Closed eddelbuettel closed 8 years ago
Good point. Will fix in the next version
:)
More methodological question: score is sum of subtests? Should we score [average] rankings?
This is really useful by the way. Sitting at a new desktop that should be a screamer but isn't.
Yes, the y-axis (score) is just total time of all tests. This time is dominated by just four tests. My plan is as the sample size increases, the resolution of the reporting is more detailed.
I like the idea of scoring the ranks (will investigate).
Since I have your attention two quick questions:
R CMD config BLAS_LIBS
works under Windows/Mac?R_COMPILE_PKGS=3
in the Renviron or byte compile. Is there a downside to byte compiling I don't know about.Don't know about either of those two bullet points, sorry. I guess I could try Windows in the VM at work tomorrow.
Fixed by 570157be72ec367551abfad26efefd47e46687cd
Those atom results pretty much destroy the useful information in the plot.