Author: Carl Friedrich Bolz <[email protected]>
Branch: extradoc
Changeset: r4646:63be3ddd3aa9
Date: 2012-08-16 19:55 +0200
http://bitbucket.org/pypy/extradoc/changeset/63be3ddd3aa9/
Log: it really doesn't make sense to run Fix16 with LuaJIT
diff --git a/talk/dls2012/paper.tex b/talk/dls2012/paper.tex
--- a/talk/dls2012/paper.tex
+++ b/talk/dls2012/paper.tex
@@ -955,9 +955,9 @@
\hline
sqrt(float) & 14.99 & 1.37 $\pm$ 0.001 & 0.89 $\pm$ 0.000 & 1.06 $\pm$ 0.013 &
0.83 $\pm$ 0.010 & 0.85 $\pm$ 0.088\\
\hline
-sqrt(int) & 13.91 & 3.22 $\pm$ 0.033 & 2.65 $\pm$ 0.001 & 1.06 $\pm$ 0.009 &
0.83 $\pm$ 0.014 & 1.25 $\pm$ 0.053\\
+sqrt(int) & 13.91 & 3.22 $\pm$ 0.033 & 2.65 $\pm$ 0.001 & - & - & 1.25 $\pm$
0.053\\
\hline
-sqrt(Fix16) & 463.46 & 5.12 $\pm$ 0.005 & 2.96 $\pm$ 0.007 & 12.80 $\pm$ 0.080
& 1.14 $\pm$ 0.009 & 1.34 $\pm$ 0.061\\
+sqrt(Fix16) & 463.46 & 5.12 $\pm$ 0.005 & 2.96 $\pm$ 0.007 & - & - & 1.34
$\pm$ 0.061\\
\hline
\end{tabular}
}
@@ -993,8 +993,7 @@
a single implementation of the benchmark that gets specialized
depending on the class of it's input argument, $y$, while in C,
there are three different implementations. In Lua there is no support for
- integers so only two versions are provided: float and Fix16. Here Fix16 is a
custom class
- that implements scaled floating point arithmetic.
+ integers so only the floating point number is provided.
\item {\bf conv3}$\left(n\right)$: one-dimensional convolution with fixed
kernel-size $3$. A single loop
is used to calculate a vector ${\bf b} = \left(b_1, \cdots, b_{n-2}\right)$
from a vector
${\bf a} = \left(a_1, \cdots, a_n\right)$ and a kernel ${\bf k} = \left(k_1,
k_2, k_3\right)$ using
_______________________________________________
pypy-commit mailing list
[email protected]
http://mail.python.org/mailman/listinfo/pypy-commit
