My plots for pango-profile

[Kalle Vahlman <kalle vahlman gmail com>]

Here's my charts:

http://iki.fi/zuh/kuvat/orig-vs-patches.png

Very cool results, barring the two red ones.

Both of these were applied:

fmq-easy-script-and-paired-chars-tables.diff
fmq-gunichar-to-glyph-index.diff

I ran the test with just one of them, but the results got b0rked
somehow. The number of characters was somehow below the 400k in most
languages, so they were not getting plotted. I'm still uncertain what
caused this, the run should have been similar with the one I got the
"good" data on.

Suggestion: the benchmark data should include the run options for the
benchmark for future reference.

--
Kalle Vahlman, zuh iki fi
Powered by http://movial.fi