Lies, damned lies and benchmarking

In his classic paper on benchmark skewing, Twelve Ways to Fool the Masses, Bailey cites as an example the dubious practice of presenting the timings from the inner most parts of a solver and reporting them as though they're the performance of the entire application.

It seems the lesson has been well learnt: one of the cell processor benchmarks has been caught playing with time in the same way. Thus, the application reports that performance scales linearly as more PEs are enabled, whilst the wall-clock time of the application remains relatively constant thanks to the dominance of start-up costs etc.

Very naughty.

BA on Journey's End

Bad Astronomy has a good post that really nails everything that was good and bad about Journey's End.

I totally agree that the moment when Davros taunts the Doctor with his failings was excellent and necessary. The Doctor gets things his own way far too often and needs to be taken down a peg or two from time to time, if only to prevent him from becoming completely insufferable. What's more, Davros' points were completely valid: for someone who abhors violence, the Doctor seems to be responsible for rather a lot of it and the fact that it is enacted by proxy in no way diminishes his ultimate moral responsibility for it.

Perhaps the Doctor's moral culpability is no less than that of Davros, for although the latter explicitly sought to create killing machines to do his dirty work for him, his hands remain almost as clean (or as dirty) as those of the Doctor. Better yet, Davros doesn't seek to evangelize and to project standards to which he himself fails to adhere. Davros is a monster, plain and simple. He doesn't pretend to be otherwise — except when he was pretending to the be The Great Healer, of course! — but the Doctor, on Davros' evidence, is worse: he's hypocritical.

Also, Gwen? From Cardiff? Yes it ties up a loose end but, given her accent, not particularly convincingly...