On 2013-04-11 13:56, Lars Vingli Odsæter wrote:
However, as Atgeirr pointed out, you can not expect the exact same
results from two equal runs.
An issue to discuss is therefore if unit tests with CTest are
sufficient, or if we also should include a test environment for
executables? If yes, then how could this be done? Any thoughts or
If there is a possibility of slight deviations (and there is when you
are doing floating-point), then the test must be a program that knows
how to execute the simulator, read the reference test and compare them
within some tolerance. Both the output may of course be filtered before
they are compared, but it is probably easier to have some common,
machine-parsable format to work from.
--
Roland.
_______________________________________________
Opm mailing list
[email protected]
http://www.opm-project.org/mailman/listinfo/opm
