[Rcpp-commits] r603 - papers/rjournal

noreply at r-forge.r-project.org noreply at r-forge.r-project.org
Sat Feb 6 16:51:01 CET 2010


Author: edd
Date: 2010-02-06 16:51:01 +0100 (Sat, 06 Feb 2010)
New Revision: 603

Modified:
   papers/rjournal/EddelbuettelFrancois.tex
Log:
updated performance numbers based on
edd at ron:~/svn/rcpp/pkg/inst/examples/ConvolveBenchmarks$ ./buildAndRun.sh
[...]
                                user.self sys.self elapsed user.child sys.child
Writing R extensions                0.032    0.004   0.032          0         0
Less careful use of R API           0.248    0.000   0.248          0         0
RcppVector<double>::operator()      0.352    0.000   0.354          0         0
Rcpp::NumericVector::operator[]     0.048    0.004   0.052          0         0
Rcpp::NumericVector::begin()        0.032    0.000   0.033          0         0
All results are equal
edd at ron:~/svn/rcpp/pkg/inst/examples/ConvolveBenchmarks$


Modified: papers/rjournal/EddelbuettelFrancois.tex
===================================================================
--- papers/rjournal/EddelbuettelFrancois.tex	2010-02-06 15:42:57 UTC (rev 602)
+++ papers/rjournal/EddelbuettelFrancois.tex	2010-02-06 15:51:01 UTC (rev 603)
@@ -725,18 +725,19 @@
 table below:
 
 \begin{center}
-\begin{small}
-\begin{tabular}{lrr}
-\toprule
-\textbf{Method} & \textbf{Time} (ms) & Ratio \\ 
-\cmidrule(r){2-3}
-R API & 34 & \\
-\code{RcppVector<double>} & 353 & 10.38 \\
-\code{NumericVector::operator[]} & 55 & 1.61 \\
-\code{NumericVector::begin} & 36 &  1.06 \\
-\bottomrule
-\end{tabular}
-\end{small}
+  \begin{small}
+    \begin{tabular}{lrr}
+      \toprule
+      \textbf{Method} & \textbf{Time} (ms) & Ratio \\ 
+      \cmidrule(r){2-3}
+      R API & 32 & \\
+      \code{RcppVector<double>} & 354 & 11.1 \\
+      \code{NumericVector::operator[]} & 52 & 1.6 \\
+      \code{NumericVector::begin} & 33 &  1.0 \\
+      \bottomrule
+    \end{tabular}
+  \end{small}
+  %\caption{Performance comparison for convolution example}
 \end{center}
 
 The first implementation, using the traditional R API, unsurprisingly 



More information about the Rcpp-commits mailing list