McMurdo's numbers on accuracy are bogus. All GPS PLBs have the same accuracy as it is not dependent upon the GPS, but rather on the limitations of the digital coding.
The difference for the quoted accuracy figures may be down to whether whether a 2D Circular Error Probability CEP or 3D Spherical Error Probability SEP is being quoted by the 2 different manufacturers. SEP is always worse than CEP by about 1.4 times which would account for the difference so it could well be that the accuracy figure of 60 metres quoted by McMurdo is not bogus at all.
Again, has absolutely nothing to do with the accuracy of the native GPS or how they compute it. The GPS itself could be accurate to within 1mm it wouldn't change a thing. The issue is the resolution that the data format can deliver. Now, if the manufacturer was to quote it's number at a specific latitude, then you could legitimately have different numbers, but this is not being done as these are world-wide devices, so worst case must be assumed, which is at the equator.
The location transmitted is offset in 4 second increments +/- from the Coarse position in 2 degree increments. At the equator, one second of latitude/longitude = 101.3 feet.
McMurdo chooses to present what it says is what an average user might expect, but IMO that is misleading unless they provide the information a consumer needs to determine what that number really means.