On Thursday, January 31, 2013 11:44:28 AM UTC-8, Jim Mathies wrote: > > > Our Talos results may be measuring imperfect things, but we have > > > enough datapoints that we can draw statistical conclusions from > > > them confidently. > > > Statistics doesn't help if you're measuring the wrong things. Whether Ts > > is measuring the wrong thing, I don't know. It would be possible to learn > > something about that question by measuring startup with a camera, > > Telemetry simple measures, and Talos on the same machine and seeing how > > they compare. > > "Ts, Paint" measures the time between a call to window.open to the first > MozAfterPaint for that window in a running process. It's analogous to > hitting ctrl-n in Firefox. The page that gets loaded is light weight. It > measures pretty much everything associated with window creation, including > widget, dom, layout, and browser front end window startup and rendering. > It's a very good test IMHO.
I looked at the code and saw that. It looked to me like it was probably measuring about the right thing, although it is measuring the times inside JS/Python code that could conceivably introduce delays. It's part of an entire system, though, and the results are only correct if the whole system works correctly, and I don't know that that's been tested recently. Dave _______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
