[webkit-help] Understanding Jetstream Performance Scores
malek.musleh at gmail.com
Tue Apr 21 08:22:19 PDT 2015
I am trying to get a better sense of how the scores for the JetStream
workloads are being computed. In general, I know that each benchmark
runs for some number of iterations (default=3 I think), and then the
geometric mean of the runs is taken and reported.
Do the reported scores for each run correspond directly to some
performance metrix cycles per instruction (CPI) or IPC, or time?
More information about the webkit-help