That’s helpful. So instead of a much larger test matrix you are using a model to reduce that to the most likely candidates, right?

[flagged]