blue / papers.hotmobile2015-numerator · Commits

Commit ffa568d23fa728eac0444c54bfb8a9a5778aaa5b

Authored by Geoffrey Challen
1 parent 3ad686ef

New.

Inline Side-by-side
Showing 2 changed files with 27 additions and 2 deletions
figures/tables/tableALL.tex
  View file @ffa568d
... ... @@ -137,9 +137,9 @@
137 137 \end{tabularx}
138 138 }
139 139  
140   -\label{table-content}
141   -
142 140 \caption{\small \textbf{Apps sorted by content energy efficiency.}}
  141 +
  142 +\label{table-content}
143 143 \end{subtable}
144 144  
145 145 \caption{\small \textbf{Evaluating Components of a Value Measure.}
... ...
results.tex
  View file @ffa568d
... ... @@ -82,6 +82,31 @@ looked better when their foreground usage was considered.
82 82  
83 83 \subsection{Content Energy Efficiency}
84 84  
  85 +Finally, we the data we collected by instrumenting the
  86 +\texttt{SurfaceFlinger} and \texttt{AudioFlinger} components to compute a
  87 +simple measure of content delivery. We measure the audio and video frame
  88 +rates and combine them into a single measure by using bitrates corresponding
  89 +to a 30~fps YouTube-encoded video and 128~kbps two-channel audio, with the
  90 +weights representing the fact that a single frame of video contains much more
  91 +content than a single sample of audio. We use this combined metric as the
  92 +value measure and again use it to weight the energy consumption of each app,
  93 +with the results shown in Table~\ref{table-content}.
  94 +
  95 +Comparing with the foreground energy efficiency again shows several
  96 +interesting changes. Yahoo Mail, which foreground energy efficiency marked as
  97 +inefficiency, looks more efficient when content delivery is considered. While
  98 +it is possible that one \PhoneLab{} participant uses it to read email very
  99 +quickly, it may be more likely that it uses a ``spinner'' or other fancy UI
  100 +elements that generate artificially high frame rates without delivering much
  101 +information. The inability to distinguish between meaningless and meaningful
  102 +video frame content is a significant weakness of this simple approach.
  103 +YouTube and Candy Crush Saga both earn high marks, which is encouraging given
  104 +that they are very different apps but also might be a result of overweighting
  105 +screen refreshes. The Android Clock is also an unsurprising result, as it
  106 +requires almost no energy to generate a relatively-large number of screen
  107 +redraws.
  108 +
  109 +\subsection{Survey Results and Discussion}
85 110  
86 111 \begin{figure*}[t]
87 112 \centering
... ...