Sure, not saying there isn't any discrepancy-- just that the discrepancy is in the measuring methods. I was showing that there _is_ a way to produce a 16GB number (albeit in terms of GB, not GiB) given the number of blocks physically available, that's all.