> I am gleaning that from personal testing. That test script is very > system-dependent, unfortunately, so not meaningfully distributed. But anyway would it be possible to provide to us sample data to do similar testing? -- Frédéric