possible SBWS measurement quality regression

With the release of v1.1 to longclaw seeing evidence of measurement quality degradation, regression. Have not had time to work up an analysis.

Recommend running Matt's original scanner and comparing raw bandwidth results.