWT discussion
> Some "FastMath" functions are slow
> 
>
> Key: MATH740
> URL: https://issues.apache.org/jira/browse/MATH740
> Project: Commons Math
> Issue Type: Wish
> Reporter: Gilles
> Priority: Minor
> Fix For: 4.0
>
> Attachments: jafamatests.zip
>
>
> From the two benchmarks we currently have in "FastMathTestPerfomance", we have that the
following functions are much slower in "FastMath" than in either "Math" or "StrictMath" (the
performance *loss*, for each of the benchmarks, is given in parentheses):
> * log10 (46%, 36%)
> * log1p (68%, 112%)
> * tan (11%, 61%)
> * atan (26%, 125%)
> * atan2 (44%, 40%)

This message was sent by Atlassian JIRA
(v6.3.15#6346)
