No, what I am saying is that the # used in the GG is the same as the one used in other places. A number was generated, probably before the last design iteration, and then was never corrected. A 10% variation I could live with, which would be 4 ounces. I think in this case they probably just didn't weigh the right bag. Sloppy?
This is just a guess, but the point here is if Marmot can't be trusted to do something as simple as weigh the right bag, why put any credence in their so-called down testing program?